The Sporting News turned 140 this year.

There’s not a print edition anymore, of course. But I’d bet no paper has shown up in more of my posts.

Back in July, I made a video about how a St. Louis weekly became “the Bible of Baseball,” and how it lost that title. I promised more clippings later. This is later. If you missed the video, you can check it out here:

Can You Still Read It?

Yes, and for free.

The paper’s archive lives at archive.sportingnews.com. Some issues have been rescanned from the original newsprint instead of the old microfilm, and the difference is night and day. The April 3, 1919 issue I showed back in March looks like it was printed last week. (See more in Improved Sporting News Archives)

A few things I’ve learned from using it:

• Not every issue has been rescanned. The pre-1919 papers are often too fragile, so you’re mostly reading microfilm before then.

• Go by the date. The paper came out weekly, so a game played on a Monday might not show up until the following week’s issue, or the one after that.

• Read the whole issue, not just the story you came for. The off-season issues are full of old-timers talking about the game they played.

• Tom Hall, the paper’s chairman, told readers in our comments in March that searchable text was coming, working back from the present to about 1945. I believe we’re still waiting on that.

I use The Sporting News almost daily for article ideas and in order to bring my own projects to life.

What Did It Tell Me About Mathewson and Brown?

Last Friday’s Great Story was the September 4, 1916 game when Christy Mathewson and Three Finger Brown faced each other one last time. The Chicago and Cincinnati papers had the game story. The Sporting News had the verdict.

Ten days later, the paper looked at a 10–8 slugfest between two of the best pitchers who ever lived and asked readers to imagine either of them, at his best, “allowing any such bombardment.” It also repeated Matty’s promise in the clubhouse: if he ever pitched again, he’d buy every one of his teammates a suit of clothes. He never had to.

That’s what The Sporting News did best. The dailies told you what happened. The weekly told you what it meant, a week later, with every box score in one place.

Where Did Conlon’s Photos Come From?

Every Friday, I pick a Charles Conlon card and test the story behind it. Those cards exist because of The Sporting News.

Conlon shot ballplayers in New York from about 1904 into the early 1940s. Not long after he died in 1945, his negatives became the property of The Sporting News. They sat in the paper’s files for decades. When Megacards put out the Conlon Collection sets in the early 1990s, the cards carried The Sporting News name.

The paper later sold the negatives to a collector named John Rogers. After Rogers ran into legal and money trouble, a court let the collection be sold, and Heritage Auctions sold it as a single lot in August 2016 for $1,792,500.

So every time I hold up a Conlon card on Friday, I’m holding up a little piece of The Sporting News’ photo files.

Since the Video

• The free archive got better, as above. For research on the 1950s in particular, I don’t know a better source.

• The paper turned up in two Great Stories, and I expect it in plenty more.

• I wrote about what the paper did so well in July, using a December 1945 issue that had a Kip Selbach interview and a look back at the Mathewson trade.

Can You Solve This One?

Here’s a loose end from the 1916 research.

In the September 7, 1916 issue, W. A. Phelon wrote that the Reds were booked to play an exhibition at Henry, Illinois, on September 6. The posters promised “Christy Mathewson, greatest of all pitchers, will pitch!”

As far as I know, Matty never pitched again. So what happened in Henry? Did the Reds play there? Did anyone ask for their money back?

If you’ve got newspaper access in central Illinois, or a local historian in your back pocket, I’d love to know.

And a bonus question: what’s the best thing you ever found in an old issue of The Sporting News?

The best answers this week could end up on the podcast and here on the blog.

Leave a comment

Supporters help pay for the newspaper subscriptions behind posts like this one, which is how a 110-year-old column ends up on your screen.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

“I have no ballclub anymore.”

He meant it literally. Clift spent ten seasons with the St. Louis Browns and parts of three with the Washington Senators. By the 1980s, both teams were gone. The Browns moved to Baltimore in 1954. The Senators left for Minnesota in 1961. Clift was a widower living in a mobile home in Yakima, Washington, getting by on Social Security and a small pension.

But in 1982, APBA gave him a ballclub back.

The question is why APBA could give Harlond Clift a card, but not John McGraw. McGraw has the third-highest on-base percentage in major league history. He is in the Hall of Fame. And he isn’t anywhere in the Original Franchise All-Star set.

What’s on the Envelope?

The first OFAS team I’m covering is the St. Louis Browns. Or, if you go by the envelope, the Baltimore Orioles.

OFAS came without lineup sheets, so the roster is printed on the envelope the cards came in. It reads: “Original Franchise All-Stars (St. Louis) Baltimore Orioles.” St. Louis gets the parentheses. Hang on to that.

This franchise has always borrowed its names. It started as the Milwaukee Brewers in 1901, moved to St. Louis in 1902, and took the name “Browns” from the old National League club that had just renamed itself the Cardinals. After 52 seasons in St. Louis, it moved to Baltimore and borrowed another name: the Orioles.

The 25 players split 15 Browns and 10 Orioles. But look at how they split.

Eleven of the 15 hitters are Browns, most of them from the hitting boom of the 1920s. George Sisler hit .407 in 1920. Ken Williams hit 39 home runs in 1922. Baby Doll Jacobson, Johnny Tobin, Harry Rice, Marty McManus and Hank Severeid all hit .300 or better in their card years. The Orioles only add four hitters: Brooks Robinson, Boog Powell, Ken Singleton and Bobby Grich.

The pitching flips it. Six of the ten pitchers are Orioles, led by Jim Palmer, Mike Cuellar and Dave McNally. The Browns get Urban Shocker, two pitchers from the Deadball Era in Harry Howell and Jack Powell, and Satchel Paige, who made the 1952 All-Star team at age 46.

If you want to know what the Browns were, there it is. They could hit, they couldn’t pitch, and they lost.

Who Was Harlond Clift?

I’ll admit it. Clift was the one name on this envelope I didn’t know.

He was born in El Reno, Oklahoma, in 1912 and grew up on an apple orchard near Yakima. People kept getting his name wrong. It’s Harlond, but the papers kept writing “Harland,” and he gave up correcting them.

His card year is 1937. That season he hit .306 with 29 home runs and 118 RBI. The 29 home runs were a record for a third baseman. He broke his own record the next year with 34.

He could field, too. In 1937, he turned 50 double plays and made 405 assists at third base. Both were major league records, and both stood until 1971.

And he did all of this for one of the worst teams in baseball. The 1937 Browns went 46-108. In Clift’s ten years in St. Louis, the Browns finished higher than sixth place only once, when they finished third in 1942. The 1939 team went 43-111.

Clift also walked. He drew 115 walks in 1936 and topped 100 walks six times in his career. His career on-base percentage was .390. Remember that number.

So why did Seitz pick 1937 instead of 1938, the year he hit 34 home runs? My guess is the batting average. APBA is mostly a batting-average simulator. In 1937 Clift hit .306. In 1938 he hit .290, with the same 118 RBI. If you’re building a card, .306 is the better season.

Then came the cruelest part of the story. On August 18, 1943, the Browns traded Clift to Washington. In 1944, the Browns won the only pennant they ever won in St. Louis.

Clift wasn’t there to see it. He came down with the mumps and didn’t play for Washington until July 16. Twelve games later he was thrown by a horse, and his season was over.

Clift or Brooks?

In Now Taking The Field, Tom Stone ranks the Orioles franchise’s third basemen in this order: Brooks Robinson, Harlond Clift, Manny Machado. When Stone compares other people’s lists, only two include Clift. One is from The Baseball Research Journal. The other is a 1983 history of the St. Louis Browns.

That tells you who remembers Harlond Clift. The historians do. Almost nobody else does.

Seitz remembered him, too. So now every OFAS manager who picks up this team has two third basemen: Brooks, with his 1964 MVP season and one of the best gloves anyone has ever seen, and Clift, the slugger with the walks. OFAS doesn’t come with a lineup sheet, so the choice is yours.

I don’t think anybody can look past Brooks Robinson at third base. But you have to give respect to Clift’s power numbers. And it’s quite impressive that a relatively obscure player from such an awful team is even in this discussion.

Where Are McGraw’s Orioles?

To understand why John McGraw isn’t on this team, you have to know where the name on the envelope came from.

Before 1954, “Baltimore Orioles” meant one team. Ned Hanlon’s National League Orioles won pennants in 1894, 1895 and 1896. They had five future Hall of Famers: John McGraw, Willie Keeler, Hughie Jennings, Joe Kelley and Wilbert Robinson.

These guys could hit. Keeler hit .424 in 1897. Jennings hit .401 in 1896. McGraw reached base 54.7 percent of the time in 1899, and he finished his career with a .466 on-base percentage. Only Ted Williams and Babe Ruth are higher. Clift’s .390 was very good. McGraw’s was in another league.

Remember the intro, when I said no OFAS player comes from before 1894? The Orioles won their first pennant in 1894. Their best years fit inside the set’s window.

And none of those five players has an OFAS card. Not one.

Here’s why. After the 1899 season, the National League cut down from twelve teams to eight, and Baltimore was one of the teams that got dropped. The team was gone.

McGraw came back in 1901 to run a new Baltimore Orioles team in the American League. It didn’t last either. McGraw feuded with league president Ban Johnson, and in July 1902 he left to manage the New York Giants.

On July 17, 1902, the Orioles’ majority owner sold his stake to Andrew Freedman of the Giants and John T. Brush of the Reds. They released the team’s best players. Joe McGinnity, Roger Bresnahan, Jack Cronin and Dan McGann went to the Giants. Joe Kelley and Cy Seymour went to Cincinnati.

That same day, the Orioles didn’t have enough players to take the field. They forfeited the game. Their opponent was the St. Louis Browns.

After the season, the American League moved the franchise to New York, where it became the Highlanders and, later, the Yankees. Even that connection is disputed. Baseball-Reference stopped counting the 1901–02 Orioles as part of Yankees history in 2014.

So McGraw’s Orioles died twice. The first team was cut out of the league. The second was gutted and shipped to New York. McGraw spent the next 30 years with the Giants, but as a manager, and managers don’t get cards. Keeler, Jennings and Kelley all played well elsewhere, but their great seasons belonged to a team that no longer existed.

So the name on the envelope belongs to McGraw’s Orioles. The history behind it doesn’t.

No Ballclub Anymore

So, back to the question. Why does Harlond Clift have a card when John McGraw doesn’t?

Because OFAS follows franchises, not cities and not names. Clift’s franchise survived by leaving town. McGraw’s died twice. The Original Franchise All-Star set can only honor the teams that lasted, and the greatest Orioles of all never got the chance to last.

Clift had a card, and almost nobody remembered him. Everybody remembers McGraw, and he has no card.

And it was the Orioles, a franchise wearing McGraw’s team’s name, that gave Harlond Clift his ballclub back. Even if St. Louis has to sit in parentheses.

Brooks or Clift at third? And which other great player lost his OFAS card because his team disappeared? Tell me in the comments.

Leave a comment

The best comments each week get featured on the podcast and here on the blog.

If you missed the start of the series, check out Original Franchise All-Stars: An Introduction.

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. And, of course, supporters are the reason I can spend my weekend digging into the life of a third baseman nobody remembers.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

That’s because they also gave up 1,199 runs, the most of any team since 1900. No team since has given up more.

So here’s a simple test for any baseball game: can a team lose 102 games with a lineup like that?

In Clifford Van Beek’s National Pastime, the 1930 Phillies win 72.

How Did I Test It?

I’ve written a lot about National Pastime over the years. You can find some of my prior National Pastime posts here.

This time I wanted an answer to the biggest question of all: how well does the game reproduce the season it was made for?

One replay can’t answer that. Luck alone can push a team ten games off in a single season, or could significantly impact the runs scored and runs allowed totals. So I had Claude Code rebuild the game and play it. That way I could replay the entire season a few hundred times.

I played the original National Pastime season 200 times in all:

• 100 seasons of the game as Van Beek designed it. His original cards, 18 per team. Fixed lineups, a five-man rotation, and no pitching system at all.

• 100 seasons with Bill Staffa’s Skeetersoft pitching system added. Same cards, but every pitcher gets a Skeetersoft level, and the manager pulls starters and sends up pinch hitters about as often as 1930 managers did.

Those of you who purchased the rerelease of National Pastime sometime in the early 2000s likely remember Staffa’s pitching system. This involves a handful of pitcher cards that replace base hit numbers with outs, or, in some cases, add extra base hit numbers to hitter cards.

What Does Van Beek Get Right?

More than you might expect for a game from 1930.

Run scoring is close. Real 1930 teams scored 5.56 runs a game. National Pastime teams scored 5.66.

The best team wins. The Philadelphia Athletics won 102 games in 1930. In my replays they averaged 100 and took the American League pennant 66 times out of 100.

Most teams land close to where they finished. The average team missed its real win total by about 6 games. Luck alone would account for about 5 of that, so Van Beek’s cards don’t leave much on the table here. (More on that in the box at the end.)

That’s consistent with what I’ve experienced while playing National Pastime on the tabletop. Though there’s not any sort of pitching system out of the box, the game is still remarkably good at replicating batting averages.

What Does He Get Wrong?

Three things stand out.

• Too few walks. Real 1930 teams drew about 479 walks. National Pastime teams drew 377. That’s about 100 walks a team gone missing.

• Too many home runs. 1930 was a home run year, but National Pastime goes further. Teams averaged 119 home runs against a real 98, about 22% too many.

• The standings are too bunched up. Good teams don’t win enough, and bad teams don’t lose enough.

That last one is the big problem, and the Phillies show it best.

To see why, you need to know what a pitcher does in National Pastime. When the other team bats, the answer is nothing. Every result comes off the hitter’s card. Lefty Grove and the worst pitcher in the league get the same results against the same batter.

So the 1930 Phillies get to keep their .315 lineup and leave their pitching behind. They win 72 games instead of 52.

The Chicago Cubs get the same deal. Hack Wilson hit 56 home runs and drove in 191 runs that year, and the Cubs led the National League with 171 home runs. They finished second. In National Pastime they win the pennant 72 times out of 100. The St. Louis Cardinals, the real champions, win it 11 times.

So the game knows who can hit. It has no idea who can pitch.

What Does Skeetersoft Change?

Bill Staffa’s system gives the pitcher a say, and the first change is easy to see.

The walks come back. With Skeetersoft, teams drew 480 walks a season. The real number was 479.

The order of the standings gets better too. The Cardinals go from 83 wins to 93, against a real 92. Cleveland goes from 75 to 82, against a real 81. Measured by how well the replay standings line up with the real ones, the match improves from .87 to .93.

But the standings swing too far the other way.

• The Brooklyn Robins won 86 games and finished fourth. With Skeetersoft they win 104 and take the pennant 70 times out of 100.

• The St. Louis Browns won 64. With Skeetersoft they win 47.

• The Phillies won 52. With Skeetersoft they win 37.

Without a pitching system, the gap between good and bad teams is too small. With Skeetersoft, it’s too big. The average miss even grows a little, from 5.7 games to 6.2.

So what happened?

Why Do the Good Teams Get Too Good?

The answer likely comes from the roster itself.

Van Beek made 18 cards per team. Real teams used 25 to 40 players. That leaves room for about five pitchers on each club.

So who got those five spots? The pitchers who threw the most innings, which usually means the best ones.

Look at Brooklyn. Its five pitchers are Dazzy Vance, who led the league with a 2.61 ERA, and four solid starters. Not one of them is a bad pitcher. Those five threw about three-quarters of Brooklyn’s innings in 1930. Somebody else threw the other quarter, and none of those pitchers has a card.

In the replays, Vance and his four teammates throw every inning. Brooklyn’s runs allowed drop from 4.79 a game to 4.14, and the Robins run away with the pennant.

Washington has the same story. All five Senators pitchers sit close together on the good end of the Skeetersoft scale. No mop-up man means no bad innings.

At the bottom, the Phillies get the opposite. Three of their five pitchers sit at the worst level on the scale. In the replays, those three throw more than 600 innings with ERAs over 11. In 1930 they were bad, but their ERAs ran from about 6.8 to 7.7.

This isn’t Bill Staffa’s fault. His system rates each pitcher the way it should. The trouble is that it only has Van Beek’s 18 cards to work with.

So, How Accurate Is National Pastime?

As Van Beek designed it, National Pastime gets the run scoring right and most of the standings close. It knows who can hit, but it can’t tell a great pitching staff from a terrible one.

Skeetersoft fixes the pitching and brings the walks back. Then Van Beek’s short rosters turn that fix into an overcorrection.

The 1930 Phillies sum it up. Van Beek’s game gives them 72 wins. Skeetersoft gives them 37. The real team won 52.

Next Monday, I’m going to look at those 18 cards and how Van Beek built them.

Let’s Talk About It

Did you ever replay a season with National Pastime, or with Skeetersoft? How did the standings come out? And if you added your own pitching rule, what was it?

The best answers this week could show up on the podcast and here on the blog.

Leave a comment

For the geeks

• Average miss: for each team, the gap between its average replay wins and its real 1930 wins, averaged over all 16 teams. Van Beek’s game: 5.7 wins. Skeetersoft: 6.2 wins.

• Luck line: a real season is one roll of the dice. Even a perfect game would miss each team’s real total by about 6.2 wins on the standard-deviation scale (the square root of 154 × 0.5 × 0.5), or about 5 wins on the average-miss scale.

• Correlation of replay wins with real wins: .87 for Van Beek’s game, .93 with Skeetersoft. Skeetersoft puts the teams in a better order.

• Spread of the standings (standard deviation of team wins): real 1930, 15.5. Van Beek’s game, 12.9. Skeetersoft, 20.9. That’s why a better order still comes with a bigger miss.

• The Van Beek runs used a balanced 154-game schedule. The Skeetersoft runs used the real 1930 schedule.

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. Reader support kept 200 seasons of 1930 baseball running for today’s post.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

The newest answer anybody could name was about twenty years old.

That’s this week’s podcast. You can listen above.

What Games Came Up?

If the episode sends you looking, here’s where to start. One line each, so you can judge for…

Read more

Last year I wrote about horror stories from league play: the cheating, the threats, the guy who seemed to win a few too many dice rolls. This time I want the stories from when it was just you, a board, a stack of cards and a season you’d sworn you were going to finish.

What’s Your Replay Horror Story?

Every replayer has one. Maybe it was one of these:

• The dice. A streak so cruel you started checking them for weights, like fourteen straight outs from your cleanup hitter, or an ace who gave up a grand slam every time you needed one out.

• The injury. Your best player went down in the second week of April, and the rest of the season felt like a funeral.

• The paperwork. Scoresheets lost in a move, a coffee spill across July, a dog that ate the Cubs.

• Real life. A job change, a new baby, a basement that flooded. The replay stopped on June 3rd and never started again.

• The season that went wrong. The team was supposed to win 105, finished fourth, and you had to keep rolling anyway.

Tell me three things:

• The game and the season you were replaying.

• What went wrong.

• Did you finish it?

I’ll go first. The thing that haunts me more than anything else is playing with the 1930 New York Yankees in National Pastime. The reason? Lou Gehrig keeps injuring Babe Ruth. It’s like a never ending curse.

Why Do We Keep Rolling?

Most of us kept going anyway. We rebuilt the scoresheets and played out the schedule with a backup shortstop hitting .190.

I think that’s part of why we do this. Real seasons have bad breaks too, and the replays I remember best are usually the ones where something went wrong.

So tell me about yours. If it’s too painful to finish telling, I’ll understand.

The best stories this week might get a spot on the podcast and here on the blog, so give me all the gory details.

Leave a comment

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. Reader support keeps this table open, even on the nights the dice turn on us.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

The poster promised a duel. The final score was 10-8.

The card

Every Friday, we take one of Charles Conlon’s cards and go looking for the game behind it. This one has a famous back story: the farm accident that took Mordecai Brown’s right index finger, and the curveball he built around what was left. You can read it right on the back of the card.

The photo on the front is from 1915, when Brown was with the Chicago Whales of the Federal League. The game I want to look at came a year later. It was the last one he ever pitched in the major leagues.

The story

Here’s how the game was sold.

On the left is Christy Mathewson, by then the manager of the Cincinnati Reds, in a September 1 dateline from Cincinnati: “You can positively count on my pitching against Brown on Sept. 4th.” On the right is Joe Tinker, Brown’s manager with the Cubs, writing from Chicago: “Mordecai Brown will be ready to battle against Mathewson Labor Day.” In the middle is a giant picture of Brown’s hand.

Brown and Mathewson had been rivals for a decade, and both ended up in Cooperstown. Mathewson had mostly quit pitching when he took over the Reds. He agreed to come back, the Tribune said, “just for another chance at Brownie.” The Enquirer said the day “will be known as Mathewson-Brown Day.”

That’s the billing: two of the best pitchers of the dead-ball era, one last time. Did the game live up to it?

Finding the game

There’s no detective work on the date. The game was Monday, September 4, 1916, at Weeghman Park, the Cubs’ new home. (It had been the Whales’ park the year before, when Brown pitched there for the Federal League champions.)

The only puzzle is which game. Labor Day was a doubleheader, and nobody seemed sure which half would get the old rivals.

• A Chicago Tribune ad on September 3 said, “Brown vs. Mathewson, Tomorrow in Double Header.”

• The Cincinnati Enquirer on September 4 said Mathewson would pitch the first game, with Fred Toney and Mike Prendergast in the second.

• The Chicago Tribune that morning said it didn’t matter which half, since the two games were being “doubled up as an afternoon bargain,” and a wire story said it “had not been decided” which section the old rivals would work.

In the end, Prendergast shut out the Reds 3-0 in the opener, and the two veterans took the second game.

The Tribune said the two would face each other “once more today and probably for the last time,” and the Enquirer said it would “probably” be Mathewson’s final appearance as a big-league pitcher. So everybody knew it was a farewell before it started.

The game

It had rained. The Tribune wrote that “a great group of baseball rooters stood in the rain for an hour before time to start the games and, soon after it was decided to play, filled the plant.” The Enquirer said there were “over 17,000 paid admissions,” and the crowd swarmed onto the field before the second game. Hits into the crowd were ground-rule doubles.

Before the game, Charles Weeghman, the park’s owner, gave Brown and Mathewson huge bouquets of flowers. Both teams gathered around home plate. The two old pitchers had their picture taken hand in hand.

Then they pitched. It wasn’t pretty.

The Cubs scored twice in the first on a triple by Kelly. The Reds answered with one run in the second and two in the third, and were ahead for good by the fourth. Cincinnati’s Ivy Wingo hit safely four times in four tries. The papers don’t agree on the hit totals: the Tribune‘s box score has the Reds with 18 and the Cubs with 14, and others say 19 and 15. Either way, that’s more than 30 hits. Reds 10, Cubs 8.

Neither man would come out. The Tribune noted that the first substitution on either side didn’t come until the ninth, “although it was not expected either of the past master pitchers would finish the way they were being hit.” Each man walked exactly one batter. Brown even picked up two hits and scored twice.

That ninth inning is the part I like best. The Reds led 10-5. Brown led off with a single, and when he got to base, Manager Tinker wanted to send in a pinch runner. According to the Enquirer, “Brownie shook his head, sticking to the paths and eventually scoring his run. He wanted to stay in the game to its completion and did so.” Then, with two out, Vic Saier hit a home run over the short right-field fence, and three runs came in. Two more singles put the tying runs on base. Mollwitz, batting for Pechous, flied out to Neale, and that was it.

Afterward, in the clubhouse, Mathewson told his players: “Boys, I thought I could pitch a few more games, but I find I haven’t got the stuff any more.” He added that if he ever pitched again, he would “buy every one of you a suit of clothes.” He never pitched again.

What the papers said

The 1916 papers argued about it, too.

I. E. Sanborn of the Tribune called it the “Brown-Mathewson swan song” and then wrote: “Gone are the days when Matty and Brownie could give the greatest batsmen in the game a winning argument nine times out of ten.” The Enquirer said “neither man was the pitcher of old. Their stuff was gone, their speed was lacking.” And The Sporting News complained: “Imagine either of them, when at his best, allowing any such bombardment as that!”

Sporting Life saw it differently. It said the lesson of the game was “the marvelous control of the old masters of the curve,” and it counted one walk apiece. That’s true. They did throw strikes. The Reds and Cubs just hit them.

Verdict: Embellished. The farewell was real. Mathewson kept his promise, both men finished what they started, and 17,000 people came to say goodbye. The duel is what got embellished. The papers said so at the time, and the box score agrees. People came for the sentiment, and they got it.

This feels like one of those games late in your replay season where you’re just trying to get to the end of the project. Just leave the guys in and forget about the score.

About the photo

The photo on the front is from 1915, a year before the game, when Brown was pitching for the Chicago Whales. That was the Federal League team he led to the 1915 championship. A year later the Federal League was gone and Weeghman had bought the Cubs. Brown ended up back with his old team, in the old Whales’ ballpark, for one last game.

Let’s argue about it

Mathewson said he’d buy every man a suit of clothes if he ever pitched again. He never had to. Would you rather see a great pitcher get a farewell game, even a bad one, or walk away quietly? And which farewell do you remember best?

Leave a comment

The best comments each week get featured on the podcast and here on the blog, so tell me what you think.

Sources

• Chicago Tribune, Sept. 3, 4 and 5, 1916 [CHECK page numbers: Sept. 3 Section 3 p. 7 (ad); Sept. 4 p. 11; Sept. 5 p. 15]

• Cincinnati Enquirer, Sept. 4 and 5, 1916, p. 8

• Cincinnati Post, Sept. 5, 1916, p. 8

• The Sporting News, Sept. 7 and Sept. 14, 1916

• Sporting Life, Sept. 16, 1916

• Poster, Hall of Fame Library

• Baseball-Reference box score, Sept. 4, 1916

• SABR BioProject biography of Mordecai Brown

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. The Brown card above is from my collection, and supporters help me keep adding to it.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

Forget the two separate scores for offense and defense. Just take each team’s run differential and look at how far it stood from the rest of its league.

I tried it. The 1939 Yankees, the team with the best run differential in my entire database, fall all the way to 49th. And the new No. 1 is a team almost nobody would pick: the 2015 Toronto Blue Jays.

I’m going to look into that criticism of the Domination Index, as well as other criticisms, in today’s post.

Any ranking has its weak spots. I’ve been championing this one for years on this blog, so I might as well be the first one to poke holes in it. Here are the four biggest problems I know of, how much each one matters, and what I’ve done about them so far. Then you can decide.

A Quick Refresher

The Domination Index asks one question: how far did a team stand from the rest of its league?

Every team gets two scores, one for scoring runs and one for preventing them. Each is measured in “steps,” where a step is the normal spread between teams in that league that season. Add the two together and you have the index. (If you missed it, the 1939 Yankees post walks through it with real numbers.)

Why Not Just Use Run Differential?

Anyone with a statistics background would point out two things about my method, and they’d be right on both.

First, two step-scores added together aren’t a step-score. When I say the 1939 Yankees scored 3.5, that isn’t “3.5 steps.” It’s a combined score, a combination of two different scores. A step on either the runs scored side or the runs allowed side of the equation isn’t the same as a step in the combined score. In a strict sense, the combined score isn’t a standard deviation score anymore, even though Rob Neyer and Eddie Epstein called it an “SD score” in Baseball Dynasties.

Second, a run scored and a run saved are worth exactly the same in the standings. Run differential treats them the same. However, the Domination Index usually doesn’t treat them as the same. For example, in 1939 one step of run prevention in the American League was 150 runs, but one step of offense was only 88. So every run the Yankees saved counted for less than every run they scored.

However, these two concerns are relatively minor. Adding two scores like this is a standard way to build a combined measure, and it’s exactly how Rob Neyer and Eddie Epstein built the SD score in Baseball Dynasties. And the Neyer-Epstein method agrees closely with measuring the standard deviation of run differential. In fact, across all 2,852 major league team-seasons, the correlation between the two measurements is .985, where 1 would be a perfect match.

But, if you look at the list of top teams using both methods, you’ll see that they disagree:

These are the plain versions of both formulas with no mathematical adjustments, so it’s a fair fight. That’s why the order is a little different from the list in the 1939 Yankees post. In that post I made a few changes to adjust for league size and strength of schedule.

Four teams make both lists. The deadball clubs (the 1906 Cubs, 1902 Pirates and 1917 Giants) and the 1927 and 1939 Yankees drop off the run-differential list entirely. Even the 2022 Dodgers fall to 18th.

Why? It comes down to how each league’s offense and defense moved together. In some leagues, the teams that scored the most also tended to prevent the most runs, as in the 1939 American League. There the index stacks the two scores and a great team gets a very large total. In other leagues the two were unrelated, as in the 2015 American League, and run differential gives a team more credit than my index does.

The 2015 Blue Jays are the example. They had one of the great offenses of all time, nearly 2.9 steps above the league, with a pitching staff that was only a little better than average. Run differential loves that. My index likes it, but not as much.

How much does it matter? It’s mostly a matter of taste. The two methods answer slightly different questions, and both are fair ones.

Is It Harder to Stand Out in a Small League?

This is the most technical problem on the list, and the one I’ve worked hardest to fix.

Imagine an eight-team league, a league where one team is incredible and the other seven are identical. No matter how many runs that one incredible team scores, it can never be more than about 2.65 steps above average on either side. In a 16-team league, the ceiling is 3.87.

That happens because the team is part of the average it’s being measured against. The better it is, the more it drags the average up and stretches the spread, and the harder it gets to stand out.

That’s a real problem when the list mixes eight-team leagues from 1927 with 15-team leagues from today and six-team leagues from Japan. And so I gave the index a size adjustment. It works in two steps:

1. Compare each team with the other teams in its league only, so it isn’t measured against itself.

2. Ask how rare a gap that size would be in a league that size, and put that rarity on one common scale for every league.

For most teams, the adjustment barely moves anything. It adds about a tenth of a point to the 1939 Yankees.

But the fix has its own problem. Close to the ceiling, the “how rare” math balloons.

Take the 2000 Nippon Ham Fighters. They scored 771 runs in the six-team Pacific League, far more than anybody else (the next-best offense scored 638). That put their offense 2.21 steps above average, just under that league’s ceiling of 2.24. Their pitching was below average.

On the unadjusted index, the 2000 Fighters rank 471st out of 3,755 team-seasons. After the size adjustment, they’re 29th. They didn’t even win the pennant; in fact, they finished in third place, 4 1/2 games behind the Fukuoka Daiei Hawks.

They aren’t the only ones. Japanese teams make up about a quarter of the seasons in the comprehensive Domination Index database. On the unadjusted index, four of those Japanese teams make the all-time top 25. After the adjustment, seven do. Nine of the ten biggest boosts in the whole database go to teams from six-team Japanese leagues.

How much does it matter? A lot, if you’re comparing a six-team league with a 15-team league. Very little within the majors.

Is the adjustment a fair reward for doing something nearly impossible in a small league, or is it overdoing it? I go back and forth on this one myself. Let me know what you think.

Should a 60-Game Season Count?

When I ran the top 10 in the 1939 Yankees post, the 2020 Dodgers came in at number one. I said I’d come back to them, and here we are.

Now here’s the surprise. In the unadjusted index, the 2020 Dodgers rank ninth among major league teams. Their jump to No. 1 (from 3.57 to 4.20) has almost nothing to do with the size adjustment. It comes from the schedule adjustment.

Once upon a time Major League Baseball teams played a balanced schedule. However, that hasn’t been true for a long time. Expansion and the move to two divisions per league in 1969 created a significantly imbalanced schedule. This was gradually rebalanced over time in both leagues. However, the advent of interleague play in 1997 completely destroyed any pretense of a balanced schedule.

In the current world of Major League Baseball, not only do teams play teams from the other league, but some teams might have good records simply because their opponents were relatively weak. To account for this, I adjust each team’s runs for the strength of the opponents it actually played. In a normal year, that adjustment is small.

In 2020 it wasn’t. Teams played only within their own region, so the East, Central and West never met. Game for game, the 2020 adjustments were more than four times their usual size. That’s the season where the adjustment has the least to go on, and it’s the season that decides my No. 1.

What about the 60 games themselves? Nothing in the index adjusts for season length. That sounds bad, but short seasons don’t actually crowd the top of the list. The shortened seasons (1918, 1919, 1981, 1994, 1995 and 2020) make up about 5% of all major league team-seasons, and they hold two of the top 25 spots: the 2020 Dodgers and the 1995 Indians.

The real problem is trust. Some of the gap between any two teams is plain luck, and luck has fewer games to even out in a short season. By my estimate, over 162 games about a quarter of the spread in run differential between teams is luck. Over the 1981 strike season’s 105 or so games, it’s about a third. Over 60 games, it’s about half.

For what it’s worth, the Dodgers backed it up two years later. In 2022 they had the best unadjusted index score of any team in the database.

One more thing about that schedule adjustment. While I was writing this post, I found a hole in it. The adjustment only covered 1997 and later, but part of what it measures applies to every season ever played: a team never gets to play itself. A team with a great pitching staff has hitters who spent the whole year facing weaker-than-average pitching, and a team with a great offense has pitchers who never had to face it. Since 1997, the best teams were being docked for that. Older teams weren’t.

So I ran the adjustment for every season in the database, using the real schedules back to 1969. It shakes up the list more than I expected, especially among the great Japanese teams and the deadball clubs. Next week I’ll show you the new rankings.

How much does it matter? A lot for one team, not much for anyone else. Should a 60-game season count fully, count for less, or carry an asterisk? Let me know what you think.

Does Beating Up on Bad Teams Count?

Bad teams warp the index in three different ways, and they don’t all push in the same direction.

They stretch the yardstick. This is the 1939 story. The Browns and the A’s each allowed more than 1,000 runs, which stretched the American League’s spread so far that the Yankees’ 244 fewer runs allowed than average counted as only 1.6 steps. This hurts the 1939 Yankees.

They pad the totals. This one has bothered me for a long time. I think of it like a college football team deliberately running up the score at the end of the game in hopes of getting additional votes in the polls. It just feels wrong.

In 1939, runs scored against the Browns and A’s count exactly the same as runs scored against the Red Sox, and nothing stops a team from adding six more when it’s already up by 15. In fact, in the space of only three games that season, the New York Yankees added 62 runs to its run differential. This came from 21–0 and 23–2 over the A’s, and a 22–2 at Detroit. That accounts for about 15% of their famous +411 run differential that season. And, of course, a 21-0 victory is worth exactly the same as a 1-0 victory in the standings.

To be fair, those runs aren’t meaningless. Obviously, a 21-0 victory means that you’ve got a good team. But is it significantly better than a 10-0 victory? One way to test this would be to run these numbers again, but to put a cap on all margins of victory - perhaps at 10 runs. But, then again, you’d also have to consider the run environment in which each team plays, since it was almost certainly easier to win a game by 10 runs in the 1930 National League than in the 1908 National League. Anyway, this is still a theoretical mess, and it’s a subject we’ll come back to later.

In 1939, these first two problems partly cancel out. The Browns and A’s shrank the Yankees’ score on the yardstick and padded it on the runs. I’m not sure if the Yankees really come out even on the deal, but it’s good enough for now.

Every team is judged only against its own league. The index can’t tell a strong league from a weak one. The 1890s National League, the Federal League, the wartime seasons and the Japanese leagues all sit on one list, each measured in its own units. The schedule adjustment helps with the strength of opponents within a league, but it can’t tell a strong league from a weak one.

Imagine what would happen if we added the 1987 Pioneer League to this list. It’s possible that dominant teams from that “rookie level” league might look like some of the greatest teams of all time. You’d probably be upset with me if I started printing stats like that.

There’s a strong argument on the other side, too. In his classic book Full House, Stephen Jay Gould argued that as the talent pool deepens, teams and players bunch closer together, so standing out gets harder over time. Measuring in steps is meant to give modern teams credit for exactly that.

How much does it matter? This one is mostly a matter of taste, with one piece I can actually test.

A Few Smaller Complaints

• Runs aren’t wins. The index gives no credit for pennants, one-run games or October.

• Ballparks. A hitters’ park inflates both runs scored and runs allowed. The total mostly washes out, but the split between offense and defense gets tilted. The index doesn’t adjust for parks.

• “Steps” aren’t odds. A score of 3 doesn’t mean “a one-in-a-thousand team.” The size adjustment does lean on bell-curve math to judge how rare a team is, which is part of why it balloons near the ceiling.

• Multi-year runs. How should a three-year or five-year stretch be measured? I’m going to look at that later on.

So Where Does That Leave the Index?

I don’t think the index is broken. When it comes to the run-differential question, it’s mostly taste - though I admittedly agree more with the results of the classic Neyer-Epstein calculation than with the run differential calculation. The two places where my own fixes deserve the hardest look are the small-league adjustment and the 2020 schedule adjustment. And the blowout question is one I’d like to put to the test later on.

The index isn’t finished, either. It keeps growing. I’m working through problems with minor league statistics so more leagues can join the list. Once I’ve got a little bit more time on my hands, I’m going to start digging into the old International League, and I’ll see if I can overcome its lack of runs scored and runs allowed totals on the team level. Of course, the more leagues I add, the more these questions matter: small leagues, short seasons, and leagues of very different strength. And that’s a problem we’ll have to address as we go.

You Decide

For discussion: Which of these problems bothers you most? And is there a team you think the index treats unfairly, in either direction?

Some of the best comments each week end up on the podcast and here on the blog, so make your case.

Leave a comment

Coming up next Thursday: The Domination Index, rerun with a schedule adjustment for every season. We’ll move the worst teams post back a week.

You can check it out here:

The short version: History Maker grades pitchers mostly on ERA. ERA already includes the defense behind them. Then the game rates the fielders too. So great defenses get counted twice, and the 1981 Astros come out stingier than they ever were.

That’s the why. This post is about the what. If you play History Maker and want your replay to look more like the real season, here’s what our 42,000 games say you can try tonight. No new cards needed.

How to read the labels

• Tested: we ran it in our main setup, 20 seasons of 1981 with real playing time and trades.

• Tested earlier: we ran it in our first, simpler setup. We expect it to hold, but we haven’t rerun it yet.

• Untested: an idea that comes straight out of the results. That’s where you come in.

1. Play without the fielding ratings

Tested. Easy: just ignore GOLD and IRON.

This is the big one. Houston went from 31% stingier than its league to 25%. The real Astros were 23%. Across all 26 teams, the typical miss on runs allowed fell from 0.43 runs a game to 0.23. League scoring barely changed. The runs just moved to the right teams.

The cost is flavor. A Gold Glove shortstop should feel like one, and this rule takes that away.

2. Skip the SEMI-GILDED adjustment in 1981

Tested earlier. Easy: leave it out.

The set suggests SEMI-GILDED for the 1981 NL, plus a die roll for it before each AL game. It hands every pitcher a few extra singles. In our tests, every step toward GILDED added runs to a league that already scores too many. No adjustment was the most accurate setting.

Our guess is that the 1981 cards already carry 1981’s low scoring on both sides. GILDED makes more sense for cross-era games, like a 1981 staff against a 1930 lineup. That’s our reading, not the rulebook’s.

3. Turn off the colored mini-charts, if you want the stat sheet right

Tested earlier. Easy: skip them.

With all three mini-charts off, league scoring landed within about 0.2 runs a game of real 1981, and pitchers’ ERAs lined up better with their real ones. The standings didn’t change at all.

The cost is real. The charts are a big part of the game’s personality, and with them off, batting average runs a few points low. So this one depends on why you play. Playing for the stories? Leave them on. Playing for the stat sheet? Turn them off.

4. Keep your pitchers as they are

Tested. The easiest of all: change nothing.

We tried regrading every pitcher on FIP, a stat that leaves the defense out. It sounds like the perfect fix, and it was a mess. 246 of 331 pitchers changed grades. Dennis Eckersley, with a real 4.27 ERA, became a STAR. Ryan’s ERA jumped to 3.27. If you’re tempted to regrade the staff, don’t. Fixing the gloves does more.

Experiments for you to try

These are untested, and they’re yours. In our replays, a GOLD rating was worth about 28 runs a season, against about 16 runs of real defense. So the gloves count for too much. Our fix was to ignore them, but there may be gentler ways to turn them down.

A pitcher’s walks are also already in his ERA, and his CONTROL rating cuts them again. So control pitchers may beat their real ERAs.

If you find a house rule that fixes either one, we’d love to hear about it. Try it for a few games, and tell us in the comments how it went. We’ll run the best ideas through the computer and report back.

What I’d use

For my own 1981 replay, I’d drop the fielding ratings and skip SEMI-GILDED. I’d keep the mini-charts, because I play History Maker for the stories.

Let’s argue about it

Have you tried a house rule for GOLD and IRON? Did it work? And would you give up the mini-charts for a more accurate stat sheet?

The best comments this week could end up on the podcast and here on the blog, so don’t be shy.

Leave a comment

So tell me if I’m wrong.

Coming up

We’re going to keep testing History Maker with other seasons, starting with a very different kind of year than 1981.

For the old argument behind all this, see How Much Of Baseball Is Pitching? (https://baseballreplayjournal.substack.com/p/how-much-of-baseball-is-pitching).

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. Every house rule in today’s post was tested over 42,000 games, and reader support is what keeps the dice rolling.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

In 1982, APBA sold a card set with no lineup sheet.

Most APBA sets include lineup sheets, of course. Those of us who are long time APBA Baseball fans remember getting the yellow (or sometimes orange) sheets with the cards. In fact, if you’re old enough you probably remember APBA sending out the lineups as a promotion for the new card set.

But this set was different.

This set is the Original Franchise All-Star set, or OFAS for short. And, over the next several months, we’re going to go through it one team at a time.

What Is OFAS?

OFAS covers the sixteen franchises that held their places from 1901 through 1952: eight in the National League and eight in the American. Each team gets 25 cards, including 15 hitters and 10 pitchers, totaling 400 cards in all. Hitting and pitching ratings come from one season per player. Fielding comes from the player’s whole career.

The set was first released in early 1982, and retailed at $20. That’s not much today, but it was pretty expensive for 1982. Then again, it was cheaper than buying a set of 16 Great Teams of the Past from APBA, which would have run you $24 back in 1982 - and you would have received 80 fewer cards for your money.

Why These Players?

Based on what I’ve been able to find, APBA creator J. Richard Seitz started selecting players for this set sometime in either 1980 or 1981. No player appears on more than one team. Each chosen player goes to the team where he spent most of his career, with a few exceptions. Jim Bunning is a Phillie, for example, even though he spent more time in Detroit. There are no one-season phenoms, and no player comes from before 1894.

Of course, controversies do remain. Take Cleveland, for example. Seitz chose Luke Sewell and Steve O’Neill to catch for the Indians. Jim Hegan was left out, but not because he wasn’t a great catcher. Rather, it’s because Hegan never really had a good enough offensive year.

Seitz focused on the most impressive offensive cards and largely ignored defense and other intangibles. And this tension between the card and the career runs through the whole series.

The Puzzle

In a 1991 APBA Journal article, David Lawrence described what he called “the All-Star Paradox.”

When a pitcher is given a good grade in APBA, that grade cuts into the singles he allows, but not the home runs. And so, when you’re playing in a league like OFAS, a bigger percentage of runs scored comes from extra base hits.

This has two effects. The first is that batting averages are generally depressed, since those singles are prevented by the pitching grades. Meanwhile, the great hitters drive in fewer runs in general, since there are fewer runners on base to drive in.

These are basic illustrations, of course, and not measurements. But, using modern technology, it’s possible for me to measure these effects more precisely. And that’s what we’re going to do in this series.

Where This Series Is Going

This series is going everywhere.

We’re going deep into the original APBA OFAS. We’re going to cover OFAS II and OFAS III, the two sequels that came out sometime around 1995.

And then we’re going to switch to other games to see how their versions compare.

It will take a bit of time, but, in the end, we’re going to have the greatest study of OFAS anywhere.

At the start we’re going to look at each of these OFAS selections in its 1982 context. We’re not only going to second guess J. Richard Seitz, but we’re also going to look for those special stories that we’d otherwise miss. You know, like Jim Hegan missing out on glory because his bat wasn’t sharp enough.

Coming up: the first team, on Tuesday, November 3.

If you were choosing the 1982 hitters yourself, which franchise would have given you the hardest decisions? Tell me in the comments.

Leave a comment

The best comments each week get featured on the podcast and here on the blog, so this is the post to tell me what you remember about the set.

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

Put his Strat-O-Matic card up against Tom Seaver’s, and he’d hit about 16 over a full season. That’s as many as Rose ever hit in a real one. He topped out at 16, in 1966 and again in 1969.

So what happened? Yesterday I asked Which Card Lied to You? This one didn’t lie. It just added when it should have multiplied.

Last week we looked at Bill James’s log5 method, and how it differs from the way dice games add a batter and a pitcher together. Check out Bill James on Baseball Matchups if you missed it. I also promised on Wednesday that we’d come back to the Bob Gibson problem: when a great hitter meets a great pitcher, who should win? Today we do it with real cardboard.

The Four Cards

These are four basic cards from the 1980 Strat-O-Matic set, issued early in 1981:

• Mike Schmidt, Phillies. 48 homers, most in the majors, and the MVP.

• Pete Rose, Phillies. He hit .282, with one home run in 739 trips to the plate.

• Tom Seaver, Reds. 10-8 with a 3.64 ERA, but he gave up 24 homers in 168 innings. That’s one every seven innings.

• Ed Whitson, Giants. 11-13 with a 3.10 ERA, and only 7 homers allowed in 212 innings. That’s one every 30.

If those pitching cards look odd to you, you’re not alone. This is the set that set off the great pattern controversy of 1981. I wrote about it in Strat-O-Matic’s 1980 Card Scandal. And, of course, this was the season of Strat-O-Matic famously sticking all the pitcher hit numbers in the same column on the advanced side, a scandal many of us have never forgotten about.

How the Cards Settle it

In the basic game, the white die picks the column. A 1, 2 or 3 sends you to the batter’s card. A 4, 5 or 6 sends you to the pitcher’s. Then the two red dice pick the row.

So every plate appearance is half batter and half pitcher. The two cards are simply added together.

Now count the home runs:

• Schmidt’s card has 14¾ home run chances out of 108. They’re all in column 1, on rolls of 6, 7 and 8.

• Seaver’s card has 5.8, in column 5.

• Rose’s card has none.

• Whitson’s card has none.

Here’s the catch. Seaver’s 5.8 chances are on his card. They’re there for whoever’s batting. Schmidt gets them, and so does Rose.

What log5 Says

Log5 multiplies instead. Remember the coupon from last week: a pitcher’s weakness is worth the most to the hitter who can use it.

Seaver gave up homers. Schmidt hit homers. Put them together, and log5 says the damage should compound. Rose couldn’t hit homers, so log5 says Seaver’s weakness should barely matter to him.

Here’s what each method gives over 600 trips to the plate. For comparison, the average National Leaguer in 1980 hit about 10.

Four Things Jump Out

• Rose vs. Seaver: 16 on the cards, 2 by log5. This is the slugger in the title. On the cards, Seaver’s home run chances don’t care who’s batting. Log5 says a whole season of Rose against Seaver would produce one or two homers, not sixteen.

• Schmidt vs. Whitson: 41 on the cards, 21 by log5. Here’s our Bob Gibson problem. Whitson was very hard to take deep, but on the cards he can’t touch Schmidt’s 14¾ chances. Half the rolls never reach Whitson’s card at all. On paper, the great pitcher barely slows the great hitter. Log5 says he should cut Schmidt’s homers in half.

• Rose vs. Whitson: 0 on the cards, 0.4 by log5. This is last week’s zero, on real cardboard. Neither card has a home run, so Rose can never homer off Whitson. Log5 says he’d do it about once every 1,600 trips. Rare isn’t the same as never.

• Schmidt vs. Seaver: 57 on the cards, 86(!) by log5. If only Michael Jack could hit against Terrific Tom every time up!

Actually, I’d take that 86 with a grain of salt. This is exactly the case where critics like Leslie Morey and Mark Cohen argue log5 overpredicts things. When two extremes meet like this, you start getting crazy numbers.

But the general direction is right. Seaver was actually in real trouble against Schmidt in 1980. Mike was 2 for 4 in 5 plate appearances against Seaver in 1980, hitting .500, driving in 1, and hitting both a double and a home run. Of course, the rest of the time Seaver had his number:

One more thing. Against an average opponent, the two methods mostly agree, because Strat builds its cards that way. Rose is where that breaks down. To give him just one homer a season against average pitchers, his card would need fewer than zero home run chances. The lowest a card can go is none. So even against an average pitcher, Rose’s card works out to about five homers per 600 trips, not one. The extremes are where adding goes wrong, and the extremes are where pennant races get decided.

Did Strat-O-Matic Know?

I think Strat-O-Matic knew all about this.

Look at Rose’s advanced card from the same set: it’s marked Power-W.

Seaver’s advanced card prints his home runs as N-HR.

Under the advanced rules, a W hitter who rolls an N-HR off the pitcher’s card gets a single instead. And, of course, all home runs on the pitcher’s card had a N-HR back in those days.

That fixes Rose vs. Seaver. But it’s a switch, not a dial. A hitter either gets the pitcher’s homers or he doesn’t.

Log5 is a dial. In toher words, when you use log5, every hitter gets exactly as much of a pitcher’s weakness as his own strength earns him.

And, of course, nothing on any card fixes Schmidt vs. Whitson.

So, back to the question: when a great hitter meets a great pitcher, who should win? Log5 says the pitcher’s edge and the hitter’s edge both count, and they multiply. Strat-O-Matic splits the plate appearance down the middle, so neither one can take away what’s printed on the other’s card.

For the geeks

Each column comes up 1 time in 6, and the two red dice give 36 combinations, so each card has 3 × 36 = 108 chances and a matchup has 216. A roll of 7 counts 6 chances, 6 and 8 count 5, and so on down to 1 for a 2 or a 12. A split like “HOMERUN 1-15” counts 15/20 of the row.

Schmidt: 5 × 15/20 + 6 + 5 = 14.75. Seaver: 4 × 14/20 + 3 = 5.8. So Schmidt vs. Seaver is (14.75 + 5.8) ÷ 216 = 9.5%, and Rose vs. Seaver is 5.8 ÷ 216 = 2.7%, or about 16 per 600 trips.

Log5 uses each man’s real 1980 rate per plate appearance. Rose: 1 ÷ 739 = 0.14%. Seaver: 24 ÷ 692 batters faced = 3.47%. NL average: 1,243 ÷ 74,060 = 1.68%. Plug them in and M = 0.28%, or about 1.7 per 600 trips.

Strikeouts and walks tell the same story, only smaller. Rose’s card has no strikeouts at all, so every one comes off the pitcher: 17 chances against Seaver (about 47 per 600) against log5’s 30. Rose struck out 33 times all year.

Let’s argue about it

Has a slap hitter ever taken your ace deep in a game that mattered? Did it feel like a bug, or just baseball?

Some of the best answers will get featured on the podcast and here on the blog, so tell me your story.

Leave a comment

Coming up: On Wednesday, we’ll ask how you’d fix History Maker Baseball. Next Monday, we start our latest deep dive into National Pastime (yes, we’re going back again). But don’t worry; log5 is going to come back very soon.

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. Reader support keeps old cards like these four coming out of storage and onto the page.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

Maybe it was a .230 utility man who hit like Ted Williams in your league. Maybe it was a 20-game winner who couldn’t get anybody out all summer. You checked the dice. You checked the card. Everything was right, and he still played wrong.

So here’s today’s question:

Which card lied to you?

Tell me three things:

• The player, the season and the game: APBA, Strat-O-Matic, Statis Pro, Replay, whatever you played.

• What he did for you, and what he did in real life.

• Why you think it happened.

I’ll go first.

My favorte crazy card is the original single column 1927 Lou Gehrig card, the old Great Team of the Past card.

Gehrig was great in 1927, but he sure wasn’t this great:

That 1-1-3-5-5-6-6 power has never been seen since in the world of APBA, to my knowledge. This card is basically the Holy of Holies.

Luck, or a lie?

Before you blame the card, remember how small a season is. A regular gets about 600 trips to the plate. Give a true .280 hitter a perfect card, and about 19 seasons out of 20 he’ll finish somewhere between .240 and .320. The other year, he’ll land outside even that.

So a .280 hitter who hits .245 for you may be telling the truth. A .245 hitter who hits .330 for you two years running probably isn’t.

And sometimes it’s neither luck nor a bad card. A card can be built perfectly and still play wrong, because of who it’s facing. Tomorrow I’ll show you exactly how that happens, with four 1980 Strat-O-Matic cards: Mike Schmidt, Pete Rose, Tom Seaver and Ed Whitson. One of them turns into a slugger he never was.

The best stories this week could end up on the podcast and here on the blog, so don’t hold back.

Coming up: Tomorrow, why Strat-O-Matic turned Pete Rose into a home run hitter. On Wednesday, we’ll look at how to fix History Maker Baseball.

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. This Journal runs on readers, and today it runs on your stories in the comments, too.

Leave a comment

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

He once said something to the effect that you tend to lose interest in a sports sim once you figure out how it works. On Wednesday I said I think he had it backwards. This week’s episode takes that argument a lot further.

If you missed it, Wednesday’s post on what makes Strat-O-Ma…

Read more

The card

Every Friday, we’re going to take one of these “Great Stories” cards and go looking for the game behind it.

The photos are Charles Conlon’s. He was a New York newspaper proofreader who photographed ballplayers as a hobby from about 1904 into the early 1940s. The Sporting News later owned his negatives, and starting in 1991 an obscure company called Megacards turned them into one of the best-looking sets of the junk wax era.

The story

Davy Jones told this one to Lawrence Ritter for The Glory of Their Times in the 1960s. It’s on the back of the card, so give it a read.

Here’s the short version. In 1902 Chicago had a left-handed pitcher named Jimmy St. Vrain who batted right-handed, and badly. One day against the Pirates he struck out his first two times up. Manager Frank Selee joked that a lefty pitcher ought to try hitting lefty. St. Vrain took him seriously. He tapped a slow roller to Honus Wagner at shortstop, then took off for third base.

It’s one of the funniest stories in baseball. Is it true?

Finding the game

This turned out to be easy, because St. Vrain barely had a major league career.

He pitched 12 games, all in 1902, between April 20 and June 14. He started against Pittsburgh exactly twice.

• April 27, at Chicago. We can rule this one out. Davy Jones wasn’t on the team yet. He was still playing for the St. Louis Browns, and he walked out on them on the morning of May 14. He played center field for Chicago that same afternoon.

• May 30, at Pittsburgh. If it happened, it’s got to be this game. It was also the only game St. Vrain ever played in Pittsburgh. That matters, as you’ll see in a minute.

If this happened at all, it happened in the morning game on Decoration Day, 1902.

The game

Back then, Decoration Day was always May 30, whatever day of the week it fell on. In 1902 that was a Friday, and the Pirates sold separate tickets for a morning game and an afternoon game.

The Pirates were the best team in baseball by a mile that year. They went 103-36 and won the pennant by 27½ games.

About 7,000 people saw the morning game. The pitchers were two lefties, Ed Doheny for Pittsburgh and St. Vrain for Chicago. The Pirates scored four in the fourth, helped by two errors by a 21-year-old shortstop named Joe Tinker, and three more in the sixth. Chicago came back with five runs in the seventh. St. Vrain was hit by a pitch to load the bases in that rally, and he came around to score. Fred Clarke pulled Doheny and brought in Jack Chesbro. Chicago loaded the bases again in the eighth and the ninth, and couldn’t get the hit. The last chance went to a pinch-hitter batting for St. Vrain: Frank Chance. He struck out. Pittsburgh won, 8-6.

Now look at the box score.

• Wagner, ss. Honus Wagner played shortstop that morning. This was the part I worried about most. In 1902 Wagner wasn’t the Pirates’ regular shortstop; Wid Conroy was. Wagner spent most of that year in the outfield and at first base. But Conroy was out of the lineup that day, so Wagner moved to short.

• One assist. Wagner had exactly one assist all morning. So he threw out exactly one runner. If St. Vrain’s roller went to Wagner, that’s the play.

• No strikeouts for St. Vrain. The box lists every strikeout by name. Doheny struck out five Chicago batters and Chesbro struck out four. St. Vrain isn’t one of them. He came up four times: a single, the hit-by-pitch, and two outs that weren’t strikeouts.

That last point is a problem for the story on the card. St. Vrain never struck out that day, let alone twice.

Here’s a fun one. The Pittsburg Press wrote that “in the final chance, St. Vrian fanned.” It even misspelled his name. The box score and the Chicago Tribune both say Frank Chance batted for him in the ninth and struck out. So even the next day’s paper got the details wrong.

The 1906 version

Ritter’s book wasn’t the first time Jones told this story. The Detroit Free Press printed it on March 17, 1906, just four years after the game.

In that telling, there are no strikeouts and no wisecrack after the second one. Selee suggests switching sides, and “St. Vrain fell for the suggestion, and went to bat that way the first time up. He managed to meet the ball, tapping it to Hans Wagner. But he forgot about his shift, and tore down the line for third base.” Wagner “almost forgot to make the play to first.”

And where did it happen? “One day, in Pittsburg.” St. Vrain only ever played one game in Pittsburgh.

So the 1906 version fits the box score: the right city, the right shortstop, one assist for Wagner, and no strikeouts needed. The 1966 version doesn’t fit.

Even in 1906, though, the Detroit writer wasn’t buying it. He suspected Jones had been reading Ted Sullivan’s book of ballpark yarns. The headline was “Davy Jones Resurrects Old One.” Jones then followed it with a story about a ball getting stuck in a tomato can, and the writer said they were “rapidly losing faith down here in the entire truthfulness of the young man from Tomah.”

What the 1902 papers said

Nobody wrote up the play. I went through the Pittsburg Post, the Pittsburg Press, the Chicago Tribune and the Chicago Inter Ocean for the next morning. None of them mention St. Vrain batting left-handed or running the wrong way.

Two lines in the Tribune‘s “Notes of the Colts” are worth reading, though. The first: “St. Vrain is the poorest hitter in the league according to the form shown here.” That’s an odd thing to say about a man who singled and scored that morning. The second: Selee “was disposed to rap St. Vrain for his pitching. He blamed him for helping to lose in the morning.” Selee wasn’t happy with “Jeems” that day.

The story says / The record says

Verdict: Plausible, but embellished. If it happened, it happened in the morning game on May 30, 1902, and the box score has room for it. The version Jones told in 1906 holds up. The version he told Ritter 60 years later added two strikeouts that never happened. What keeps this from “confirmed” status is that nobody in 1902 wrote it down.

About the photo

The photo on the front is from 1912, ten years after the story, when Jones was playing for Detroit. Conlon worked in New York, so it was taken during a Tigers trip there. 1912 was a lively year for Jones. He was the teammate who reportedly egged Ty Cobb on before Cobb went into the stands after a heckler at Hilltop Park that May.

One more St. Vrain story

St. Vrain has one more legend. In Tacoma in 1904, his manager swore St. Vrain was on second base in a close game when a teammate lined one over his head, and St. Vrain reached up and caught it. We’re still digging on that one.

By the way, St. Vrain was out of baseball by 1906. The St. Joseph club released him on June 27, 1905. A Des Moines paper’s notice of his release said “in his day, St. Vrain was one of the great luminaries in the baseball firmament.” (Sioux City Journal, June 28, 1905, p. 7)

Let’s argue about it

The Glory of Their Times is one of the best baseball books ever written, and it’s full of stories like this one. How much of it do you take at face value? And which old-timer’s story would you like to see us check next?

Leave a comment

Sources

• Pittsburg Post and Pittsburg Press, May 31, 1902

• Chicago Tribune and Chicago Inter Ocean, May 31, 1902

• Detroit Free Press, March 17, 1906, p. 10

• Sioux City Journal, June 28, 1905, p. 7

• Baseball-Reference: box scores for May 30, 1902, and the St. Vrain and Davy Jones game logs

• SABR BioProject biographies of Jimmy St. Vrain, Davy Jones and Honus Wagner

• Lawrence Ritter, The Glory of Their Times (1966)

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. The Conlon card above is from my collection, and supporters help me keep adding to it.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

When Rob Neyer and Eddie Epstein wrote Baseball Dynasties back in 2000, they picked the 1939 New York Yankees as the greatest team of all time.

If you’ve never heard of that book, you can read my review here.

It’s hard to argue with Neyer and Epstein. After all, the 1939 Yankees went 106–45, won the pennant by 17 games, and swept Cincinnati in the World Series. That was New York’s fourth straight championship, which sure sounds to me like a dynasty. The Yankees did it after losing Lou Gehrig, who played eight games and never played again. Joe DiMaggio hit .381 in 1939 and won the MVP.

And then there’s the number that jumps off the page. The 1939 Yankees scored 967 runs and allowed 556. That’s a difference of 411 runs.

No team in my Domination Index database has a bigger run differential. Not the 1927 Yankees (+371). Not the 1890s Orioles of Wee Willie Keeler. No team in any major league season from 1893 through 2026 can compare with the 1939 Yankees.

And so, when I ran the 1939 Yankees through the Domination Index, I expected to find them at the top.

They came in 11th.

What the Domination Index measures

The Domination Index is my name for the “SD score” that Neyer and Epstein used in their book. It asks one simple question: how far did this team stand from the rest of its league?

It measures that distance in steps. A step is the normal spread between teams in that league that season. (For you geeks out there, statisticians call this a standard deviation; those of us who speak actual English can think of it as a “step.”) When the teams in a league are bunched close together, a step is small. When they’re spread far apart, a step is big.

A team gets two scores: how many steps better than average it was at scoring runs, and how many steps better than average it was at preventing them. Add the two together and you have the Domination Index.

Let me put this in context. When we stretch from 1893 to 2026 (yes, including this year’s stats), we have a total of 2,852 team-seasons (one team-season per team per season). And, of those 2,852 team-seasons, only 46 teams have hit a Domination Index score of 3.0.

The 1939 Yankees come in at 3.6.

Now, 11th out of 2,852 is still about as good as it gets. But it isn’t first. And why not?

The Yankees’ two scores

Scoring runs. The Yankees scored 967 runs. The average American League team scored about 800. That’s 166 runs better than average. In 1939, one step on offense was about 88 runs, so the Yankees were nearly 1.9 steps above the league.

Preventing runs. The Yankees allowed 556 runs. The average team allowed about 800. That’s 244 runs better than average, an even bigger gap than on offense.

But on defense, the Yankees were only about 1.6 steps above the league.

How can a bigger gap count for less? Because in 1939, one step on defense was a whopping 150 runs.

Blame the Browns and the A’s

Here’s the 1939 American League. Look at the bottom two rows:

The St. Louis Browns allowed 1,035 runs. The Philadelphia Athletics allowed 1,022. Two teams in the same league gave up more than a thousand runs.

Those two pitching staffs stretched the whole league apart. In a typical season, one step on defense is about 75 runs. In the 1939 American League it was twice that.

I’ll put it a little bit differently. There are 263 individual leagues in my database. The 1939 American League has the second biggest runs-allowed step out of all 263 of those individual leagues. The only one that was bigger was the 1899 National League, a league that featured the 20-134 Cleveland Spiders.

And so the Yankees’ pitching and defense were measured with a stretched yardstick. Allowing 244 fewer runs than average is spectacular. But in a league where two teams were each allowing more than 220 more than average, it doesn’t stand out quite as much.

Is that fair to the Yankees? I can see it both ways.

The case for the index: standing out is easier when your league is lopsided. If two teams are falling apart, being far ahead of the pack means a little less. The index is doing exactly what it was built to do.

The case for the Yankees: they didn’t pick their opponents. A team that allows 556 runs in a league that averages 800 is doing something remarkable no matter who else is in the league. Maybe the 1939 Yankees would have done the same in 2026.

And here’s the catch: the Yankees didn’t just play against a stretched league. They feasted on it.

Sunday at Shibe Park

On Sunday, August 13, 1939, the Yankees came to Philadelphia for a doubleheader. They were 72–32 coming into the series. The A’s were 36–69 and had lost eight in a row. Even so, 34,570 people jammed into Shibe Park, and the gates closed early in the first game with thousands more turned away.

The first game was a slugfest, and the A’s won it. Frankie Hayes, the A’s All-Star catcher, hit a home run in the bottom of the ninth to win it, 12–9, and snap the losing streak.

Even the 1939 Yankees could give up 12 runs on a bad afternoon.

Then came the second game.

Red Ruffing took the mound for New York, walked three batters in the first inning, and got out of it without a run. That was the most trouble he was in all day. He allowed three hits, struck out five, and also had four hits of his own at the plate.

The Yankees scored five runs in the second inning, five more in the third, and kept going. Babe Dahlgren, the man who had replaced Gehrig at first base, hit two home runs. Joe DiMaggio hit two, one of them an inside-the-park job in the sixth, the other a drive in the eighth that cleared the roof of the left field stands.

The game never reached the ninth inning, by the way. Philadelphia’s Sunday baseball law required play to stop by 7 p.m., and the umpires called the game at 6:49. Final score: Yankees 21, Athletics 0.

It was the most lopsided shutout in the majors since 1901. In fact, the A’s were completely smothered:

That second game is the 1939 Yankees in a single box score: 21 runs scored and none allowed. That’s both halves of the Domination Index in one afternoon. And it came against one of the two teams that stretched the yardstick.

So were they the greatest?

On a single season, my index says the 1939 Yankees were one of the dozen most dominant teams ever, but not the most dominant. Ten teams stood farther from their leagues:

1. 2020 Los Angeles Dodgers: 4.20 (60-game season)

2. 2022 Los Angeles Dodgers: 3.91

3. 1917 New York Giants: 3.89

4. 1906 Chicago Cubs: 3.88

5. 1902 Pittsburgh Pirates: 3.86

6. 1927 New York Yankees: 3.79

7. 1998 New York Yankees: 3.74

8. 1923 New York Yankees: 3.73

9. 1984 Detroit Tigers: 3.71

10. 1986 New York Mets: 3.71

11. 1939 New York Yankees: 3.61

And, yeah, I know that the 2020 season only lasted 60 games. We’re going to talk about that later.

But the 1939 Yankees didn’t happen in a single season. Their 1937–39 stretch ranks 3rd among all three-year runs in my database, and their 1935–39 stretch ranks 3rd among five-year runs. We’ll come back soon to these Yankees and ask a different question: were the 1936–39 Yankees one great team, or several?

For discussion: Should a team get less credit because the bottom of its league was awful? Or is the index right to measure the 1939 Yankees against the league they actually played in? Let me know in the comments.

Leave a comment

Coming Up Soon: the worst teams in the Domination Index. Hint: one of them played in Philadelphia in 1939 - and it wasn’t Connie Mack’s team.

For The Geeks

The 3.61 in the table is the adjusted version. It compares each team only with the other teams in its league, and it puts eight-team and sixteen-team leagues on the same scale. For the 1939 Yankees, the adjustment adds about a tenth of a point.

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it. Supporters keep the digging going: the Domination Index behind today’s post, and the 1939 newspapers behind that Sunday at Shibe Park.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

June 12, 1981, was a day that lived on in infamy.

That was the day Major League Baseball players walked off the job. Big-league baseball disappeared for seven weeks.

WITS in Boston, the principal station of the Boston Red Sox, kept broadcasting anyway. Ken Coleman and a young Jon Miller sat in a studio on Lansdowne Street and called Red Sox games played out with Strat-O-Matic dice. Miller even did the Sherm Feller public-address voice. In Minneapolis, the Star printed Strat box scores almost every day of the strike.

I wrote more about that summer in The Year Strat-O-Matic Became a Household Word; check it out if you missed it.

Think about that for a second. Fans would rather listen to cardboard and three dice than to nothing at all.

Anyway, last year I made a video asking why. That video focuses on the basic game, not the advanced rules, because I think that’s where its charm shows best. And you can check it out here:

What’s in the video

• How a single roll works. The white die picks the column, and the other two dice pick the row.

• Why you shouldn’t trust whatever jumps off a card. Mule Shirley’s 1924 card has a triple sitting in a spot that comes up less than once every 200 plate appearances. Shirley only had 87 plate appearances in real life that year; good luck.

• The 20-sided die. It let Hal Richman fine-tune results well beyond the double-column innovation J. Richard Seitz brought to APBA in the late 1950s. The 20-sided die was really Strat-O-Matic’s “killer app,” and makes its appearance in one form or another in every single one of the company’s games.

• The “lucky column.” Most Strat hitters and pitchers keep their best results bunched in a single column: Maris in 1, Mantle in 3, Frisch in 2.

• The X chart. Only about a third of the results come off the pitcher’s card. That’s the secret reason why the 1968 Bob Gibson can never be quite as dominant in Strat as he was in real life.

For the nerds: a 7 comes up on two dice 6 times in 36, and any one column comes up 1 time in 6. So any single row-7 result shows up 1 time in 36, or 2.78%. A 2 or a 12 shows up 1 time in 216, or 0.46%.

The part I keep thinking about

The original Strat roster sheets didn’t come with recommended lineups, so you had to set your own. That’s a huge departure from APBA, where every season includes a sheet that listed the most commonly used lineups. And that bred a completely different style of gamer.

In the video I argue that a whole generation of Strat players learned, by fiddling with their batting orders, that putting the guys who get on base at the top works. They learned that years before anyone had heard of some guy with a beard named Bill James, let alone anybody called the science sabermetrics.

Of course, many of you were actually there, while I wasn’t born yet. So tell me if I’m right.

Since the video

• I hadn’t come across the WITS broadcasts when I made it. They deserve a post of their own, and if anyone has a tape, I’d love to hear from you.

• A small fix: David Madsen’s 3.24% figure for shortstop X-chart chances is per plate appearance, not per at-bat.

Let’s argue about it

Bill James once wrote that you tend to lose interest in a sports sim once you figure out how it works. I think he had it backwards: once you crack the code, you start trying to fix the game.

So: did figuring out a game ever kill it for you? Or did it hook you for good?

A bonus question: in your first Strat or APBA league, who batted leadoff, and why?

Leave a comment

Coming up: We’re going to dive deep into that Bob Gibson problem in next Monday’s post, where we’ll ask who should win when a great hitter meets a great pitcher. We’ll then look next Wednesday at History Maker Baseball, which is like Strat-O-Matic without the numbers.

The Baseball Replay Journal is free, and it always will be. It keeps going because readers chip in $5 a month to support it.

Become A Supporter

Rather leave a one-time tip? Buy Me a Coffee · PayPal. Tips go to the Acquisitions Fund for old magazines, game sets, and reference books.

Subscribe now

If you’re anything like me, you’ve always wondered who wins when a great hitter faces a great pitcher.

You know the drill. What happens when a fastball hitter faces a guy with a 103 mile per hour fastball? What happens when you get a crafty lefty with good control up against a guy with a good eye?

The board games we grew up with gave us a pretty simple answer to this problem. In most of these games you roll the dice, and half the time you read the batter’s card, half the time the pitcher’s. The result lands somewhere in the middle. It’s fair, it’s easy, and it has worked well enough for generations.

But in 1981, Bill James offered a better answer. He called it Log5. Let’s take a look at what it is, what makes it different, and why you should care.

What Is Log5?

Log5 is a way to predict what happens when two sides meet, using only how each side does against an average opponent.

Now, wait a second. That sounds like Strat-O-Matic, right?

Yep, it does sound like Strat-O-Matic baseball. Except that it’s not the same. The math works differently.

Bill James first wrote about this idea in the 1981 Baseball Abstract. Unfortunately, those of us who are not millionaires have no way to read his original words. James explained it again on pages 12–13 of the 1983 Bill James Baseball Abstract, the mass-market edition most of us actually own. And this isi what he had to say:

I’ve gotten better at taking these photos, right? I unfortunately don’t have a high quality scan of this book, and I’m not willing to sacrifice my precious copy for the sake of a blog post, so this is what we’re going to get.

Buy Me A Coffee

Anyway, Bill James started with teams. So let’s look at it that way. Suppose a .600 team plays a .400 team. How often does the better team win?

Most of us would guess 60%. Log5 says about 69%. Here’s the reasoning in plain words:

• A .600 team wins 60% of the time against average teams.

• A .400 team is worse than average.

• So against this opponent, the good team should win more than its usual 60%.

For those who like a pencil, here is James’s formula, where A is the better team’s winning percentage and B is the other team’s:

If you plug in .600 and .400, you get this: (.600 − .240) ÷ (1.000 − .480) = .360 ÷ .520 = .692.

James’s colleague Dallas Adams (who was a long time contributor to The APBA Journal) later stretched the idea to individual stats, like batting average or home runs, where the league average isn’t .500. That’s the version that matters for batter against pitcher.

Adding versus multiplying

Now comes the interesting part.

The old dice-game way (i.e. Strat-O-Matic) adds the batter’s and pitcher’s strengths together. Log5 multiplies them.

So what?

Well, think of it this way. Think of two store coupons, each worth 50% off. If you were going to add them together, you’d get your thing for free, right? I get 50% off the $100 retail price of that box of baseball cards with this coupon, and 50% off the same retail price for the same box with that coupon. Makes sense, right?

Any cashier will tell you that’s not how it works. Instead, you’d see 50% off with the first coupon, and then 50% off the new price with the second. That’s the multiplication method.

Log5 treats batters and pitchers the way the cashier treats coupons.

An Example

Let’s use a totally made up example. Let’s say we’ve got an imaginary league where the average hitter homers once every 36 trips to the plate.

We’ll look at four different kinds of players:

• A slugger who homers twice as often as average.

• A gopher-ball pitcher gives up homers twice as often as average.

• A slap hitter homers half as often as average.

• A stingy pitcher who gives up homers half as often as average.

PayPal Donation

If we use the Strat-O-Matic method of adding, we add together the extra home runs by the hitter with the extra home runs by the pitcher. We wind up getting three times the average. In other words, we start with the league average, we add on the extra helping of home runs from the hitter, and we add the extra helping from the pitcher.

Under Log5 we multiply and divide to get our finding.

Now, this is going to make us use the Dallas Adams version of the formula. But it’s pretty easy to understand. It’s the exact same concept that Bill James explained, and there’s a shortcut to help this make sense.

The matchup rate is M. We take the batter’s rate of home runs (or whatever we’re trying to figure out), multiply it by the pitcher’s rate of home runs, and divide it by the league average. Here’s a shortcut version of the formula:

B is the batter, P is the pitcher, and L is the league average.

With a slugger and a home run happy pitcher, Log5 gives us four times the average once the math shakes out. That’s because two times two is four.

Now, if you look at the other side, the part where you’ve got a bad hitter and a really good pitcher, the additive method runs us into a wall. Subtract the slap hitter’s shortfall and the pitcher’s shortfall, and you hit zero.

If you use Log5 instead, you get that half times half is a quarter effect. That’s what we just saw with the coupons, of course. That gives us an event that is rare, but is still possible.

This chart might help you visualize all of this:

If we’re talking about a single trip to the plate, well, all bets are off. But, if we’re looking at 600 trips to the plate, the slugger hits about 50 home runs off the gopher ball pitcher using the added method, but about 65 using the Log5 method.

One more thing to notice: when either player is plain average, the two methods give exactly the same answer. Or, in other words, this only helps us out at the extremes.

Why anybody should care

So why do we care about this?

Because the extremes are the part of baseball we remember.

Nobody replays a season to watch average hitters face average pitchers. We replay it for the slugger who feasts on a bad pitching staff, and for the ace who makes a singles hitter look helpless. And those are exactly the matchups where the adding method squeezes everyone toward the middle.

The zero problem is real, too. A game that adds can’t let a weak hitter homer off a stingy pitcher at all. Card makers have had to invent workarounds to fix that. In fact, when we come back to this topic again in a few days, we’ll take a closer look at a real life example.

Log5 is also the starting point serious researchers use when they study batter-against-pitcher matchups. Some call it the “odds ratio” method, something you might have read about elsewhere.

For example, Bill James used it to ask how Barry Bonds’s 73-homer season of 2001 would have looked in the National League of 1920. His answer: about 16 home runs per 500 at-bats. The real 1920 leader, Cy Williams, hit 15.

Now, to be fair, Log5 has critics. In 2015, two researchers, Leslie Morey and Mark Cohen, argued that it overshoots when two extreme players meet. Bill James himself wrote a pointed reply - and imagine what it would be like to have the Guru himself weigh in to argue with you. We’ll go over all of this later.

What Comes Next

When we come back to this topic we’ll put real cards on the table. I’ve got a real world example for you that I’m sure you’ll enjoy, using Strat-O-Matic cards.

Leave a comment

Subscribe now

I’ll make this short and sweet.

The Los Angeles Dodgers of the last 10 years have been the greatest dynasty the sport has ever seen.

It doesn’t really matter how you measure it. As long as you’re focusing on the regular season, this is an undisputed fact. The Dodgers are the supreme dynasty, especially if you look at it decade by decade.

Buy Me A Coffee

I’ve done a little bit of fiddling around with data over the past few days. And the Domination Index makes it quite clear that the Dodgers have dominated baseball like no other team before them.

If you count teams from every league, including both Japanese leagues, the 2017-2026 Dodgers were still absolutely the most dominant team in baseball history. Here’s the top 25:

Some of these teams will be familiar to you, of course. The Dodgers were better than both of the unstoppable New York Yankees runs on either side of World War II. The Cardinals of the 1940s can’t touch them. The great orioles teams of the late 1960s and early 1970s also can’t touch them.

PayPal Donation

The Dodgers of the past 10 years were greater than the most legendary Giants teams under John McGraw. They were better than the old days of the $100,000 infield of Connie Mack. They were a lot better than the Yankees in the early days of Steinbrenner. And they were a lot better than the amazing run the Astros had not long ago.

They were even better than the 1904-1913 Cubs, which includes the incredible 1906 Cubs team. That was a team that played magnificent baseball for several years.

This is the team the Milwaukee Brewers swept.

There’s a lot of talk out there about whether this changes the discussion about a potential salary cap. It shouldn’t, in my opinion. The Dodgers were completely dominant because they were able to exploit a loophole that was created for themselves, one that really needs to be dealt with one way or another.

I don’t have the right answer necessarily. But I do know dominance, and I’m absolutely sure that the Dodgers dominated.

I’ve got more on this topic coming before too long.

Leave a comment

Subscribe now

You’re going to like what is coming soon. Here’s a quick preview. Stay tuned!

Leave a comment

Subscribe now

I had an email the other day from somebody who wanted to share their latest web based baseball management sim with me.

This game is called Ninth Inning Drill. It’s free; you can find it on this site. It’s entirely internet based, but it’s still pretty interesting.

Buy Me A Coffee

I’ve only messed around with the “daily” scenarios. However, I can tell you that they’re pretty interesting.

Basically, this game takes situations directly from Retrosheet records, puts you in the difficult situation, and see how you do.

I haven’t talked with the creator all that much about how this game determines results. The choices you have are usually limited: maybe you can bring in a pitcher or two on defense, and maybe you can bring in a pinch hitter on offense. You can also always bunt or call a hit and run.

PayPal Donation

It’s not clear what the “optimal” settings are for each circumstance. That’s a good thing, since it gives you a bit of “fog of war.” It’s also confusing, though, because it usually is “optimal” to call for a hit and run, though precisely how that is supposed to work in every situation isn’t clear.

There also seem to be a few bugs. I got the first scenario wrong and we wound up not winning the game, but I was told that my answer was somehow “right.”

Having said all of that, I do think this is a really cool system. This is something I’ve been hoping to see with general baseball sims. It would be amazing to have a game that allows you to play out a certain situation in a certain game rather than just having you replay the entire season.

Anyway, let me know what you think. And, if you’ve got your own project, please contact me and tell me about it. I’d love to feature it!

Leave a comment

Subscribe now

I came across this Jeff Passan article the other day.

Passan’s article is a bit long, and I think he skirts around some of the important points. However, rather than drafting an actually coherent response, I’ll write a few general notes here in the interest of time.

The 2026 Brewers Were More Dominant Than The Dodgers

Despite our englihtened sabermetric ways, we still live in a world where most baseball analysis at the team level begins and ends with won-loss records.

The 2026 Dodgers won 100 games and lost 62. The Brewers were three better, winning 103 and losing 59.

Most analysts would look at these numbers and instantly conclude that the Brewers were only marginally better than the Dodgers.

However, if you look at the Domination Index, you’ll see that the Brewers were actually quite a bit better than the Dodgers:

Of course, the fact that the Dodgers are coming off the most dominant decade in baseball history (and I really mean that; I’m not exaggerating) likely has a lot to do with their postseason expectations.

The Yankees Lost Because Of Injuries

If you look at the American League Domination Index for 2026, you’ll notice that the Yankees were on top:

Of course, the Yankees were snakebitten by injuries near the end of the regular season, which is likely the biggest reason for their lac of postseason success.

Buy Me A Coffee

Cleveland, meanwhile, is an interesting story in that it was actually a bit below average in terms of general domination in 2026. I haven’t crunched all the numbers yet, though my guess is that there aren’t a lot of teams in baseball history who had postseason success despite playing this poorly in the regular season.

The Postseason is a Horrible Way to Measure Parity

I’ve got to be blunt here. Nobody in baseball cares about the regular season anymore.

Yeah, I know it’s sad. It wasn’t that long ago that people really cared about what happened during the regular season. However, when the owners unilaterally decided to push us into the NFL-esque world of three divisions back in the mid-1990s (and kicked out the Commissioner to be able to do so), they sucked away any meaning that the regular season had once had.

Baseball’s regular season means less today than it ever had in the past. And, if the owners do manage to expand the stupid wild card round to a 7 game series, we’ll see the regular season mean even less in the future.

The irony here is that the additional television revenue from expanding the postseason really isn’t worth the effort. The league gets a marginal increase at best from the extra games. Fans in general grow weary of the endless postseason rounds. Personally, I completely skipped the two wild card rounds, largely because I’m not interested in watching teams that played barely over .500 compete to advance to the next round.

The biggest problem, though, is that postseason success just feels random. A mistake here or an injury there becomes extremely significant in a short series. And, as fans of the old Moneyball Atheltics remember all too well, the stuff that works over the course of the regular season becomes irrelevant when you hit October.

The irony here, by the way, is that the expanded playoffs help the rich teams more than anybody else.

When the owners dragged us kicking and screaming into the wild card era, the New York Yankees had gone over a decade without making the postseason. Back in 1994 it had been 13 years since the Yankees had their most recent postseason appearance - and that appearance only came about because of the stupid split season approach to 1981. Had the owners decided on using the overall record to rank the teams in 1981, the Yankees would have finished 4th in the American League East.

In general, expanded postseasons tend to favor richer and more established teams. It serves as a sort of guarantee for larger teams. Depending on the sport, if you’re able to be one of the top 6 or 8 teams in your league on a consistent basis, you’re able to get the extra revenue that comes from postseason appearances. This is true in every single sport, and is probably most obvious in the world of European soccer, where the post-European Cup international club play format has allowed the biggest clubs to make exponentially more money each year than their smaller competitors.

Or, in other words, the ridiculous 24 team college football proposal will likely only benefit the Big Ten and SEC.

The Payroll Problem Is Real

Assuming that the Spotrac numbers are accurate, the 2026 postseason has been a remarkable Cinderella story.

The Guardians have an estimated $74,459,014 payroll, which is dead last in the Major Leagues. The Rays are only two spots above that, with $95,615,469. Amazingly, both teams are behind the awful Athletics, who barely qualify as a Major League club these days.

Milwaukee managed to have its success depsite a payroll of only $146,206,320, which ranks 19th.

The Dodgers, of course, finished second in payroll figures, with $339,319,244. The Mets were about $7 million higher than that, despite having an extremely poor season.

Now, you might look at this and conclude that payroll simply doesn’t matter. And there is a bit of truth to that. Payroll only matters when it comes to signing free agents, and the structure of Major League Baseball’s free agent system basically guarantees that most big money free agents are in their decline phase. The best teams have known for years that youth is the key to success, and the days of the Yankees splashing crazy cash on every big name in the game are long gone.

PayPal Donation

But there’s still a major problem here.

The problem is most apparent when it comes to teams like the Pirates, the Athletics, the Reds, and a number of other clubs that have been mediocre for years upon years. If you find yourself among the mediocre teams in the modern Major League Baseball economic system, you’ll quickly discover that there’s really no way out.

You want to either be horrendously bad because you’re rebuilding entirely with youth, or you want to splash the cash in an effort to compete with the big boys. If you’re middle of the road, chances are that you’ll simply be mediocre for years and years.

In the NBA there is a way to take a bad team and make it good quickly. The salary cap helps with this (though not all teams are capable of turning things around; look at the Kings for an example). In Major League Baseball, I’m not convinced that there are realistic ways for every team to turn things around.

“Large Market” and “Small Market” Is Meaningless

The “large market” and “small market” terms come from the discussions of the mid-1990s. And, honestly, they’re meaningless these days.

The only reliable “large market” teams in modern baseball are the teams in Los Angeles and New York. Every other team has the ability to claim “small market” status. That includes the Boston Red Sox, who spent like a “large market” team for years; the Philadelphia Phillies, who were 4th in payroll in 2026 (and finished at 88-74), the Toronto Blue Jays, who have a television deal almost as sweet as the Dodgers (minus the insane bankruptcy loophole) and who finished below .500, and quite a few other teams.

The truth is that the size of the metro market doesn’t matter as much in modern baseball as you might think. The big money comes from media deals, not from gate revenue. We know that this is the case, by the way. If this weren’t true, the owners wouldn’t be proposing shrinking the regular season.

To move the discussion forward, we need to stop talking about “large market” and “small market” teams.

We also need to stop trying to convince ourselves that the randomness of the postseason somehow fixes the obvious lack of parity in the sport.

Frankly, if all we care about is the postseason, we might as well cut the regular season by half or more. If the regular season is supposed to still mean something, then we should make changes to the sport to make it important again.

And, as I’ve said before, I firmly believe that the best changes would involve bringing back balanced schedules, eliminating the pointless interleague games, cutting back on the postseason rounds, and giving people a reason to watch regular season games again.

If we keep going down this road, even more fans will feel disillusioned by the sport.

Leave a comment