The best of BP 2025: one poke across the line | Baseball prospectus

There was, believe me, a time before the Internet. If you wanted to keep up with the news, you had to listen to the radio, watch TV or read a newspaper or magazine. If you followed baseball, the newspaper was the only place you could get daily updates on box scores and standings.

And if you wanted a more comprehensive list of baseball statistics than the occasional top ten list, you had to wait for the Sunday paper. There, buried in the sports section, there would be a list of batting statistics (at bats, runs, hits, home runs, RBI and batting average) for teams and for batters with more than a certain number of at-bats. They would be ordered on average. For pitchers, the ERA series would give you innings, hits, walks, strikeouts, wins, losses, and ERA, first for teams, then for all pitchers above a decision threshold (not inning!) Here’s a pretty garbled sample from Gainesville Sun on July 10, 1983.

Sorry that’s so hard to read. But you get the idea. Before Baseball Prospectus, FanGraphs, Baseball Reference, MLB.com and elsewhere, the Sunday newspaper was the place to get comprehensive baseball statistics. The sports news had them too, but you had to wait until it was on the newsstand or delivered to your mailbox. The Sunday newspaper was immediate: it contained everything during Friday evening’s matches!

In the American League, the best hitter mentioned above was California’s Rod Carewwho was hitting .403 at that point in the season. (He would finish at .339, second only to Wade Boggs‘ .361.) At the other extreme was Chicago’s Julio Cruz.234. However, the SunFor space limitations, you can cut the frame. Clevelands Gorman Thomasfor example, hit .191, but was not listed. Neither does Minnesota’s Tom Brunansky.199.

Newspapers from the larger cities, which had more columns, mentioned more names. So while Thomas and Brunansky weren’t mentioned in Gainesville, they were in the papers in the cities where teams play. And the players saw the numbers, just like the fans.

In 1979, short stop in Seattle Mario Mendoza was at or near the bottom of those lists. In the Sunday papers of May 13 of that year, his batting average was .202. It was .189 the next two Sundays, .186 the week after that, and then .185 in the Sunday, June 10 papers. (He finished the season at .198.)

Mendoza’s Mariners teammate Bruce Bend coined the term ‘Mendoza Line’, which indicates the batting average of the light-hitting shortstop. If you were worse then Mario Mendozaat the bottom of the list on Sunday you were below the Mendoza line.

When Kansas City traveled to Seattle for a four-game series beginning May 14, Mariners presumably left outfielder Tom Paciorek warned the royal family George Brett that he was at risk of falling below the Mendoza Line. (Brett was hitting just .257 at the time and entered the series with an 8-for-38 slump. He went 6-for-16 in the series and finished the year at .329.) Brett told ESPN SportsCenter host Chris Berman, and the term entered the baseball lexicon.

(Mendoza, by the way, ended his MLB career in 1982 with the Rangers, returned to his native Mexico, where he played for seven years, and was inducted into the Salón de la Fama de Beisbol Mexicano, Mexico’s Hall of Fame, in 2000.)

The Mendoza Line represents a batting average of .200. If you hit under .200, like you did Bo Naylor (.195) and Michael Conforto (.199) Among players with at least 400 plate appearances this year, you would be below the Mendoza line.

Of course, 1979 was almost half a century ago. At the time, Sunday newspapers reported batting averages because that was how batters were judged. We’ve come a long way since then. So at the suggestion of my friend José Hernández, editor of BP en español, I set out to define a new standard for ineffective hitting.

As you probably know, Bill James noted that teams’ on-base percentage is much better correlated with their run production than their batting average. I discovered that while James’ observation was true at the time he wrote it, slugging percentage has exceeded OBP in recent years. Here are the correlations between team runs and five commonly used batting metrics for the 810 team seasons in the 30-team era (beginning in 1998), excluding 2020.

Slug percentage has the advantage of being easy to calculate (just two numbers, total bases and at bats). But OPS and wOBA have better correlations. The problem is that not all fans know what OPS and wOBA are like they know AVG or even SLG. But OPS is mentioned more often in broadcasts and in articles than wOBA, so if we’re going to work with a simple rule of thumb for the audience, I think it’s better to use the more familiar figure, even if it’s a bit shy of wOBA’s precision.

The question then becomes: What do we use for the OPS equivalent of a .200 batting average? Specifically: What easy-to-remember value for OPS yields a level that almost no one falls below (but a few do)?

I looked at the different OPS levels in the five post-pandemic seasons and the number of players with at least 400 plate appearances who failed to achieve them.

Obviously .550 is too low. (The three players who had an OPS below .550 over the past five years are 2021 Jackie Bradley Jr. (.497), 2025 Nick Allen (.535) and 2022 Geraldo Perdomo (fourth in the NL MVP voting three years later: .547). Likewise, .650 casts far too wide a net.

Take the center of that distribution. I think the best figure is .590: an average of only two or three players, but almost never zero. It is an attainable, if not undesirable, standard. But .590 is an awkward number. .600, a nice round number, is not. So let’s go with .600. Any OPS under .600 is under… wait, we need to come up with a name for the rule!

The first player I thought of was Allen. He is the modern Mendoza: good field, no hit infielder. The problem is that he took the no-hit thing to the extreme. He never had an OPS above .600. He’s never even had an OPS above .550. His career average is .536. Mendoza’s career batting average was .215. You can name a batting average of less than .200 after a guy with a career mark of .215. You can’t name an OPS less than .600 after someone who never came close to that level.

Among active players with at least 1,000 plate appearances are the players closest to .600 Taylor Walls (.584), Sandy Leon (.585), Lucas Maile (.597), Billy Hamilton (.617) and Martín Maldonado (.620). The problem with that quintet is that Walls is more of a utility man than a regular and the others are nearing the end of their careers. Something about the Walls Line and the Maile Line sounds good, but may not be right.

There’s one more. (Bonus points if you get the reference.) Jonah Heim is, to be clear, a useful player. As recently as 2023, he generated 4.1 WARP on a .259/.318/.439 slash line, 18 home runs, 95 RBI and 17.1 defensive runs prevented. He’s a pretty good framer, and while his arm isn’t the best, it’s good enough to deter base thieves. All told, he has generated 9.1 WARP over just under five full seasons, always above replacement level.

But his play has declined since 2023. His defense hasn’t been as strong, and more importantly, his OPS has been an identical .602 in both 2024 and 2025. Almost .600 on the nose. And no matter how pleasant it is to the ear, the Walls Line or the Maile Line cannot compete with the Heim Line. It sounds a bit like hemline or timeline! It rhymes!

These are our 21^st century analogous to the Mendoza line. This year, Nick Allen (.535), Joey Ortiz (.593) and Ke’Bryan Hayes (.595) had an OPS below .600.

They were under the Heimlinie.

Make it so.