BOOG and the Hall of Fame, Part 1 (Introduction)

In 1889, Hall of Famer John Clarkson had one of the best pitching seasons ever by WAR. He won the Triple Crown with a 2.73 ERA, 284 strikeouts, and a win-loss record of 49-19. He also threw 620 innings, the most in the league and a total that represented more than half the combined workload of his team’s entire pitching staff. Although no one knew it at the time, his retroactively calculated ERA- was 64 and his FIP- was 82; both of those rates led the league among qualified pitchers. If the Cy Young award had existed in the National League at that time, he almost certainly would have been the recipient.

In 2025, Tarik Skubal had a dominant season. He didn’t repeat his Triple Crown achievement from 2024, but his ERA and strikeout numbers both improved from the prior year. His 195 1/3 innings were tied for fourth in the majors, and his 54 ERA- and 58 FIP- not only led among all qualified AL pitchers, but also significantly exceeded Clarkson’s 1889 performance. Skubal’s performance was more than good enough for him to win his second consecutive Cy Young award.

Clarkson’s pitching WAR was 10.9; Skubal’s was 6.6. Whose season did more to bolster his Hall of Fame case?

Baseball has a rich tradition of attempting to quantify as many aspects of player performance as possible, and the Hall of Fame is no exception. Not only are inductees’ plaques chock full of statistical facts and figures, but the process of determining Hall-worthiness has been analyzed by scores of sabermetric luminaries (as well as in countless bar room debates). To this day, a number of Bill James creations are listed on Baseball Reference pages, including the Black Ink Test, which compares the number of times a player led the league in various statistical categories to the average Hall of Fame career, and the Hall of Fame Monitor, which uses a point system to measure the likelihood that a player’s career will warrant a place in Cooperstown. It assigns points to players for reaching statistical benchmarks or achieving other notable criteria — such as winning MVP awards, making All-Star teams, being a regular starter on pennant-winning clubs — and then adds up the points. A player with a score of at least 100 is considered “a good possibility” to make the Hall of Fame, while a score of 130 or better suggests that player’s Hall of Fame case is “a virtual cinch.”

In more recent times, Jay Jaffe’s development of JAWS has become the gold standard for measuring a player’s statistical case for induction because it accounts for both longevity and peak performance. Specifically, it takes a player’s career WAR and averages it with the combined WAR of his best seven seasons. JAWS is widely cited in sabermetric circles and was likely a significant factor in the elections of Tim Raines and Larry Walker by the BBWAA on their 10th and final ballots (in 2017 and 2020, respectively) after years of Jaffe and others beating the drum for them.

In this article I present an alternative perspective on measuring Hall of Fame worthiness, one that takes the philosophical underpinnings of WAR as a concept and arrives at a unique and statistically valid approach to balancing career longevity and peak value, distinctly different from JAWS.

Let’s start with the obvious point: JAWS is a better Hall of Fame evaluator than career WAR. John Olerud has more career WAR than Todd Helton, but Helton has a higher JAWS because he had a superior peak, and thus, he’s in Cooperstown while Olerud is not. That said, despite all that JAWS has done to advance how voters measure Hall-worthiness, it still papers over what I believe to be a fundamental flaw in using WAR for this process: Replacement level is the wrong baseline.

In the context that WAR was developed to measure — a players’ value to his club — WAR makes perfect sense. You’re comparing to the hypothetical freely available replacement player who can be slotted in at virtually no marginal cost to the club’s resources. But when we talk about Hall of Fame worthiness… what is a “replacement level” player in that context? Do we declare players Hall-worthy because they are vastly superior to Quad-A cup-of-coffee guys? No, we declare them Hall-worthy because they are head and shoulders above the rest of the players we see every day. When I was sitting in the Camden Yards bleachers punching out All-Star ballots by hand with a pencil in the 1990s, I was seeing Ken Griffey Jr.’s name next to the other regular AL outfielders and comparing his play to theirs. (I was voting for Brady Anderson and B.J. Surhoff anyway, but I hope you’ll excuse some youthful homerism in this example.) We look at wRC+ and ERA-, stats that compare to major league average, and those play a significant role in determining who gets accolades and statues and numbers retired and plaques in upstate New York. I posit that for end-of-season awards and career retrospectives, Wins Above Average (WAA) is a superior metric for measuring how dominant a player performs relative to his league.

JAWS attempts to correct for this inherent problem with WAR in the context of total career evaluation by adding a peak component, and there are also versions of JAWS that adjust for changing pitcher workloads, measure relievers on a scale that’s closer to starting pitchers or position players, and include framing stats for catchers. But at the end of the day, all of those are essentially ad hoc adjustments to a metric that is still based on an underlying number that has limited applicability to this context.

Dan Szymborski wrote about using WAA in assessing Cooperstown candidacies earlier this year. He plugged in WAA to a JAWS-style calculator and called it altJAWS. I contend that a simpler adjustment (described below) is more appropriate for this purpose. Because WAA is already inherently measuring peak performance by only giving a player credit for being above average, adding an extra bonus for seasons that are among a player’s best is putting a hat on a hat. Raw unadjusted WAA is more flexible to the variety of shapes that an MLB playing career can take. A player whose peak output is concentrated into four or five years, as well as a player who maintains a high level of performance for a decade, is served imperfectly by the seven-year peak component of a JAWS-style system. A simple WAA accumulation treats each season individually and needs no adjustments to determine what counts as “peak” and what counts as “compiling”.

So, can we just calculate a player’s career WAA total and call it a day? Not quite. We don’t want to unduly penalize a player for hanging around long after his peak production has ended, even if that results in some seasons of play where he is below average. Hanging on for a few seasons in post-peak decline in order to hit some counting stat milestones that may be personally meaningful to a player, even as his overall rate of production dips below average, should not result in a ding to his WAA-based statistical case. Therefore, I make an adjustment to zero out any negative WAA seasons. You can never have banked value taken away from you, just like a fan’s memory of seeing their favorite player in his prime can never be dulled by the modern reminder of his lessened capacity to play the game. This approach is favored by no less a sabermetric thought leader than Tom Tango, who said on social media, “The worst thing we do with HOF evaluations is using WAR (Wins Above Replacement) instead of positive-WAA” and has blogged about a quick and dirty Hall of Fame estimator based on similar concepts.

In the long tradition of semi-tortured backronyms for sabermetric inventions, and with love and respect to Tango, I suggest we refer to this metric as BOOG Score (Better than Other Ordinary Guys) instead of positive-WAA.

Perhaps readers take issue with “major league average” as the baseline level for Hall of Fame analysis. Maybe you think All-Star performance is table stakes for Cooperstown hopefuls, and thus, there should be a higher level of production needed to boost a Hall of Fame résumé. Maybe you think there’s no ability like availability, and we need to give more credit to guys who were able to stay on the field for a long time even if they weren’t elite players. Choosing a different threshold as the baseline for this type of metric is absolutely a valid approach. The choice of zero point is somewhat arbitrary, within reasonable bounds; it simply reflects your opinion of “how good is good enough” for various baseball honorifics. As Tango muses in the blog post linked above, “Maybe you would use WAR, and zero-out below that level? Doing so helps pitchers with longevity (like Don Sutton). Maybe you would use a higher standard and zero-out that level? Doing so helps pitchers with shorter careers (like Koufax). There’s no right or wrong answer here.”

For the purpose of introducing this metric, I chose to use a baseline that is relatively easy to calculate from public FanGraphs data, which gave me a high degree of confidence that my implementation was internally consistent and properly calculated. Other baseball minds may come to different conclusions and suggest alternative “zero points.” BOOG score is a conversation starter, not a conversation ender.

For those interested in the nitty gritty details of calculating BOOG score, the next paragraph is for you. Others may feel free to skip down to the tables below.

Dan Szymborski’s article noted that FanGraphs does not provide an easily referenced WAA number on its leaderboards, so he approximated with a value of 2.0 WAR per 200 innings pitched or 650 plate appearances. However, there is publicly accessible FanGraphs data to allow a more accurate calculation with some intermediate steps. Batter replacement level is calculated at 20 runs per 600 PA, which can be subtracted from a player’s RAR (Runs Above Replacement) total and then converted to wins based on season-specific run environment data. For pitchers, I calculated the league total of WAR per IP for every league-season that’s publicly available at FanGraphs, then multiplied each pitcher-season’s innings total by that ratio to find how much a pitcher’s single-season WAR total should be debited to convert to WAA. By definition, this results in a league-wide total of 0 WAA per season as intended. I also reproduced Dan’s JAWS calculations with FanGraphs data, but unlike his results, I did not exclude 19th century pitching. This makes the difference in JAWS vs. BOOG Score for pitchers even more apparent. Like Dan, I chose to use fJAWS for this comparison instead of the more commonly cited “standard” version of JAWS (which uses Baseball Reference data), so that the differences between the lists are only due to the WAA vs. WAR methodology, sidestepping any underlying “disagreement” between Baseball Reference’s WAR model and FanGraphs’.

Below are the top 50 hitters and pitchers by both fJAWS and BOOG Score, and how they compare. Active players are evaluated through the end of the 2025 season.

Let’s look at an illustrative example from this list to demonstrate the impact of zeroing out negative WAA seasons: Albert Pujols. He’s listed as the 24th-best position player in MLB history by both BOOG score and fJAWS; if you buy what fJAWS is selling, then the BOOG score ranking must also feel sensible. However, if we instead counted up his career WAA total without excluding negative seasons, his career total would be 18% lower and he’d be ranked 42nd. On the other hand, if you chose to exclude negative WAR seasons from his fJAWS total, it would change by less than 2% — not enough to affect his placement on the list. It’s simply a lot harder for an MLB player to be significantly below replacement level than to be significantly below average. The zeroing-out step is relatively inconsequential for JAWS, but a critical part of the process for BOOG.

Taking a step back, two things stand out about the hitter and pitcher comparisons. First, the batting lists are much more similar to each other than the pitching lists are. This reflects the fact that pitcher usage patterns have changed much more dramatically across MLB history than batter usage patterns have, and BOOG is less influenced by differences in volume than JAWS. It’s a natural consequence of the fact that BOOG scores are, at first pass, all about peak – at least when it comes to individual-season totals. You can’t rack up a high career BOOG total without sticking around in the league for a long time, but you have to do that while continuing to be a notably above-average player. High quality major league batters don’t see as much variability in their playing time from year to year as pitchers do, especially when measured on the large scale of decades and centuries. The fact that JAWS and BOOG agree in broad strokes about the top position players in MLB history is a helpful sanity check that the underlying WAA methodology arrives at roughly the same conclusions as JAWS, at least when presented with player-seasons that have more consistency in terms of total time on the field.

Second, and relatedly, the JAWS pitcher list is much more dominated by players from the first 70 years of MLB history (1871-1940) than the BOOG list is. Again, this is due to the different choice of baseline in the underlying WAR vs. WAA numbers. By less heavily rewarding high-WAR seasons from old-school pitchers whose value to their team was more dependent on volume of innings thrown, BOOG scores consider the “peak vs. longevity” divide on an individual-season level. JAWS rates Christy Mathewson (~320 innings pitched per year) as the seventh-greatest pitcher of all time and Pedro Martínez (~200 innings pitched per year) as the 14th best; BOOG rates them as 17th and fifth, respectively. Koufax is not among the top 50 pitchers of all time according to JAWS; by BOOG score, he is. We will dive further into the implications of this “anti-era effect” as it relates to pitchers specifically in a later article.

Using WAA allows us to more naturally contextualize changing pitcher workloads across different eras of baseball, and creates a level playing field where performances from 100 years apart can be stacked up against each other with fewer adjustments needed to account for structural changes in the game. When evaluating batters, a WAA-based metric spits out rankings that do not drastically differ from JAWS, which serves as an informal proof of concept that the underlying methodology is more or less sound. These two facts combine to make BOOG scores a powerful method of analysis for providing context to discussions about the Hall of Fame.

Returning to the question from the beginning of this article: John Clarkson’s 1889 season was worth 4.7 WAA as a pitcher. Tarik Skubal’s 2025 season was worth 4.7 WAA as a pitcher. So as pitchers, they both added the same amount to their BOOG totals in those respective campaigns. I think that feels about right!

I’ve set up a simple web page where you can query for the BOOG score of any player in MLB history who reached at least 10.0 BOOG in his career (5.0 minimum for relievers), as well as the BOOG of all active players who are above 0. It displays their career accumulation on a year-by-year basis, and you can also compare their totals to the median Hall of Fame player (medians defined for the groups of all starting pitchers, all relievers, or all position players).

In later articles, I will dig deeper into Hall of Fame BOOG thresholds at specific positions, highlight specific cases of players considered by BOOG to be overlooked, and introduce a BOOG-based Hall of Fame induction estimator.