Bill James For the Hall Of Fame
I am far from the first to write that Bill James should be in the Baseball Hall Of Fame. People have been saying this for more than a decade. For example, see Tom Van Riper, Forbes, October 31, 2013, “Bill James For the Hall Of Fame,” Graham Womack, The Sporting News, March 31, 2017, “What’s keeping Bill James out of the Baseball Hall Of Fame?” and James Malinowski, The New York Times, January 20, 2019, “Bill James Has the Required Numbers,” and most recently John Thorn, Medium, “Pioneers, Season Two: Bill James.” Many of the pieces I’ve seen making this argument have been short, relying on (accurate) summary assertions about how “his analytics revolution pervades baseball” or that he is “the Godfather of stats.”
In nearly a half century of published writing about baseball, James rarely addressed his subjects in shorthand or bullet-point fashion. And so it seemed to me that if one wants to make the case for why he should be enshrined, that calls for a somewhat longer exposition. What appears below is a revised version of a blog post I wrote earlier this year, after he had shut down his web site Bill James Online and after we learned that this year’s edition of The Bill James Handbook would be the last one. In a quixotic gesture, I sent a copy of this piece to the good people who run the Hall Of Fame for distribution to whomever it may concern.
I know not everyone shares my admiration for James or the work he has done. Even though the analytical approach to evaluating players has been widely incorporated by MLB front offices and many baseball writers, there are still former players and managers, as well as some fans and a few scribes, who view sabermetrics as anathema. This poses a problem for James’s HOF chances. As Graham Womack noted in his Sporting News piece, the “Eras” committees of the HOF are populated by a small group of “veteran baseball insiders, some of whom might actively resent what James has done for baseball.” Joe Morgan has been a prime example of this. In a 2005 interview with Tommy Craggs, he rejected what he characterized as computer-based analysis of player performance and value. “I have a better understanding about why things happen than the computer, because the computer only tells you what you put into it. I could make that computer say what I wanted it to say, if I put the right things in there.” Morgan thus implicitly recognized that, while a computer may perform calculations, it is some human being who has studied, thought about and. selected the factors being used to evaluate players. At bottom, Morgan was saying he knows better than a bunch of sabermetricians who never played the game the things that make a ballplayer valuable. As long as this attitude has influence in Hall of Fame decisions, there will be resistance to recognizing the contributions of Bill James and other sabermetricians.
James also made some enemies along the way with his sometimes unpopular opinions and disdain for those he viewed as fools. When the evidence summarized in the Dowd Report convinced almost everyone that Pete Rose had bet on MLB games, James wrote a detailed response challenging the evidence as insufficient to prove the case against Rose. Even some who otherwise were supporters of James viewed him as having stepped off the ledge on this issue. And he managed to irritate a good number of people in baseball—players and others—when he asserted in writing on two separate occasions (30 years apart) that if all the active players in MLB retired, the game would go on barely missing a beat. (“The players are NOT the game, any more than the beer vendors are.”) As Curt Schilling proved, stating unpopular views can be enough to keep you out of the Hall.
It is not even clear what the route would be for James to be get in. Under the revamped (2022) structure for selecting inductees to the Hall of Fame outside the writers’ balloting process for players, the Contemporary Baseball Era Non-Players Committee (1980 to Present) is charged with considering “retired managers, umpires and executives whose greatest contributions to the game were realized from the 1980-present era.” James worked in the front office of the Boston Red Sox for 17 years until he left the organization in October 2019. But, while he made contributions to the Red Sox in that role, his case does not really rest on being a ground-breaking executive. It is the collective weight of his ideas, insights, concepts and constructs—before, during and after his employment with the Red Sox—that merits a plaque on the wall in Cooperstown.
So James faces a tough road to induction. The post-1980 non-player committee will next consider candidates in 2026. Both the members of that committee and the nominators who will construct the ballot have yet to be determined, but both groups generally include Hall of Fame players and managers, major league executives and veteran media members/historians. One or two members of either committee who hate sabermetrics could easily scuttle his candidacy.
I have no expectation that what I have written is going to move anyone at the Hall who has a vote. Still, I am hopeful. One of my family’s Thanksgiving traditions is listening to Arlo Guthrie’s circa 1970 song “Alice’s Restaurant.” It’s a 20-minute-long shaggy dog story about petty crime and ending the war in Viet Nam. Near the conclusion, Guthrie talks/sings that if one person were to walk into his draft board and sing one bar of “Alice’s Restaurant” and walk out, “they may think he’s really sick.” And if three people do it, “they may think it’s an organization.” And if 50 people a day do it, “friends, they may think it’s a movement.” So my hope is that if enough people say it enough times, a Bill James movement may sway the Hall.
Bill James (Ret.)
February 12, 2024
I recently purchased the 2024 iteration of The Bill James Handbook, subtitled “Walk-Off Edition.” As the name suggests, this is the final edition of the Handbook, which for the last 20 years has been an annual companion for those interested in viewing the game and its players through an analytical lens. James has also recently shutdown Bill James Online, his website on which both he and many others published innumerable pieces, long and short, on a wide range of baseball topics that the authors felt could be better understood by taking a thoughtful look at the relevant data. These are sad developments. But even without uttering another word, James has made contributions to our understanding of and discussions about the game that are almost impossible to overstate.
Without diminishing the work of writers like Pete Palmer, John Thorn, Tom Tango, John Dewan, numerous writers at Baseball Prospectus and many others who have advanced the study of baseball analytics, I submit that two individuals have contributed the most to the statistics of baseball: Henry Chadwick and Bill James. In the late 19th Century, Chadwick fundamentally changed the way people talked and thought about baseball and its players. Among other things, he devised the box score to record what happened in a game, introduced core statistical concepts such as batting average and earned run average, and recognized that fielding range was a better indicator of defensive prowess than fewest errors made. The concepts and analytical approach pioneered by James, who coined the term "sabermetrics,” similarly changed the way we look at and talk about the game.
Well before The Bill James Handbook, James began publishing the Bill James Baseball Abstract in 1977. It was initially a set of photocopied stapled-together pages, which James estimated sold maybe 75 copies. By 1980, it had progressed to something that looked more like a book—with better binding and printing on both sides of the page. A couple years later the Abstract was published as a trade paperback by Ballantine Books, and it finally looked like a professional publication, something a reader in the general public could take seriously.
The earliest Abstract I still have in my possession is the 1983 version. In that one, James introduced his concept of how to calculate the number of runs a team (or a player) would create based on certain hitting statistics. He actually came up with three different formulas, but the one that survived more than a single season was the simplest one: Runs Created = (Hits + Walks) x Total Bases/(AB + Walks). It was a radical approach, quantifying how hitting statistics translated into runs scored.
In the 1984 Abstract, James returned to this topic in a seminal essay entitled “Logic and Methods in Baseball Analysis.” Writing like baseball’s version of Wittgenstein, James included in this article a series of fundamental baseball Axioms, along with three Known Principles of Sabermetrics. The first axiom sounds almost childishly simplistic: “A ballplayer’s purpose in playing baseball is to do those things which create wins for his team, while avoiding those things which create losses for his team.” It is similar in its simplicity to the so-called Fundamental Theorem of Poker, first articulated by David Sklansky in 1978: “Every time you play a hand differently from the way you would have played it if you could see all your opponents' cards, they gain; and every time you play your hand the same way you would have played it if you could see all their cards, they lose.” In both cases, the follow-on questions of How? and How much? are far more complicated, and more interesting.
What James was really about in stating things like the first axiom was laying the groundwork for the next step: identifying the things a player does that contribute to winning and quantifying how much they do so. Axiom II provides: “Wins result from runs scored. Losses result from runs allowed.” From this axiomatic point, James explained, his mission was a “to develop a formula which measures the number of runs that a player has created for his team—a formula that takes the number of hits and doubles and triples and home runs and whatever else and expresses them all as runs.”
Of three runs-created formulas he had introduced a year earlier, James stuck with the one set forth above because , he said, “it works.” By that he meant that it accurately predicts, through arithmetic performed on four broad categories of hitting statistics, the number of runs that will result. Focusing on the American and National League run totals for five seasons from 1979 through 1983, James showed that his formula predicted the total runs scored within 1% accuracy (except for the NL in 1980, when the formula was off by 1.3%). The small prediction discrepancies were neither consistently high nor low. And, James said, you don’t need to know how many stolen bases they had, what their batting average was with runners in scoring position, how many home runs they hit with men on base, how good a job they did moving runners over with an out, or how many errors their opponents made. “All of those things might make marginal changes in the number of runs resulting from the four basic offensive elements, but only marginal changes.” (James did in fact come up with a “technical formula” for calculating runs created that took into account stolen bases and several other statistics, but he viewed it as a refinement of the sound simple version.)
The third Principle of Sabermetrics addressed how runs created (and allowed) translate into wins: “There is a predictable relationship between the number of runs a team scores, the number they allow, and the number of games that they will win.” Again, it seems trivial—since teams win games by outscoring their opponents, teams that score more runs and give up fewer runs will, one would expect, win more games. The key thing in James’s principle is the phrase “a predictable relationship.” As he did with runs created, he came up with a formula to quantify that relationship, one that could be tested against the data, and predict how many games a team would win (specifically, its winning percentage) based on runs scored and allowed. He explained the formula as follows: “The ratio of a team’s wins and losses will be the ratio between the square of their runs scored and the square of their runs allowed.” He dubbed this the “Pythagorean expectation” of won/lost percentage.
Each year, the Abstract brought some new statistic or analytic concept to his readers, and some found their way into discourse about the game. In the 1988 edition, he introduced Game Score—a "garbage statistic,” he called it—that was an empirical way, based on a formula he created, to evaluate the dominance of a starting pitcher’s performance. The formula was based on innings pitched, strikeouts, walks, and hits and runs allowed. When he first wrote of it, he said he was aware of only one MLB pitching performance that had reached a Game Score of 100 in a nine-inning game—Sandy Koufax’s 14-strikeout perfect game against the Chicago Cubs. There had in fact been four others before James introduced the idea, and there were 11 more in the next 35 years. Kerry Wood, in his magical 1998 20-strikeout game, recorded the highest nine-inning Game Score in MLB history—105. For several years, Game Score appeared to be catching on. Baseball-reference.com added Game Score to its advanced pitching statistics, and the box scores published every day on MLB.com for years included the starter’s Game Score, before it disappeared a couple years ago.
Despite his early deprecation, James eventually warmed to the stat. Years later, in 2011, he came up with the idea of the “World’s No. 1 Starting Pitcher.“ Why, he asked, shouldn’t we treat these “center-ring figures” in every game the way we do top golfers or tennis players? The ranking, which he updated after every start on Bill James Online, was based on a weighted rolling average of a starter’s Game Scores (with downward adjustments made for starts missed due to injury). It was one of the fruits of Bill’s fertile mind that didn’t catch on, though I thought it was a great idea and would periodically check it on his website. Now, both the ranking and the underlying Game Score statistic seem obsolete given the ever-declining role of the starting pitcher as anything other than a five-inning employee.
James abandoned the Abstract in 1989, and later he began publishing the Handbook. The latter, he explained in this year’s essay “A History of the Handbook,” was created to provide a direct alternative to the Baseball Register, which for years had been published before the beginning of each season by The Sporting News. I was a regular purchaser of the Register. It included what seemed to me a wealth of pertinent baseball information. But to James, it was an embarrassingly inadequate rendering of baseball statistics. For example, the Register did not include saves by relief pitchers, strikeout rates, grounded into double plays numbers, on-base percentage, caught stealing, sacrifice bunts or flies, and it was woefully lacking in meaningful defensive statistics. The Handbook sought to remedy these deficiencies, plus do a lot more by including new statistics such as Runs Created. Though it was not his mission—James writes that he kept buying the Register every year until it ceased publication— his Handbook played a role in driving the Register out of business.
In 2001, James introduced the statistic Win Shares, in a book of the same name. After that, it was ubiquitous in his writing. The Win Shares stat reflected James’s attempt to capture in a single metric how much each player contributed to the games his team won over the course of a season—to create one “Great Statistic” that would measure a player’s value. It would allow its users to compare the value of all players, regardless of position, home park, or the era in which they played. As James later noted, people had been trying for at least 100 years to do that. “Most of these attempts were silly, naive, amateurish and kinda stupid,” he wrote. James acknowledged that he had long ridiculed the very idea of creating such a Great Statistic, but eventually he got the bug. “I thought I could do it,” he wrote. It became his white whale. Because he is Bill James, he went at the task full bore. Roughly 100 pages of the book Win Shares is devoted to explaining how the system he devised works.
James brought out Win Shares around the same time that other sabermetricians, also seeking to draw the connection between runs created (or prevented) and wins, came up with their own all-encompassing statistic, which they dubbed Wins Above Replacement, WAR. The idea was to measure how many wins a player added to his team’s total over what the team would have done with a high-quality minor leaguer or fringe major league player in his stead. Among the several formulas they devised, two have assumed prominence—one published in Baseball-reference.com (bWAR) and one by Fangraphs (fWAR). They produce similar numbers for position players, though they employ different methods and produce more divergent results for pitchers. As it turned out, WAR rather than Win Shares attained a prominent place in the firmament of baseball statistics. It is now included in virtually all comprehensive reports of a player’s seasonal or career performance. It plays an increasingly important role in evaluation of players for the annual MVP awards and for the Hall of Fame. And it has become a significant data point used in arbitrations and negotiations over player salaries. Perhaps unfairly, it appears that Win Shares is the Betamax of baseball’s Great Statistics—an arguably superior device that lost the battle for acceptance in the marketplace.
While James did not always welcome critics of his work, he was open to acknowledging when one of his constructs was shown to have shortcomings. But he was consistently and aggressively hostile to anything labeled analysis that he regarded as wrong-headed or, worse, lazy. That explains his long running antipathy to WAR, which he viewed as both wrong-headed and lazy, even as it steadily rose in importance among baseball statistics. Among other problems with this stat, he wrote, WAR “is dead wrong because the creators of that statistic have severed the connection between performance statistics and wins, thus undermining their analysis."
Thus, it is not surprising that in his “Walk-Off Edition” James could not resist including a parting shot at WAR. In his essay on Win Shares, which he says may be “my last chance in a couple years” to address the issue, he begins by acknowledging a series of mistakes he now recognizes he made in trying to come up with a formula that would “summarize all the varied accomplishments of a baseball player into one number, one statistic.” After explaining several ways he could have made Win Shares better, he compares his approach to that of the architects of WAR, who he says are not even attempting to get it right. “What WAR does [with its imperfections] is just say: ’Oh, who cares; we’ll just stick some simple little formula in there and call it good. Good enough for government work.’” And in a fit of frustration, he writes that “they’re right, because people DON’T care.”
If there is anything that can be labeled a “great debate” in the world of sabermetrics today, it is probably whether WAR (either leading version), Win Shares, or some other single statistic best expresses a player’s total value to his team. One of many concrete examples of this debate occurred in discussions of the voting for the 2017 American League MVP won by Jose Altuve. He got 27 first place votes to just two for runner up Aaron Judge. Several sabermetrically-minded writers argued, based on WAR, that Judge was at least as deserving as Altuve. (Fangraphs' WAR formula credited Judge with 8.2 wins vs. 7.5 for Altuve; bWAR had them virtually the same.) This was "nonsense," James wrote on his website, “Aaron Judge was nowhere near as valuable as Jose Altuve. . . . It is NOT close. The belief that it is close is fueled by bad statistical analysis.”
As between Win Shares and either of the leading versions of WAR, I don’t know who is right. Some baseball-focused statisticians have done regression analyses and concluded that WAR “works.” Dave Cameron, in an October 2009 article in Fangraphs, wrote that “WAR is the best tool we have” to accurately quantify the total value a player produces. For the 2009 season, Cameron determined there was a high correlation between the total wins reflected by a team’s players’ cumulative fWAR and the number of games the team won. Three years later, Glenn DuPaul published a piece in The Hardball Times reporting that he had calculated an even higher correlation between bWAR and teams’ total wins over a span of several seasons. DuPaul noted that one reason some sabermetricians were distancing themselves from WAR was the perceived unreliability of the defensive value component, which he described as having “a crazy amount of variability.” (On this point, James agrees. In a 2023 interview, he stated that the biggest unanswered question in baseball analytics is how to evaluate fielders’ contributions.) DuPaul also acknowledged that one season’s WAR had little predictive value for the next season. Despite these limitations, however, he concluded that single-season WAR “does an awfully impressive job of describing where individual wins come from for a team.”
Even with what he recognizes as some shortcomings, James remains adamant that his methodology gets closer to the truth of total player value than WAR. But his argument also underscores one problem with the debate: “Win Shares are more accurate, but to prove that they are you are going to have to unravel the pages and pages of acrobatic mathematical approximations.” Win Shares and WAR are, for the average fan, essentially black boxes. The formulas are complex and contain so many assumptions and components it is all but impossible for most followers of the game to determine which method better reflects a player’s contribution to his team’s wins. Whether or not James is right about the superiority of Win Shares, though, his persistence in the argument and his lack of complacency about the answer to the question are part of what has made him so valuable to the ongoing study of the game.
Over the span of more than 45 years, first in the Abstract and then the Handbook, as well as in numerous books like Win Shares, The Bill James Historical Abstract and Whatever Happened To the Hall Of Fame?, James asked and then sought to answer hundreds of questions about baseball. Some were the kinds of questions to which most any fan would want to know the answer (like who really is the most valuable player in a particular season or over the course of a career); some were of interest to a narrower group of fans or baseball professionals (like what is the game situation in which a strikeout is most costly to a player’s team); and others likely of interest only to James and a handful of like-minded obsessives. His flow of ideas was seemingly irrepressible. In his book Win Shares, after writing about its titular subject at length, James included in a separate section entitled “Random Essays” 40 pieces on topics ranging from “Aging Patterns Among Great Players” to “Biases In MVP Voting” to “Steve and Milt vs. the Dodger Dons.”
While I am a long-time fan of James, there were pieces that made my eyes glaze over in a sea of detailed statistical minutiae, and some of the questions he sought to answer were of limited interest to me. But I never stopped admiring the unflaggingly rigorous approach he took to every one of these questions. Even if he was not always right and his constructs did not inevitably gain wide acceptance, his analysis always contributed something to our evolving discussion of the game.
While James's "Great Statistic," Win Shares, has not gained wide acceptance, other Jamesian constructs have. The concept of Runs Created, for which James was the groundbreaking proponent, is now an essential element of sabermetric player evaluation. Widespread acceptance of James’s “Pythagorean expectation” for team wins is the reason that when you look at league standings on a site such as MLB.com, in addition to wins and losses, you find runs for and runs against each team and their “expected” W/L record. Other James statistics in use include Range Factor (a way to quantify a player's defensive value), Similarity Scores (measuring a player's closest statistical similarity to players from all eras), Secondary Average (based on the sum of extra bases gained on hits, walks, and stolen bases), and Major League Equivalency (a metric that uses minor league statistics to predict likely performance at the major league level). And in Baseball-reference.com, there is a section on each player’s Hall of Fame credentials that includes five different ways of assessing a player’s chances/worthiness. Three of them are creations of Bill James: Hall of Fame Standards, the Black Ink Test and the Hall of Fame Monitor. Each looks at the question from a different perspective—cumulative career achievements, discrete superlative seasonal performances, and likelihood of election (regardless of whether merited).
I think some people have viewed James as a kind of soulless number cruncher, who (with his like-minded sabermetric colleagues) has caused damage to baseball by exalting esoteric analytics over other criteria, some not quantifiable, that have traditionally defined excellence in the game. This is both wrong and unfair. I think it bears repeating James’s explanation of why he did what he did with statistics: “It was never my idea that we needed to look more carefully at baseball statistics because statistics are the best way to look at baseball. It was my point, rather, that people do make judgments about baseball players primarily by statistics, not should but do, and because they do they need a better understanding of what those statistics really mean.” James has never insisted that analytics—what he calls “organized understanding”—is the only way to look at players or the game. As he explained in an essay in the 2017 Handbook, “Many times people who are not in my business, not in sabermetrics, assume that we believe that organized understanding is superior to intuitive understanding. That is not actually what we believe, at all, or at least not what I believe.”
It is impossible to read James’s writing over the years without recognizing him as a fan who appreciates the non-quantifiable beauty of the game—all its rhythms, quirks, subtleties, unpredictable outcomes and athletic grace. Consider this from James’s preseason write-up on the Chicago Cubs in the 1988 Abstract, where he observes that about the only thing the Cubs have going for them in the coming season is their announcer, Harry Caray:
“Meaningless game in early July, Sutcliffe is throwing repeatedly to first trying to take the lead away from Milt Thompson. All of a sudden, early in the move, Harry screams ‘He got him.’ (I played it back on the video tape several times, and I am certain; the ball is still in Sutcliffe’s hand when Harry says the word ‘him.’) Sure enough, Thompson is a tenth of a second late at first base; the ump signals out. . . . Sure, Harry’s had a stroke, he mixes up a player’s name now and then—but what other announcer is so into the game that he will know the very instant when the pitcher has the runner by the throat.”
This example, one of numerous asides James intersperses with his various quantitative observations and dissertations, is remarkable in several respects. First, James is a Royals fan living in Kansas, but he is watching a “meaningless” Cubs game in July. Second, he is recording it. And finally, he is excited about just how cool it is that Caray made this call. He could not be further removed from being just a soulless number cruncher.
James was ahead of the curve on other issues beyond statistical analysis. In The New Bill James Historical Baseball Abstract, published in 2001, he addressed changes he would like to see in the game. Writing in the year after average MLB game time first crossed the three-hour mark, he made several suggestions to cut down the amount of “wasted time inside baseball games.” His first proposal: “adopt a rule limiting the number of times a pitcher can throw to a base.” His second: adopt a policy that “once the batter gets into the box to hit, time will not be called.” It took MLB two decades to see the wisdom in these suggestions, which fans largely agree have improved the game. (Proving that he has his own quixotic streak, James also argued that a good way to improve the pace of games would be to shorten the breaks between innings.)
Another change proposed by James reflected his concern, shared by many observers of the game today, that the proliferation of strikeouts and home runs, and the attendant decline of balls put in play, has damaged the quality of the game. His proposal: impose both a minimum weight and a minimum handle diameter on bats. These changes, he argued, would result in lower bat speed, which in turn would lead to fewer home runs, more singles and doubles, “and more contests between fielders and baserunners.” These were among the specific goals MLB sought to achieve, largely unsuccessfully, by a different rule change: banning defensive shifts. (In 2024, the numbers of singles, doubles and balls put in play per game were at or below their levels in 2019 and for decades before that. And batting average on balls in play was the same in 2024 as the last three seasons before the shift ban was implemented.)
James’ concern for the game he loves comes across in one of the essays in his final Handbook, with the unassuming title “Career Targets.” It begins with a discussion of how, despite the changes in the game from decade to decade, certain career milestones—3,000 hits, 300 wins, 500 home runs—remained meaningful. These career targets “are derived from the game’s history.” But recently, he laments, the game is changing too fast, so that the milestones that for generations were constant targets have lost (or are losing) their meaning. James insists he does not decry change in the game; he embraces it—“until there is too damn much of it. . . . The point at which it becomes a problem is when the things you loved about the game are carried away like an old refrigerator and dumped in an unregulated landfill somewhere. If a point is reached at which Mookie Betts can no longer be lined up coherently alongside Dwight Evans and Bobby Abreu and Paul Waner and Roberto Clemente, then baseball has lost something meaningful.”
Like so many fans, he has grown tired of a game that has reinvented itself as “a collection of big guys swinging wildly at anonymous pitchers throwing 100 miles an hour for 22 pitches each.” James recognizes he is being “histrionic” with this statement, but he is serious in his concern that this is one way the game is changing dramatically from its “historic norms.” He knows why, analytically, this evolution of the game is happening, but he still hates it and would like to see it change. After noting that MLB has deservedly congratulated itself for restoring a brisker pace to the game, the fact remains, he says, that “some people do not recognize the image of the game that captivated them some time in the past. It could be, perhaps, that there is still work to be done.”
Henry Chadwick was inducted into the Hall of Fame in 1938. His plaque in Cooperstown describes him as “Baseball’s preeminent pioneer writer for half a century.” Now that Bill James has seemingly retired (or at least taken a break) after nearly 50 years of pioneering baseball writing and contributions to how we think about the game, it is time for him to take his place next to Chadwick in the Hall.