EmailPrintOpen Extended ReactionsEarlier this month, Roger Clemens visited the Milwaukee Brewers during the team's recent trip to Houston, and among the awestruck Brewers pitchers was veteran Brandon Woodruff, who relished the chance to talk pitching with the all-time great."I'm a big fan of Roger Clemens, and I watch a lot of his stuff on YouTube," Woodruff told ESPN. "I like to go back and watch highlights. I just love his demeanor."After the meeting, Woodruff spoke to MLB.com about the pitcher's appearance. In describing Clemens, Woodruff had a number of superlatives to choose from. He could have mentioned Clemens' membership in the 300-win club or the 4,000-strikeout club, or simply alluded to the Rocket's record seven Cy Young Awards.Instead, Woodruff cited another stat: "I mean, that's the eighth-highest WAR in the history of baseball right there, hitters and pitchers," he said then. "You're kind of in awe. This dude is the epitome of starting pitching."Yes, among all the historic feats and accolades that surround Clemens' on-field legacy, Woodruff went straight to WAR. Clemens' historical ranking on the Baseball Reference leaderboard slots him right behind Hank Aaron and just ahead of Tris Speaker.WAR -- or wins above replacement -- is good for a lot of things, it turns out, even articulating one pitcher's starstruck reaction to meeting a legendary peer. Still, years after the framework worked its way into baseball's mainstream, some think it could be better, and others have resisted it altogether. Multiple public versions of it exist -- most prominently Baseball Reference WAR (bWAR) and FanGraphs WAR (fWAR) -- and individual MLB teams have their own proprietary figures, too.Even Woodruff isn't necessarily a WAR aficionado when it comes to his own career."Probably in that middle ground there," Woodruff said when asked by ESPN about his WAR buy-in. "I know different sites calculate WAR differently. Some sites just value either pitch metrics or batted ball stuff differently, so I don't pay attention."As WAR continues to evolve, it's worth addressing what has made it so influential and so confounding -- and what it might look like in the future. Imperfect as WAR might be, it has become, without question, the most important statistic in baseball.How WAR went mainstreamThere's no one moment when WAR became the game's go-to metric -- it has happened slowly and persistently. FanGraphs launched its version of WAR in 2008; Baseball Reference in 2010. But if there was a "WAR moment," when its existence, uses and, yes, shortcomings, exploded into the baseball mainstream, it was probably in 2012.That's when a ridiculously talented rookie, Los Angeles Angeles outfielder Mike Trout, trounced all competitors with an MLB-leading 10.1 bWAR, more than two wins better than every other player. Yet as the season advanced and Detroit Tigers slugger Miguel Cabrera was in hot pursuit of baseball's first hitting Triple Crown since 1967, arguments raged from every corner.It was the ultimate old school vs. new school debate, and it was a hot one. In the end, Cabrera won both the Triple Crown and the MVP balloting, landing 22 of 28 first-place votes. There are still arguments about that race, though the person most affected understands Cabrera's case."The guy hit for a Triple Crown, right?" Trout said, still marveling over Cabrera's feat. "That's pretty incredible."Neither player professed to understand WAR back then and for Trout, not a whole lot has changed, though WAR flagged him as a surefire Hall of Famer even before he'd reached the required 10 MLB seasons for eligibility."I'm definitely aware of it," Trout said. "Couldn't really tell you how they calculate it. But it seems like the last 10 to 15 years, when they are looking at players, they're going off that."While the Trout-Cabrera MVP debate didn't result in either universal acceptance or rejection of the WAR framework, it certainly heightened awareness of it.Perhaps the ultimate arrival into the mainstream occurred the very next year, in 2013, when Topps first began to list WAR on the back of some of its cards, a practice that has since become standard.Why the two leading versions of WAR are so different -- and how to make sense of itFor many, confusion about WAR starts with the simple fact that the two leading purveyors of the measure -- Baseball Reference and FanGraphs -- often come to different conclusions. The differences are small, except when they're not. A few examples from the current season and beyond: Shohei Ohtani's two-way act for the Los Angeles Dodgers marks him as one of the most valuable players in baseball. This year, with 5.5 combined WAR at Baseball Reference, he's first among all MLB players -- but has only recently taken a 0.3 WAR lead on Philadelphia Phillies starting pitcher Cristopher Sanchez. At FanGraphs (5.8 combined WAR) Ohtani has been ahead longer -- and is now nearly a full win ahead of second-place Pete Crow-Armstrong of the Chicago Cubs (4.9). Sanchez is second at Baseball Reference with 5.2 WAR, but he's fourth at FanGraphs (3.9), a yawning gap of 1.3 wins. Houston Astros designated hitter Yordan Alvarez ranks sixth at FanGraphs (3.6). At Baseball Reference, his 3.6 bWAR is the same total, but ranks just 16th. On the career front, Hall of Fame right-hander Walter Johnson's 168.4 bWAR is second all time behind Babe Ruth. At FanGraphs, even after you add in his 9.5 career batting fWAR, he ranks just 13th all time at 125.9.This is cherry-picking, but you get the idea. The two leading WAR systems often get considerably different answers to the same questions.The differences are a product of different inputs, the decisions made on which ones to use and about how to use them. They are detailed here, and they are numerous. Researcher, analyst and former Toronto Blue Jays consultant Sean Smith -- known in internet circles as Rally Monkey -- created a version of the WAR framework that was adopted by Baseball Reference, and he sees the biggest contrasts in how the different systems divide credit and blame between pitching and fielding."Almost every time you see a difference between players, it comes down to the defensive number," said Smith, whose book "WAR in PIECES" explains the past, present and possible future of every facet of the framework. He maintains a reworked version of WAR and other win-based metrics at BaseballProjection.com.Baseball Reference and FanGraphs see the contrast as a feature, not a bug, and they might be right."It's definitely a feature," said Tom Tango, database architect of stats for MLB Advanced Media, whose seminal work on calculating WAR helped build the foundation of the current, public-facing versions of the system. "I'm not there to tell them what they should think. I'm simply saying, well, if you care about this, then this is how you should construct the WAR, and if you care about the other thing, that's how you should construct the WAR. FanGraphs and Baseball Reference have taken two polar opposite views, which to me is fantastic. Because then I know that the true answer is somewhere between the two."In terms of explaining WAR to a casual fan, it's a disconnect not so easy to dismiss, especially because, again, the framework is the same, as is the label on the statistic itself -- they are both WAR."It's just a way for people to assign value, which is all this game is about now, because if you can't assign a value, then you don't really know how to play something nowadays," numbers-crunching Kansas City Royals first baseman Vinnie Pasquantino said. "So that's all it is. It's a tool for the public to be able to rank players."In the early days of WAR's availability on FanGraphs and Baseball Reference, the differences were even more profound than they are now. That's because of the "replacement" aspect of WAR -- wins above replacement level. Back then, the sites used very different baselines for what constituted replacement level, or the standard of performance of "freely available players" -- the type you can summon from Triple-A whenever needed.Then, in 2013, the leaders of the two sites had a kind of stathead detente and agreed to operate off a similar measure of replacement level. Thus there is a precedent for the systems to move toward each other. But one obstacle to that happening again is that in some ways, the aims of the warring versions of WAR are not the same.In a nutshell, FanGraphs WAR functions as a true talent estimate that works well for projecting forward and making economic valuations. The Baseball Reference version is more a matter of divvying up credit -- describing what actually happened. That's an oversimplification but gets at something essential."FanGraphs is based on process," Seattle Mariners president Jerry Dipoto said. "It's what you do. Baseball Reference, on the other hand, is based on results."Among those echoing that observation is Sean Forman, president of Sports Reference L.L.C., which operates Baseball Reference."I don't know if [FanGraphs] would agree, but for me the main difference is that B-R's version tries to explain what happened on the field," Forman wrote in an email. "While FanGraphs focuses more on the players' skill level."This crucial distinction is too often overlooked by anyone who looks at the competing WAR values on the two sites and thinks of them as aiming for the same goal. But really, they aren't.FanGraphs' forward-looking emphasis is a natural for anyone looking to make decisions based on projected outcomes, a large subgroup of the baseball world that includes the scores of fans engaged in fantasy baseball."A win in FanGraphs WAR is very similar to a win in Mariners WAR, and I'm guessing that is very similar to a win in White Sox WAR," Dipoto said. "We're all operating off of the same sort of system. But through the years, you just see where it's a little bit off."By filtering out as many factors that are out of the player's direct control as possible, FanGraphs expresses a more pure estimate of the player's underlying talent level. The inputs in fWAR are more stable from a projection standpoint. That underscores its use as an estimate of player value.And if you're focused on estimating player value, then the concept of replacement level is inescapable. It renders fWAR as much an economic indicator as a baseball statistic, something that is of great use when trying to think like a baseball executive or analyze transactions."We have our own formula for WAR," Royals president J.J. Picollo said. "I personally use it more in the offseason, when you're trying to build your roster. When you're looking at how we compare to other clubs. Where are we weakest? Where do we need to make the greatest gain?"But this is not to say that FanGraphs' version of WAR is better than the Baseball Reference version, nor that of any particular team. They are all just different, and you can argue that Baseball Reference does the best job of measuring what has already happened on the field by not trying to contextualize too many aspects of a player's performance record into oblivion.Such an approach is useful when trying to suss out things such as candidacy for awards, the Hall of Fame or when doing any kind of history-based research."I encourage people to look at the underlying numbers to see how we arrived at the final value," Forman said. "We show all of our work. I like that WAR can surface performances that were much better or much worse than what the people thought at the time."The chief differences between the FanGraphs and Baseball Reference versions of WAR come on the run prevention side. Baseball Reference bases its pitcher WAR on runs allowed, adjusted for team defense, rather than the more theoretical FIP (fielding independent pitching), which FanGraphs relies on for its version of pitcher WAR. This is a profound difference in philosophy."There's so many different metrics to measure pitcher value," Chicago Cubs starter Matthew Boyd said. "There's SIERRA, there's FIP, there's ERA, there's WAR. I think that the real question is whether there's one that can wrap up a pitcher's value in total. It's just all just an algorithm, right? What's the skew in it? What do you want to value more?"What makes WAR so appealing to teams -- and even many playersIf we back away for a moment from the current WAR systems and think more generally, it's perhaps easier to see the appeal. You have one number that evaluates the performance of every kind of player and allows you to compare them from team to team, league to league, stadium to stadium and even from era to era. And it's expressed in the ultimate form of baseball currency -- wins.Such an expression is a perfect staging ground for any kind of baseball debate or investigation. Who is the MVP? How do we compare different kinds of players? How do we decide who is worthy of Hall of Fame induction? It's a vehicle for understanding and context, even for practitioners of baseball at its highest level."Trying to compare players from different positions is tough," New York Yankees' three-time MVP Aaron Judge said. "I don't know if you can really, truly put a number on what type of value guys bring, comparing left field to first base to shortstop to center to catchers to pitchers. I think it's tough, but I think it kind of gives a general idea of what type of value that people bring to the game or the team."While debates over the 2024 AL MVP race didn't reach Trout-Cabrera levels, that competition between Judge and Royals shortstop Bobby Witt Jr. was a good example of WAR doing what it's supposed to do. It created a common measurement between two very different sorts of players -- the classic power hitter against the hyper-athletic jack-of-all-trades. Judge pulled away in the WAR race near the end of the season and ended up becoming a unanimous MVP selection."It's just tough for players to kind of get on board with something that judges you, and you know that it's not 100% an objective thing, right?" Pasquantino said. "When it works well, it can take a player like Bobby with the full array of skills, and then a player like Aaron Judge, and like in the MVP race two years ago, it can do a pretty cool comparison, even though they are creating value in very different ways. That's kind of what it needs to get to."Witt, who says he defers to Pasquantino on all metrics-related matters, has certainly been exposed to WAR often enough. After all, it has been on the back of baseball cards since Witt was 13 -- and he's an avid collector."I really don't know what goes into the numbers and stuff, because I can't really control any of that," Witt said. "So I don't want to look into that and try to be like, 'Oh, if I do this, I'll get my whatever WAR.' Obviously you want to have a high WAR as a player because then you're helping the team win."If you have any doubt at all about WAR's efficacy, take the word of those who run teams, build rosters and rely on accurate valuations of baseball talent in order to keep their jobs -- baseball's lead executives."I think we're at a stage in baseball history where every club has some type of number," Dipoto said. "When I first started in scouting and player development, it was a grade on every player based on whatever your grading system was. It could be an A-B-C. It could be a 40, a 60 or an 80. I think 30 teams now have some system by which they value a player, and it's usually some form of WAR."Here it's useful to think of WAR as more of a framework than a statistic or metric -- a system of inputs combined in such a way to yield a calculation that professes to be a measure of the value of baseball production. The public sites that calculate WAR have different inputs. And so, too, do baseball's 30 teams.This is why there is no clear consensus on which version of WAR is "right" or even whether any of them are. Teams use the framework, or something very similar to it, but beyond that we can't know how they adapt it to their specific needs and cultures because that information is key to what teams hope will become a competitive advantage."All we're trying to do is just value players appropriately," Milwaukee Brewers president Matt Arnold said. "Everyone's doing that, and then it's just sort of how you do it. That involves a ton of ingredients, right?"While teams are never going to be transparent about their internal processes, the thing about the public versions of WAR is that they are attempting to do many of the same things as baseball's executives. So if we're really following baseball, or a particular team, closely, it's useful to organize our thinking through the WAR prism. It brings us closer to those making the decisions for the teams we love.While not every player is on board with WAR, many track it. For someone like Judge, who said his favorite stat while growing up was batting average, it is a kick to see how the greats of the past are measured and where he might fit into that historical context."It's always cool for me to see Barry Bonds' WAR, Babe Ruth's WAR," Judge said. "Mike Trout, even 10 years into the game, he already had like an 80 WAR, and there's some Hall of Famers that have 80 WAR. I really don't know all the numbers that go into it to calculate it, but it's just always cool for me looking at the Hall of Famers' numbers."While all baseball numbers have their own kind of beauty, there is always room for more if they enhance our understanding of the sport. The "if" in that sentence is important, however.A perfect version of the WAR system would be baseball's ultimate statistic. The problem, as so many like to point out, is the framework/system/statistic/measurement remains not only inscrutable to many players and fans, it also remains a work in progress. But the work, for countless reasons, is worth doing."I think it's cool just to see, like, maybe you see a guy not hitting well but his defense is unbelievable, so that shows how he's helping the team," Witt said. "Because essentially, WAR is like, 'What can you do to help the team?'"What the future holds for WARLove it or leave it, WAR is not going anywhere. It's not just deeply embedded in the coverage of the game and in the front offices of teams, it's literally codified into the CBA.That version of WAR, used as a mechanism for distributing money from a pre-arbitration bonus pool, is some mysterious concoction agreed to by MLB and the MLB Players Association. But if you think the players themselves aren't paying attention, you're wrong. This is especially true of young relievers who have become acutely aware that it's not that easy for them to stand out in the WAR framework. That reality is costing them money."Relievers are screwed," Pasquantino said. "I mean, they're screwed in general, but especially with that stat."If WAR is a permanent feature of the baseball landscape, it must continue to evolve, especially in an era when the proliferation of Statcast and related technologies have led to an explosion of hyper-granular measurements that allow us to see things in the game in a way no one could have dreamed of before.When it comes to the expression of win-based metrics, it's important to note that there are others beyond those using replacement level, as WAR does. Baseball Reference produces bWAR but also makes use of wins above average, though it doesn't get as much play in the coverage of the game. WAR is the go-to for most, but if you want to view the leaderboards through a WAA perspective, that's easy to do, too.And, one could argue, when it comes to things such as choosing award winners, All-Star teams or Hall of Fame inductees, something like average or absolute wins should be the measure anyway. Do we really want to be using replacement value when weighing the Hall of Fame candidacy of a player?"I think that there are arguments for using WAA in the context of HOF cases," Forman said. "As you want to identify players with peak value rather than just accumulators who were average for a very long time."It all comes down to baseball's essential currency -- wins, and the way those are measured is always subject to change."We are due to reassess and upgrade a couple of areas," Forman said. "Park factors, confirming our defensive numbers, where replacement level is now and the relative value of the various defensive positions. We are more focused on the past, so I'd like to see us continue to fill out our historical data, especially the Negro Leagues."As long as they work, win-based measures will continue to proliferate, though they are argument starters, not enders. There remain multiple ways to view a player's value -- even among those whose livelihoods are judged using WAR."RBIs ... that's usually the number one thing for me," Judge said. "It's RBIs and runs. If I'm scoring a lot of runs by touching home plate, then I think we're going to be in a good position.""OPS, on-base [percentage]," Trout said. "If your OPS is in the .900s, it means you're doing something. And getting on base is huge."If the aim is to create a "real" version of WAR -- a definitive one that achieves something close to universal acceptance -- MLB Advanced Media, fueled by the precision of its Statcast measurements, would be a likely source. The thing is, that's not necessarily at the top of the group's to-do list. There are too many areas of the game to dive into -- an endless array of questions that heretofore have never been accurately answered."Nothing technically really stands in the way," MLB.com writer and analyst Mike Petriello said. "I feel like a lot of people [thought] from the first day of Statcast that the one great number to rule them all was the goal. Really, the main goal is quantifying all the cool stuff along the way."Bill James wrote that baseball statistics have taken on the power of language. You could start chatting with a total stranger about the game and refer to "a .300 hitter" and no further explanation would be necessary. The statistic had taken on such specific meaning that it functioned like common language.Baseball WAR, or whatever shapes it takes in the future, is nowhere near that, at least not in the mainstream. It's there in baseball's inner circles, where you will hear something like, "we had him as a five-win player," and maybe the rest of baseball lovers can get there, too.The purpose of baseball's statistical archive isn't just to create lists, settle arguments and generate fodder for advanced analysis and player valuation. It's also to share memories and stories, to connect the present day with baseball's greatest asset -- its long, glorious history, both on the field and in the record books.If WAR helps to do that, its existence is redeemed.
