What Is WAR in Baseball? The Complete Guide

A view from Wrigley Field from behind homeplate.

Throughout the history of baseball, the age-old debate has existed over whether Player A was better than Player B. Was Ted Williams better than Joe DiMaggio? Willie Mays or Hank Aaron? Ken Griffey Jr. or Barry Bonds? Or today, Mookie Betts or Mike Trout?

Since people first began playing baseball seriously over 150 years ago, statistics have existed to try to differentiate between who is a good player and who is not, but due to one reason or another, it can be hard to correctly determine that answer. This is where “WAR” comes into play.

Now you might be wondering, what is WAR in baseball?

WAR or “Wins Above Replacement” is a statistic that shows how many more wins a team has with a player, than they would have with a replacement-level player in his place. The statistic factors in a player’s offensive, defensive, pitching (if necessary) and base running to produce one number.

In the short lifespan of WAR, it has already sparked numerous debates about its accuracy, usefulness, and whether it truly does accomplish what it is meant to do. However, with more and more mainstream acceptance, WAR has truly emerged as a statistic that’s here to stay.

So let’s dive into the burning question:

What Is a Baseball Player’s WAR?

In baseball, the use of sabermetric statistics has been rising constantly over the past two decades, with new ones showing up just about every year.

All of these intend to evaluate a player in a manner that hasn’t been done before and/or is impossible to do using regular baseball card stats. WAR, however, is at the forefront.

WAR was created in the early 21^st century to try to paint a better picture of how good a certain player is compared to a so-called replacement-level player. The statistic uses one singular number to determine a player’s value in all facets of the game.

There are two key parts in the definition to look at. The first is the “replacement-level” phrase.

What this means is that the statistic essentially assigns an overall value to a run-of-the-mill minor league player and determines how much better an active Major League Baseball (MLB) player is than a theoretical minor league fill-in.

The other key part is that the statistic encapsulates all facets of a player’s game and combines his total contributions.

That way, a player who is excellent either offensively or defensively but struggles on the other side of the ball (or on the bases) would see his WAR suffer as a result.

On the other hand, the players with the best WAR are players who are excellent in most or all facets of the game.

Hence, players like Mike Trout and Mookie Betts who are excellent in hitting, fielding, and baserunning are among those with the highest WAR in MLB.

Also, it should be noted that because pitchers rarely hit anymore and are a minimal factor defensively, their version of WAR is calculated entirely differently than that of position players.

How exactly do those formulas work? We’re going to tackle that question next.

How Do You Calculate WAR in Baseball? (Position Players)

First baseman in grey and green prepares to catch the throw to get the runner out.

Unlike other statistics we’ve dived into such as ERA, OPS, WHIP, and others, WAR is different in that there is no cut-and-dry formula to determine a player’s WAR. There are many factors and moving parts that go into the calculation.

According to MLB, WAR is “(The number of runs above average a player is worth in his batting, baserunning and fielding + adjustment for position + adjustment for league + the number of runs provided by a replacement-level player) / runs per win.”

Did you get all of that? Well, if you answered no, you’re certainly not alone. To put it lightly, the calculation is not simple at all. There is not even a commonly accepted formula for WAR.

Two forefront statistic websites that compute WAR: Baseball-Reference and FanGraphs, have slightly different formulas.

Both sites have posted how they calculate WAR. Fangraphs has a somewhat lengthy explanation which helps, but still doesn’t explain everything. Baseball-Reference has a much more in-depth explanation of their WAR calculation, which covers over 6,600 words, so we obviously won’t dive too deep.

Baseball-Reference also has an in-depth chart comparing components of both sites, as well as Baseball Prospectus’ Wins Above Replacement Player (WARP).

Many components are similar, but there is enough variation in others to cause WAR outputs to vary by source.

However, for the sake of this exercise, because Baseball-Reference’s WAR (abbreviated as bWAR) is used more often (notably by ESPN) than FanGraphs’ WAR (fWAR), we will tackle the components that go into bWAR specifically.

As mentioned, because the Baseball-Reference formula is thousands of words long, we are just going to scratch the surface and briefly look at their six primary components: batting runs, baserunning runs, runs added/lost due to double plays, fielding runs, positional adjustment runs, and replacement level runs, as well as what they mean.

It should be noted that all of these components have other complicated statistical terms and formulas within them, so for the sake of brevity and relative clarity, we are going to omit the exact terms and equations and use layman’s terms to illustrate more or less what goes into them.

Batting Runs (Rbat) measure a player’s effectiveness while hitting. This number is adjusted to the league each season depending on if offense is up or down. It also removes pitchers’ batting stats from the equation. It also differentiates strikeouts and batted outs, infield and outfield singles, and factors in when a player reached on errors.

Baserunning Runs (Rbr) factors in not just stolen bases and caught stealing but also how adept baserunners are at gaining extra bases (i.e. first-to-third on a single, scoring from first on a double, etc.) while also factoring in how often they are thrown out on the bases, while weighing the occurrence of these events against how many occurrences a player has at a certain base, as well as the league’s numbers as a whole.

Grounded in Double Play Runs (Rdp) is the measure of whether a player is good at avoiding double plays or not. This factor is determined by how many double plays a player hits into whenever he hits a groundball where at least one out is recorded whenever there is a runner on first with less than two outs.

Fielding Runs (Rdef) factors in a player’s defense. This may be the most complicated, as it is largely based on eight mostly-invisible and somewhat subjective factors that go into the overall number. The overall goal is to essentially determine how many runs a player saves or costs his team due to his defense.

Positional Adjustment Runs (Rpos) is an adjustment that adds or subtracts total runs solely based on what position a player plays (except pitcher). More difficult positions have positive numbers as multipliers, while easier spots have negative numbers. These are pro-rated based on what percentage of 1,350 innings (150 games) were played at a certain spot, which is calculated separately for players who played multiple positions.

Replacement Level Runs is the final component, which is the overall basis for the rest of the calculation. However, because it applies similarly to both hitters and pitchers, we’ll come back to it in a little bit.

In the meantime, let’s look at how things go for the men on the mound.

How Do You Calculate WAR for Pitchers?

MLB pitcher throws a pitch at Coors Field.

While pitching WAR is still not a cut-and-dry statistic, the formula for the pitching side is a little less cluttered, because a pitcher’s WAR only factors in two primary inputs: pitching and fielding.

In a nutshell, pitching WAR is calculated using a pitcher’s runs allowed and innings pitched, while adjusting the numbers to more accurately determine the effectiveness of a pitcher against the other teams in his league and adjusting for team defense and the pitcher’s role.

So as stated a couple of paragraphs ago, it boils down to pitching and fielding, but like position player WAR, pitcher WAR has several smaller sub-categories to factor in.

The main goal, as Baseball-Reference explains, is to determine how an average pitcher would perform in the same circumstances. Once again, we will use Baseball-Reference’s WAR (bWAR) for this exercise.

The number that results is the expected runs allowed (xRA), which separates players by their league (American or National League), removes all team stats associated with their team, removes interleague games, adjusts for park factors (whether a park favors pitchers or hitters), and also factors in how good or bad a pitcher’s defense is behind him.

There is also an index based on whether a pitcher is a starter or a reliever. Because starters pitch more innings and tend to have higher ERAs than relievers, there is a slight adjustment made to give starters more of a benefit of the doubt.

To counteract this, there is also a leverage index assigned to each situation (i.e. how big of an impact does a certain outcome in a certain situation have on a game) that accounts for a large portion of the WAR for late-inning relief pitchers.

All of this is combined, weighed, and compared with the pitcher’s actual performance and what a replacement-level pitcher would be expected to do. Now, it’s time to figure out exactly what replacement-level means.

What Is a Replacement Level Player?

As we said earlier, the logic behind replacement-level is what the baseline of expected production would be if a player were to theoretically be replaced by a standard minor league player.

Replacement-level is determined by Baseball-Reference to be a .294 winning percentage, as in that is what a team of explicitly replacement-level players would be expected to accomplish (in other words, the rough equivalent of going 48-114 over a full season).

Replacement level rarely changes, but it will be slightly different based on shortened seasons (in the case of strike-shortened seasons or the 2020 COVID season) or if the league talent pool is depleted, which was the case in World War II.

The .294 replacement-level figure is a key figure used in WAR calculation. The reason why is because when you subtract it from .500 (the winning percentage of a truly average team), then multiply the result by the 4,860 team games in a normal MLB season, you get a total of 1,000 WAR across MLB in a given season.

Why is 1,000 WAR relevant? Well, we’ll explain.

How Is WAR Calculated?

Because WAR is a statistic that is based on a complex calculation, it is not easy to track in real-time. Also, WAR is unique in that it is a statistic that is allocated on a player-by-player basis, not compiled like other stats. What does this mean?

WAR is unique in that it is not created or destroyed, meaning that as one player’s WAR goes up, another’s must go down so that the league total equals 1,000 WAR for a season. Therefore, a player can only earn WAR for continually being a good (or better) player.

This all factors into the pool of 1,000 WAR, which only changes if a season is shortened or if fewer teams are playing. The shorter the season, the less amount of WAR there is to go around.

Additionally, the WAR distribution in nearly all seasons has been different between the American and National League. Why so?

Well, because the AL had the designated hitter for several decades and the NL didn’t, the WAR formula is designed for offensive production by pitchers to be zeroed out, so that any offensive production on their part is a net positive, rather than holding them to the same standard as position players.

In other words, WAR accounted for one more offensive contributor on American League teams than for National League teams, with the same amount of pitching contributions.

WAR is also unique in that players can earn negative WAR. If a player is generally a poor player all-around, he can finish in the red on WAR, making it one of the few stats where a player can finish with a negative number.

Is WAR Useful in Baseball?

Baseball player gets in the ready position on the left side of the infield.

Many people are still on the fence about how useful WAR is, in large part because how it is calculated is so ambiguous and hard to comprehend for many.

However, WAR can be used as a helpful tool to determine how players match up against both contemporary and historical peers, especially in situations where run-scoring environments are drastically different and traditional stats would be misleading.

For example, we took a look at every qualified pitcher (one inning pitched per team game) who compiled an ERA of exactly 3.00 since 1920. We found two examples, Chuck Dobson of the 1968 Oakland A’s, and Odalis Perez of the 2004 Los Angeles Dodgers.

These pitchers had virtually identical strikeout rates, and Dobson threw just three more innings than Perez.

However, 1968 was the most pitching-dominated season since World War II, so Dobson’s accomplishments only earned him a 0.9 WAR.

Meanwhile, the 2004 MLB season was still in the later years of the high-scoring “Steroid Era”, so Perez’s work is much more impressive in context, and indeed, he put up a very solid 4.7 WAR.

What Is a Good WAR in Baseball?

Because WAR is allocated on more-or-less a per-game basis, season length will greatly affect WAR. However, in standard seasons, there is a rather clear benchmark on what constitutes a good, bad, and excellent WAR.

As a rule of thumb, the average WAR for an everyday position player or an average starting pitcher is 2.0. A WAR above 3.0 is good, above 4.0 is very good, and above 5.0 is excellent. Once a player’s WAR surpasses 6.0, they are a legitimate MVP or Cy Young Award candidate.

As FanGraphs once studied, in an average season, the vast majority of players in any given season (many of whom played sparingly), record a WAR between -1.0 and 1.0, symbolizing that the majority of the league is somewhat close to, if not below, replacement level.

On the other end of the spectrum, an extremely high WAR can be attained, with figures over 10.0 being exceptionally rare, and anything north of that being virtually unheard of.

With this information, the next time you see a player’s WAR, you will now have a better idea of whether he’s a star or likely destined to head back to the minors.

Odds and Ends About War

The highest single-season WAR in modern MLB history (since 1901) belongs to pitcher Walter Johnson, who posted a 15.1 WAR season for the 1913 Washington Senators. He had the second-best season by a pitcher the year before, putting up a 13.2 WAR in 1912. The highest mark by an offensive player is the 14.1 WAR posted by Babe Ruth in 1923. Ruth holds five of the top ten offensive WAR marks in MLB history.
Ruth also holds the record for the highest career WAR, with 182.5. Johnson leads all pitchers with a 164.5 WAR. Among active players, Albert Pujols leads position players with 100.6 WAR, while Justin Verlander is tops among pitchers with a WAR of 71.8.
By Baseball-Reference standards, the average member of the Baseball Hall of Fame has a 69.0 WAR. Remarkably, this figure is virtually identical for both pitchers and hitters.
In MLB history, 32 players have a career WAR over 100.0. In the modern era, there have been a total of 57 offensive seasons with at least one position player earning a WAR of 10 or higher and 52 pitching seasons with at least one pitcher earning a WAR of 10 or higher, an average of almost one 10-WAR season every year. However, just six have occurred since 2005.