Mike Trout and The 1% Club

Like many other followers of FanGraphs, I never seem to get enough articles about Mike Trout — the general theme being: “another way Mike Trout is so great” — so I want to share one of my own. Contrary to popular belief, Trout did not invent WAR; even more shocking, the metric wasn’t developed for the primary purpose of showcasing how valuable he is. But that of course is what it does do. Many readers may know that beginning with his first full season, at age 20, and continuing up to his most recent, age 26, Kid Fish has had the most WAR through every season but one — his age-25 season, when he missed nearly fifty games with an injury — than any other player in history. So it shouldn’t be surprising that he is one of just 20 members of what I call The 1% Club: players who have equaled or exceeded 0.010 WAR per PA for their entire careers.

In fact, he’s currently second only to Babe Ruth:

This list is based mostly on FanGraphs’s fWAR, and is for players with a minimum of 50 career WAR. However, two players, Nap LaJoie and Jackie Robinson, don’t quite qualify by fWAR, but they do if we use Baseball Reference’s rWAR (all the others on the list qualify using either metric). Trout is the only active player on the list, though if we lower the bar to 30 career fWAR, three more players appear, including Mookie Betts, who is at 0.0104. Even lowering the criterion to just 10 fWAR adds just one more active player, Aaron Judge (0.0103).

No doubt as Trout ages and his performance declines, his value will drop somewhat. But it’s hard to imagine Trout dropping off the list entirely. None of the other players on the list saw more than a 15% drop in their WAR/PA from their age-26 season to the end of their careers. In fact, in addition to the 19 players in the table who averaged 0.010 WAR/PA or higher at age 26 (Wagner is the lone player to reach this for his career while not achieving it at age 26), there have been only eight other players who achieved this milestone at age 26 (minimum 30 fWAR) while falling below 0.010 by the end of their careers: Mel Ott, Arky Vaughan, Eddie Mathews, Dick Allen, Ken Griffey Jr., Frank Thomas, and Albert Pujols (still active). Only the final three of these suffered a major decrease in WAR/PA (30-35%) during the period from age 27 to the end of their careers. Griffey’s performance was compromised by injuries; Thomas, in addition to declining as a hitter, lost considerable value by moving to DH; and Pujols is the poster boy for age-associated decline in performance.

The 1% Club can be understood in another sense as well: year to year, roughly 1% of all players achieve a WAR/PA value of > 0.010. For example, in 2018 there were eight qualified players who finished the season at or above this level, out of 877 players on Opening Day rosters; in 2017, there were nine players out of 868 total at the beginning of the season.

The value of 0.0100 isn’t just another round number or pretty face, though; it has an interesting implication. If we assume that during a full season, a player gets roughly 700 PA, a WAR/PA of 0.0100 corresponds to about 7.0 WAR per 162 games. This value is often considered the threshold for a serious MVP candidacy. For example, the six AL players with a WAR/PA > 0.010 in 2018 finished in the top seven places in MVP voting that year, accounting for nearly 80% of the total points. The national league leader in WAR/PA won the MVP for that league. In 2017, four AL players exceeded 0.010 WAR/PA, and they took the four top spots in MVP voting. The NL leader won that league’s MVP.

So the players on the list above have basically averaged an MVP-caliber season throughout their entire careers. One can hardly be more awesome than that. All of them except Trout, who of course is active, and Joe Jackson, of the infamous Black Sox, are in the Hall of Fame. In fact, most of them, arguably, are inner-circle HOFers.

WAR/PA and the Hall of Fame

The WAR/PA metric — which converts what is basically a counting stat to a rate stat — may also be a helpful way of evaluating lesser players for their Hall of Fame prospects.

Most HOF voters base their decisions to some extent on both a player’s peak years as well as his entire career. Jay Jaffe’s JAWS system makes this approach explicit by determining the mean between the cumulative WAR of a player’s best seven seasons (WAR-7) and the WAR for his entire career. The idea is to give players some credit if they were exceptional for a sustained period of time, as well as to reward them for being very good over an entire career.

WAR/PA might be thought to favor players with a high peak, but this is the case only if they have a truncated career as a result of either retiring early and/or missing large chunks of time in their 30s due to injuries. For players with a career of more or less average length — which, as I will discuss below, is about 9000 PA for HOFers at most positions — WAR/PA should incorporate both peak and full-career contributions. That is, a high peak will contribute a high WAR/PA value for those years, while steady value over a long career will minimize the reduction in WAR/PA that usually occurs both pre-peak, when a player is still developing, and post-peak, when age-associated decline has set in.

With this in mind, I determined WAR/PA values for all 159 position players currently in the HOF, as listed in BBRef’s JAWS tables. This system, of course, uses rWAR, not fWAR, but the two metrics track each other fairly closely. The table below shows the mean WAR/PA value for each position, as well as the number of players at that position above 0.0100, 0.0070, and 0.0060.

Notice first that the mean WAR/PA values are fairly similar for each position, ranging from a low of 0.0060 for second basemen to 0.0072 for shortstops. What might surprise many readers is that one of the highest mean values is for catchers, 0.069. Based on WAR or JAWS, HOF catchers are a breed apart; they have the lowest average career WAR (53) as well as the lowest mean JAWS (44). In contrast, the mean career WAR values for HOFers at other positions ranges from about 65-71, and mean JAWS from 54-58. So by career WAR or by JAWS, HOF catchers are about 20-25% lower than HOFers at other positions. Catchers are also the only group with no HOFers reaching 0.010 WAR/PA, or 100 career WAR (Johnny Bench, the career leader at 75 fWAR, is not even close).

As I’ve discussed elsewhere, though, a major reason why catcher WAR is low relative to that of other positions is because players at this position tend to have shorter careers. The mean number of PA for the 15 catchers in the Hall of Fame is about 7500; for other positions, it ranges from about 9000-9500. The difference of 20-25% parallels the difference in career WAR or JAWS, so when one uses WAR/PA, the effect of this difference is effectively eliminated. Peak or Jaffe’s WAR-7 values for catchers are also about 20% lower than WAR-7 values at other positions; but as discussed in the link, even at the peak of their careers, age 27, catchers lag somewhat in PA relative to most other positions, though the difference is not as great as at the end of their careers. Also, it should be emphasized that WAR-7 is a total of the best seven years of a player’s career, and some of these seasons may occur in years beyond what is usually considered a player’s peak.

While catchers lack the extremely high WAR/PA values of a few HOFers at all other positions, they compensate for this by having fewer low WAR/PA values. Note that two-thirds of all HOF catchers have a WAR/PA of > 0.0070, the highest proportion for any group of position players. And 80% are above 0.0060, a value exceeded only by shortstops and third basemen. This distribution suggests that HOF voters in effect allow for the fact that catchers have shorter careers, while judging them just as strictly as other position players for what they produce in these careers.

So WAR/PA may be another useful way of evaluating a player’s HOF candidacy, a metric that is about the same for catchers as for players at any other position. From the data in the table, we can conclude that any player in the range of 0.0060 – 0.0070 or higher should be considered a viable candidate, assuming he has close to the average number of career PA for that position. Almost exactly half of all position players in the HOF have a value > 0.0070, and about 75% are > 0.0060.

Like 0.0100, the 0.0060 – 0.0070 range can also be understood in terms of a seasonal average: it corresponds to about 4-5 WAR per 162 games/700 PA. This is around the threshold for an All Star. For example, there were 30 qualified position players last year with a WAR/PA > 0.070, and 43 > 0.0060. While the correspondence is not exact, there were 43 selected for the All Star Game. With that in mind, we can say that a typical HOF resume corresponds to All Star production averaged for an entire career, just as an inner circle HOF career corresponds to MVP production on average for that career.

I began this article by identifying The 1% Club: players who have averaged 0.010 or greater WAR/PA for their careers. I noted that these players also comprise roughly 1% of all players in any given season. The Hall of Fame is another example of a 1% club, perhaps the ultimate one. There have been roughly 20,000 players in the history of major league baseball, and including pitchers, there are about 230 former players. If we exclude Negro League players — not because they aren’t deserving, but because the Negro Leagues don’t figure in that 20,000 total — we have almost exactly 1% of all players in the Hall of Fame.