Wednesday, 17 July 2013


The amazing website Baseball Reference gives a handy guideline for understanding WAR:
<0   Replacement
0-2   Substitute
2+   Starter
5+   All Star
8+   MVP

That seems pretty straightforward. Produce 8 or more WAR in a single season, and you are probably in the MVP conversation*.

*Provided you can get through the other BWAA hoops that have been put in place over the years.

WAR is composed of, well, a little bit of everything. It is an attempt to quantify a batter's contribution at the plate, in the field, and on the bases. It combines all of these things into RAR (rawr!!), Runs Above Replacement, and then converts runs into wins to get the WAR number based on how easy runs are to come by during that season. "Replacement" is a funny concept - this is a fictitious player who, theoretically, any team could pick up at any time to get 0.0 WAR in the lineup and contribute to more losses than wins. Think of a "Replacement" team as the worst teams in history - 1962 Mets (6.5 WAR batting, 3.1 WAR pitching), 2003 Tigers (8.1 WAR batting, -1.1 WAR pitching), etc. These are teams made of players that could be had off the scrap heap, or expansion draft, or whatever. Then add players' WAR to move up in the standings.

There are multiple versions of WAR, based on different ideas and defensive metrics. I will be writing about the Baseball Reference version.

So, back to the original point. An MVP candidate is supposed to contribute 8+ WAR, or close to it. Since 1901, there have been 273 seasons of 8+ WAR for batters*.

*This list does not include Miguel Cabrera - yet. Maybe he should work on his D.

There have been 565 seasons of 7+ WAR for batters. I don't want to accumulate the entire plot, but the 7+ WAR hitsogram looks a lot like the right side of a normal distribution:

That single data point on the far right corner is Babe Ruth in 1923. Ruth was a WAR behemoth, with the top two single-season totals, 6 of the top 12, and the most career WAR (163.2)*.

*Plus he also accumulated a significant amount of WAR as a pitcher. But this post is about batters only.

Actually, there are a lot of players who were WAR champions. Recently, Albert Pujols led the NL in WAR for 6 straight seasons 2005-2010. Before that, Barry Bonds* led the NL 12 out of 15 years 1990-2004. Long before, Willie Mays led the NL in WAR for 10 of 13 years, at around the same time Mickey Mantle was leading the AL in WAR five straight years. Ruth led the AL 11 of 13 seasons at about the same time Rogers Hornsby was leading the NL 11 of 13 seasons. Honus Wagner led the NL 8 straight seasons, and 11 of 13. It isn't a mistake that I am listing legends of the game here - WAR measures the things that make a player awesome at baseball and lead to that player's team winning games. The type of player that does well at WAR gets on base, hits for power, and plays solid defence, or does one or more of those things legendarily well.

*Old man (age 36 to 39) Barry Bonds led the league in WAR 2001-2004 with a "different style of play" than when young man normal career prime (age 25 to 31) Barry Bonds led the league in WAR 6 out of 7 years 1990-1996.

Aside - I will admit to a healthy scepticism about the accuracy of defensive metrics from before TV and play-by-play records really took off. Okay.

Ruth accumulated 14 WAR (worth near MVP production from TWO players) in 1923, a season that isn't generally acknowledged as his best. He achieved 12.4 in his legendary 60 HR 1927 season. He got up to 12.9 in his legendary 1921 season, where he set the all time (post-1901) single season marks in runs, HR, RBI, SLG, TB, and probably more (TB and runs records still stand). 1923 was remarkable, and also gets boost from 1.2 defensive WAR. Hmm.

Those seasons are at the top end. The other top-end seasons?

The list is a who's who of the best seasons ever. Poor Lou Gehrig crushed 1927 with 11.8 WAR and didn't lead the league (because of Ruth, #4). Notice that every single one of these hitters led the league in OBP and SLG (and therefore OPS)*. Because the type of player that does well at WAR gets on base, hits for power, and plays solid defence.

*Except Gehrig, who finished second in those things to his teammate.

The WAR list has lots of interesting things to notice. At least that I will notice. And point out, right now. Down the rabbit hole...

The highest WAR by a player who did not lead his league in a single offensive category was Willie Mays in 1963. The 32 year old Mays accumulated 10.6 WAR with no black ink:

There were two seasons where a player had MVP Level 8+ WAR with zero home runs:

But, those seasons were both dead-ball era, so how about the fewest HR during an MVP-Level season in the modern age?
Wade Boggs hit 3 during his 8.4 WAR 1989 season (51 2B, .430 OBP)
Willie Wilson hit 3 during his 8.5 WAR 1980 season (230 hits, 15 3B, 79 SB)
And Lou Boudreau hit 3 during his 8.0 WAR 1944 season

Of our MVP candidates, the largest Fielding RAR season belonged to (tie) 1989 Young Barry Bonds and 1927 Frankie Frisch, with 37 RAR each. 1989 Young Barry Bonds managed to get to 8.0 WAR with a .248 average and 19 home runs on the strength of his 3.5 defensive WAR.

 The largest baserunning RAR contribution belonged to (surprise) Rickey Henderson with 18 RAR in 1985. It was NOT his 130 SB season. It was not a 100+ SB season. It was his 80SB-10CS 1985 season. Among our MVP candidates, there are only 10 seasons with 10 baserunning RAR or better*. It would seem that baserunning is not the most direct path to MVP-Level WAR.

*Including Mike Trout's 10.9 WAR 2012, where his 49SB against 5CS led to 10 baserunning RAR

OBP is widely considered to be the most important offensive statistic. The more outs a player makes, the less valuable that player is, and the sooner games end. The lowest OBP on our MVP candidate list was Brooks Robinson with .304 in 1968 on his way to a 8.4 WAR. However, we also know that 1968 was the Year Of The Pitcher, so that .304 OBP isn't so bad compared to a league average of .299! And we also know that Brooks Robinson, Defensive Wizard of the Century, derived an incredible portion of his value from his third base defence. In 1968 he was credited with 4.5 WAR for defence alone.

The "alltime" record for single season dWAR is apparently Terry Turner, who logged 5.4 in 1906. I don't see how it is possible to calculate dWAR from 1906. The highest dWARs that I find remotely credible are 4.9 (total of 5.0) for Mark Belanger in 1975 and 4.7 (total of 7.3) for Ozzie Smith in 1989.

But back to OBP for a moment. Only 79 of our 273 MVP candidates have an OBP under .400, which makes for a pretty great season. Only 9/273 have WAR below .360. OBP is huge in racking up a big WAR. And it is one of the keys of being an effective baseball player, contributing to actual wins.

I'm pretty sure that more fun with WAR will come later.

No comments:

Post a Comment