What’s the Pitcher-Hitting Equivalent of a .400 Hitter?

Not too long ago, FanGraphs editor Carson Cistulli was watching a broadcast of a baseball game during which the announcer remarked offhandedly that the pitcher’s batting average of .231 was equivalent roughly to a .400 mark for a position player. His interest piqued, Cistulli asked his FanGraphs colleagues: if not .231 precisely, then what is the equivalent of a .400 batting average for a pitcher? After nobody else expressed any interest in doing the same, I endeavored to answer the mostly frivolous question.

The easiest way to go about solving the problem is probably to look at percentile ranks — that is, at seasons from position players, broken into percentiles by batting average, compared to the same percentiles for pitchers. That’s where I started, at least. I looked at all qualified position players from 1986 to 2015, finding nearly 5,000 player-seasons. Then I turned to pitchers. Because no pitchers qualified for the batting title during that time range, I chose a threshold (a somewhat random figure of 50 plate appearances in a season), yielding nearly 1,500 pitcher player-seasons.

I created percentiles for both groups and set them at 10%, 33%, 50%, 67%, and 90% to yield averages. The table below shows the results:

While not a perfect methodology, we find here that, in a small sample, a .231 average for a pitcher is not the same as a .400 hitter. A .231-hitting pitcher is still a very good result for a pitcher and would be in the top 10% of seasons over the last 30 years. After all, there have been just 19 pitchers to bat over .300 in a season in the last 30 years, and Dan Haren was just two hits away from .400 back in 2010. If we set the bar at .300 instead of .400, we find that the pitcher equivalent is a .190 batting average. Of course, over the course of just 50-100 plate appearances, randomness will very well occur. Consider: a true-talent .200 hitter is just as likely to hit .400 over the course 40 at-bats as a true talent .300 hitter is over the course of 170 at-bats (0.3%).

While batting average is often used in broadcasts, it’s generally discussed at FanGraphs as one of three main slash stats. Here are the equivalencies for the other two slash stats, on-base percentage and slugging percentage:

So that .167/.200/.212 pitcher you know of? That’s the pitcher equivalent of batting .290/.362/.475 — or, at least over the last 30 years it is, which admittedly includes some pretty robust hitting years. The hitting line is also derived from considerably more talent, as those seasons all include qualified batting seasons as opposed to seasons with roughly one-tenth the plate-appearance totals

Before getting to a somewhat larger sample size, here’s a similar exercise I performed, except with wRC+ to help control a little bit for the different eras over the past 30 years. Here are those numbers:

You might notice that the 50th percentile for hitters is at 110, which is considerably above the overall average of 100. This is due in part to the the fact that good hitters get more plate appearances and are more likely to qualify for the batting title. As an example, last season the average wRC+ of the 125 hitters with between 300 and 500 plate appearances was 96. For the 177 players with between 100 and 300 plate appearances, their average wRC+ was 83 last season.

Regardless, the main lesson here is likely that pitchers are not very good hitters and that, even over the course of a full season, a decent hitting season from a pitcher doesn’t make much difference in terms of wins and losses.

Moving on to the slightly less frivolous, but still definitely frivolous, pursuit of the question above, what happens when we use multiple seasons and require more plate appearances? Here we won’t even bother ourselves with comparisons to a .400 average, as few players in modern baseball have even approached that mark in a single season. Repeating the exercise above using the same 1986-2015 seasons, but no longer splitting the seasons up, we might get a more accurate representation of a pitcher’s hitting equivalents.

I raised the bar to pitchers who’d recorded at least 250 plate appearances and made a similar raise for position players, moving them up to 2,500 PA minimum. That leaves 204 pitchers and 742 position players in the respective samples. We still won’t get quite to the true talent of the pitcher, as 250 plate appearances across several seasons will not tell us that much, but it’s more than what we saw above. Shooting for more than 250 plate appearances reduces the number of players in the pool to make the results even more meaningless.

First, batting average:

For a pitcher, hitting .207 is roughly the equivalent of a .300 hitter.

Next, on-base percentage and slugging:

At that 67% mark, the pitcher is at .157/.193/.201 while the position player comes in at .276/.347/.444. Given the era with which we’re dealing, those equivalencies can feel a bit high.

Here is wRC+:

The hitters at the 50th percentile come in much closer to the 100 average. The pitchers are simply generally bad — and even the really good pitchers are bad compared to their positional colleagues. In the past 30 years, 8,794 players have recorded 250 plate appearances in a season, and only nine have posted a wRC+ below 32. In the last 15 years, the only position players to have such seasons were Tomas Perez (26 in 2006), Reid Brignac (24 in 2011), and Drew Butera (19 in 2011).

Given the sample sizes involved, there probably isn’t a real equivalency we can find between pitchers-as-hitters and position players, but the comparisons we can make do support one point: it might be time for the universal designated hitter.