For a while it seemed Aaron Judge was going to defy history and post a winning batting average despite a strikeout rate horrendously high even in our modern-day, whiff-happy context.
Judge’s average was up to .347 on July 12. Since then, entering play on Thursday, it sunk to .291. In that period, he’s hit just .230 while striking out 87 times in 196 at bats (a strikeout every 2.25 at bats). That rate was a K every 2.96 ABs previously.
This raises two questions we should know in fantasy since strikeout rates for hitters stabilize early in seasons — what is the expected average for various levels of Ks and what kind of trade-off can we expect in increased homers?
A third question has to do a major component in batting average — average on balls in play. Theoretically, players who strikeout swing harder and thus hit the ball harder (when they hit it) increasing their BABIP.
The answers are that Ks are closely correlated to batting average and this type of cratering by Judge should have been expected. And the higher the Ks, typically the higher the expected homers. Finally, and most surprisingly, BABIP doesn’t increase with strikeouts but there is a sweet-spot where slightly higher K hitters do seem to achieve bettable BABIP gains.
Since 2000, there are only 10 hitters at the levels of ABs/K that Judge has been mired in for two months and their average batting average was .227 with a .304 BABIP.
From 2.51-to-3.00 ABs per K, those numbers increase to .239 and .305 and the sample grows to 45 hitters. The next strata to 3.5 ABs/Ks was .256 and .312 (122 players). The BABIP sweet spot is from 3.51 to 4.00, where BABIP is .315 and overall average .266 (237 hitters). Finally 4.01-4.5 is .268/.309 (321 hitters) and 4.51-5.0 (317 hitters) is 275/.309.
Note the average ABs/K this year for the qualifying 159 hitters is 4.98, an all-time high.
The interesting thing about Ks and homers, which we assume are connected since both are increasing in tandem, is that the game’s most dominant long-ball specialist, Giancarlo Stanton, is hitting them at a record pace of late while actually lowering his K rate. I wrote about this on Thursday in The Wall Street Journal.
Historically, the homer-bump for hitters at that highest-K level is significant — about 30%. This group includes Judge and other sluggers Joey Gallo, Keon Broxton, Chris Davis, Miguel Sano and Trevor Story. But a bad batting average is an unavoidable trade off, as Judge owners are now realizing. And Judge is still inflated and he’s likely to hit about .240 for the remainder of the year. That should be his projection in 2018, too, unless he significantly alters his fan-tastic ways.
There are only 21 hitters with a AB/K below 3.50. They average 24 homers to date. But their average is .254. This indicates that Chris Taylor’s average of .303 is inflated given an AB/K of 3.4 that’s one spot lower than Cody Bellinger. Similar to the larger sample, the nine hitters between 3.0 and 3.5 this year (Will Myers, Mark Reynolds, Tim Beckham, Hunter Renfroe, Justin Upton, Michael Conforto, Taylor, Bellinger and Jake Lamb) have the highest BABIP of all the levels (by far at .339).
Another batting average fluke to sell given his K rate is Tommy Pham, who I keep saying to sell but who stubbornly has kept his average above .300 for a month. Pham’s tier hits .262 (30 players). After Pham’s .310, the next highest average in his K tier is Stanton (.287).
Scott Schebler and Rougned Odor’s averages seem artificially low. But Odor is going to have a bad BABIP and underachieve even his K% with an infield pop-up rate of 18.6%. Schebler is on the DL with a shoulder strain but his owners have suffered unfairly in the average category this year.
The biggest outlier on the low-average side is a player who is often maligned, Yasiel Puig, who has a .262 BABIP and thus a .253 average. In 2014 when he hit .296, Puig had a higher K% and a virtually identical hard-hit rate, according to Fangraphs (33.8% this year vs. 34.6% then). However, like Odor, Puig is popping up too much at nearly double his 2014 rate. His groundball-rate is unchanged (still high) so this is unlikely the product of an altered approach, as seems to be the case with Odor.