clock menu more-arrow no yes mobile

Filed under:

Jose Altuve and the risks of contact hitters

Talking Astros Sabermetrics: Altuve was baseball's worst at the "little things" and maybe that's a risk of contact hitters

Duane Burleson

Matt Klassen at fangraphs has an award called "King of Little Things." As you can tell by the name, the award is given in a lighthearted way. That's because the award is based on win expectancy but gives players credit for "the little things," like moving runners over and bringing runners home with a sacrifice fly. The value is based upon the difference between WPA/LI and batting wins above average. The former is based on the batter's impact on win expectancy associated with the base-out state (runners on base, outs, inning). The latter is based on the the average win expectancy of the batting outcome, as derived from linear weights.

Klassen explains:

Two reminders about what "Little Things" (now the name for better or worse) covers here: a) things like grounding into double plays, sacrifice bunts and flys, and other things not currently incorporated into wRAA [weighted runs above average]; b) the player’s response to the game state beyond what traditional linear weights measure: base/out situation, relative score, and inning.

We can go to the Astros' connection now: in 2013, Jose Altuve was the worst hitter (minimum 500 PA) in baseball at doing the little things (at least based on this measuring stick). He was last, no. 141, at -1.047. Altuve was approximately 1 win below average as a batter, based on average expected results calculated with linear weights. However, he was more than 2 wins below average based on WPA/LI---meaning that his production in runners on base situations cost the Astros about 1.5 more wins than expected. One of the most damaging factors for Altuve is that he grounded into 24 double plays in 2013.

Before you get too upset, it's unclear whether this "little things" calculation is a repeatable statistic. Sure, there are some characteristics which may predispose a player to good or bad outcomes in certain base-out states; for example, certain types of players are more likely to ground into double plays. On the other hand, luck (or randomness, if you prefer) likely has a significant effect on the smaller samples of base-out states. Last year, Altuve's WPA/LI was slightly better than expected.

WPA/LI is similar to RE24 (Run Expectancy for 24 Base-Out States), which I wrote about earlier this year. RE24 focuses on the player's impact on run expectancy for each at bat's base-out state, ignoring leverage, score, and inning. In that article, I compared RE24 to weighted runs above average to evaluate whether hitters were better than average at increasing / decreasing the probability of scoring a run in the base-out situations they faced.

Jose Altuve had a wRAA of -9.11 and a RE24 of -17.62, which means that Altuve's performance based on run expectancy was about 8.5 runs worse than the average player produced with the same batting events. Since 9 - 10 runs is equivalent to a win, that means Altuve is almost 1 win worse when base-out situations (as measured by RE24) are taken into account.


Looking at the Altuve example, perhaps this shows the hidden risk of contact hitters. The risk associated with high strike out hitters is apparent. That's why many fans yearn for players who make contact and put the ball in play. While contact hitters can be beneficial, they bring risks, in terms of variability. A lot of things can happen when the ball is put in play--and sometimes the outcome is worse than doing nothing at all. The contact hitter's run production with runner's on base will be subject to BABIP variability. In addition, possibly the contact hitter is more anxious to swing with runners on base, resulting in fewer walks.

We can see some of these signs for Altuve in 2013. Altuve committed 11 of his 24 GIDPs on the first pitch. 16 of his GIDPs occurred on the first two pitches. When Altuve swung at the first pitch, his BABIP was .287. If he took the first pitch, his BABIP was .335.

To a large extent, Altuve's last place position on the "little things" list is due to BABIP variability. With runners on base in 2013, Altuve's BABIP was .290 versus a BABIP of .333 with no runners on base. In 2012 Altuve had a wRAA which was the same as his RE24. Altuve's BABIP differential, with and without runners on base, was only 4 points for that period,

For the five worst "little things" hitters iin each of Klassen's articles from 2009 - 2013, only six of the 25 hitters had a strike out rates equal or greater than league average. Thus, most of the five worst "little things" hitters had below average strike out rates, some by a substantial margin. On the surface, it appears that contact hitters are more likely to appear on the five worst list. This is ironic, because contact hitters are touted for their ability to advance runners and do the little things.

Among Astros' hitters, Jason Castro and Chris Carter lead the team in RE24 by a considerable margin--despite their respective strike out rates of 27% and 36%. Castro's RE24 exceeds his wRAA by 17%, which means that his 4.3 WAR understates his actual run production.


The manager's batting order decisions also may affect the ability of hitters to achieve or exceed average run expectancy performance. Jose Altuve's propensity for grounding into double plays varied based on his batting order position. Altuve's GIDP / PA is shown by batting order position below..

Batting Position / GIDP Rate

1st / 2%

2d / 4.2%

3d / 3.6%

Of course, it makes sense that Altuve's GIDP rate is lowest in the lead off slot, because he probably sees fewer runners on base from that slot. However, Altuve's walk rate has to increase in order to justify a lead off position. (Altuve's walk rate in the lead off slot last year was 1.5%---yikes.) If Altuve took the first pitch last year, he was almost 5 times more likely to walk during the plate appearance. If Altuve is discouraged from swinging at first pitches, his walk rate may increase and potentailly his GIDPs could decrease.