BP Comment Quick Links
September 19, 2001 Aim For The HeadHit-by-Pitch RatesThis week's question come from P. N., who asks:
I'm an Oakland A's fan and there's been a lot of talk lately about how the A's are leading both leagues in the number of batters hit by pitches. Tim Hudson has suggested that the pitching staff needs to be more aggressive in protecting the hitters. Thanks for the question, P.N.; let's take a crack at testing this hypothesis. If the A's HBP totals were solely due to seeing more pitches per plate appearance (and teams otherwise had no ability to induce more HBP), we should see a relationship between the HBP rate (HBP per PA) and the number of pitches seen per PA (NP per PA). Plotting them for each team during 2001 to date (through September 17):
There's not much of a linear relationship. We have examples of teams that see a lot of pitches, and get hit a lot (which happens to be Oakland), but we also have teams that see even more pitches, and are among the lowest teams in the rate at which they get hit (San Diego). We've got outliers at just about every corner. Computing the correlation between HBP/PA and NP/PA yields +0.130, which at best indicates a mild relationship, given these relatively few data points. Perhaps that not the right approach. Taking lots of pitches can lead to taking more walks. Could a high OBP lead to more opportunities to get hit by a pitch, thus increasing the team totals? There's a bit of a complicating factor in that getting hit by a pitch directly increases OBP, and yields another PA in which a batter could be hit. However, the frequency of HBP is small enough where we can ignore it for now. One way to test this alternative theory would be to compare HBP per game to plate appearances per game. If a team has a high OBP, they will have more plate appearances per game (excluding extra-inning games), which would in turn give more chances to get hit per game. I've done this analysis, and the correlation between the two stats (HBP/G and PA/G) is +0.199, slightly better than we saw before, but not that strong of a relationship. I've included the data for each team's HBP, PA, and NP rates at the end of the article, including how each team ranks among all 30 major-league teams. It's interesting to note how different the Astros and Padres are in the HBP/NP and NP/PA categories: Houston ranks first in times getting hit per pitch, but 28th in pitches per PA. The Padres are the opposite, ranking dead last in how likely a single pitch is hit a Padre batter, but first in how many pitches they see each time they come to the plate. Though these two examples might lead you to think that there's some kind of inverse relationship between the two categories, that's not the case. The correlation between NP/PA and HBP/Pitch is virtually zero (+0.044), indicating that the two categories have no relationship (positive or negative) to each other when you look across the entire league. So it doesn't look like the A's propensity get plunked can be blamed on their taking more pitches. There's a slightly positive relationship, but the primary explanation lies elsewhere: either in the batting styles of the A's players themselves, taking stances closer to the plate, or perhaps a coaching strategy to get them to "lean in" to a pitch that's a little too far inside. Or maybe Tim Hudson was on to something...
Team Lg G HBP PA NP PA/G HBP/G HBP/PA HBP/NP NP/PA ANA A 144 68 5497 20744 38.2 0.47 0.012 0.0033 3.77 BAL A 142 68 5330 19974 37.5 0.48 0.013 0.0034 3.75 BOS A 141 60 5466 20489 38.8 0.43 0.011 0.0029 3.75 CHA A 144 46 5416 20507 37.6 0.32 0.008 0.0022 3.79 CLE A 144 62 5616 21343 39.0 0.43 0.011 0.0029 3.80 DET A 143 46 5395 19929 37.7 0.32 0.009 0.0023 3.69 KCA A 143 39 5429 19762 38.0 0.27 0.007 0.0020 3.64 MIN A 144 59 5491 20603 38.1 0.41 0.011 0.0029 3.75 NYA A 143 55 5521 20998 38.6 0.38 0.010 0.0026 3.80 OAK A 144 81 5665 22031 39.3 0.56 0.014 0.0037 3.89 SEA A 144 54 5713 21767 39.7 0.38 0.009 0.0025 3.81 TBA A 143 44 5346 19788 37.4 0.31 0.008 0.0022 3.70 TEX A 144 68 5672 21548 39.4 0.47 0.012 0.0032 3.80 TOR A 143 64 5526 20595 38.6 0.45 0.012 0.0031 3.73 ARI N 144 50 5536 21082 38.4 0.35 0.009 0.0024 3.81 ATL N 143 38 5369 19440 37.5 0.27 0.007 0.0020 3.62 CHN N 143 58 5374 20484 37.6 0.41 0.011 0.0028 3.81 CIN N 144 59 5505 20458 38.2 0.41 0.011 0.0029 3.72 COL N 143 54 5548 20478 38.8 0.38 0.010 0.0026 3.69 FLO N 144 58 5457 20203 37.9 0.40 0.011 0.0029 3.70 HOU N 143 80 5530 20236 38.7 0.56 0.014 0.0040 3.66 LAN N 144 50 5406 20223 37.5 0.35 0.009 0.0025 3.74 MIL N 145 65 5429 20638 37.4 0.45 0.012 0.0031 3.80 MON N 144 58 5294 19635 36.8 0.40 0.011 0.0030 3.71 NYN N 145 61 5463 21088 37.7 0.42 0.011 0.0029 3.86 PHI N 144 38 5487 20868 38.1 0.26 0.007 0.0018 3.80 PIT N 144 61 5313 19866 36.9 0.42 0.011 0.0031 3.74 SDN N 144 38 5573 21945 38.7 0.26 0.007 0.0017 3.94 SFN N 144 41 5616 20773 39.0 0.28 0.007 0.0020 3.70 SLN N 144 59 5398 20020 37.5 0.41 0.011 0.0029 3.71
Keith Woolner is an author of Baseball Prospectus. You can contact him by
clicking here.
Keith Woolner is an author of Baseball Prospectus. 0 comments have been left for this article.
|