Pedroia's Power Outage
I wish I'd been able to think of a snappier title, but that one gets right to the heart of the matter. Less than a year after winning the MVP, Pedroia's Green-Monster-aided slugging stands at a paltry .392, more than .100 below the mark he put up during his 2008 campaign. Since I go out of my way to avoid watching Red Sox games, I have no idea why this might be the case. He's getting hits less often, yes, but a thirty point drop in batting average makes a hundred point slugging implosion not. The fun part of not knowing anything? We get to test out any hypotheses and play around with data! (I think that's fun)
For about two months at the beginning of the season, it was pretty difficult to escape a commercial featuring Dustin Pedroia defend himself from the accusation that he couldn't hit a high-inside fastball. With that springing to mind whenever I think of Mr. Pedroia, and because it'd be really funny if true, my belief as to his power-outage is two-fold:
- Dustin Pedroia cannot do anything with a high-inside fastball
- Pitchers saw that commercial so much they started throwing more high-inside fastballs.
Obviously this would imply that teams pay more attention to television than advanced scouting, but hey that'd be hilarious too, so I'm really hoping that we can confirm this twin hypothesis. How do we test it out, though?
For the first part, I've grabbed 2009 pitch F/X data from the StatCorner database for every pitch Pedroia's seen this year. For simplicity's sake, and because it gives a much bigger sample, I've decided to lop off the 'fastball' part of this little theory. Below is a chart of everything thrown to Pedroia this year. Also shown is the strike zone zoning I've used. Left on the graph is in, right is away, and I'm going to trust my readers enough to tell up from down.

Figure 1: All pitches thrown to Pedroia in 2009.
Obviously it's a little hard to draw any conclusions from that, but it gives you a pretty good idea as to the way pitches have been distributed - all over the place. In order to get any idea as to which ones Pedroia hits best, we're going to have to actually look at the numbers.
First, I discard everything not in the strike zone. This seems like a pretty big deal since it removes a fair amount of hits, including XBH, as well as taking out most of his swinging strikes, but when I did bother looking at it it didn't change my conclusions much -I won't bore you with extra work for no reason. Next, we chose which metrics we value.
I'd argue that the following are important (and feasible to extract from raw pitch FX data) for explaining a lack of power:
- %pitches for determining that zone's effect on overall batting line
- %swings for similar reasons
- %contact as swinging and missing is generally not a positive outcome for a hitter
- Batting average on pitches made contact with as equivalent to OBP
- Slugging percentage on pitches made contact with because that's what we're trying to work out
- Isolated slugging (slg-ba) on pitches made contact with to give us another way of looking at power
With these chosen as my value metrics, it's a relatively trivial task to set up a spreadsheet to crunch the actual data for us (which I hope I've done correctly). The table below shows each of the above numbers for each zone. Which zone is which should be self evident, and green/red cells show positive/negative outcomes from Pedroia's standpoint.

Table 1: Value metrics split by zone.
That's not what was expected, not at all. While it does confirm that Pedroia can't do much to hurt you with the high-inside pitch (an isolated slugging% of 0.091 on pitches you hit would put you in the company of such luminaries as Willie Bloomquist), a much more surprising find is that he's just as bad with pitches straight down the middle, which is an area hitters normally feast on. The fact that he's slugging just .371 with an iso of 0.086 on pitches thrown straight down the pipe is absolutely bizarre. Pedroia's best zones by my analysis are high-middle, up and away, and down and in, for which there's no obvious reason. He's hitting .556 on pitches down and in and a few inches above that he's hitting .188. This could well be a sample size issue, as the number of pitches considered in any zone is fairly small even over half a season.
You can actually get a fair amount of interesting information off of these charts, most of which is only tangentially related to my original idea. Considering the problems Pedroia's been having elsewhere and that only 7.4% of pitches hitting the strike zone are making their way up and in, it's probably fair to say that part one of my hypothesis was way off base. Sure, he can't hit very well on the high inside fastball, but Pedroia has bigger issues to worry about. So since we have the data on hand, let's go over other random things of interest:
- I'm surprised that Pedroia's power drop looks like it's caused in home runs drying up rather than doubles, as I had assumed it'd be easy to take inside pitches off the monster rather than having to yank pitches on the outer half to left field. Perhaps his homers last year were coming from those inside locations - I don't know.
- Pedroia has yet to swing and miss at a pitch at three of the corners - no guesses as to which one induces some swinging strikes.
- Pedroia should really be swinging at more pitches down the middle. I wonder if the uptick in his walk rate is to do with him just swinging less, rather than being less selective. Statcorner's 'By Pitch' tab on Pedroia shows him swinging at 38.6% of pitches in 2009, down from 44.6% in '08, with a more or less irrelevent change in the distribution of pitches he took for a ball vs. a strike.
- Pedroia has an iso of 0.036 on pitches middle-down. He has to be getting way on top of those and getting a tonne of ground balls, because that's insane.
That's part one done and more or less disproved. How about my hoping that pitchers were paying more attention to their televisions than their advanced scouts, and are throwing Pedroia more high-inside pitches this year than last?

Table 2: 2008/2009 thrown% comparison.
So that's a no then. Where's the fun in that? They've done the exact opposite, moving away from the upper left edge of the strike zone, shifting heavily down and away, which ironically is one of the locations where Pedroia is actually doing damage. Why? I'd have to imagine a guy who slugged around .500 last year scares pitchers, and when you want to limit extra base hits, staying down and outside is where you want to be. That sort of thing apparently matters more to pitchers than video game commericials, which has quite frankly ruined my day.
0 recs |
28 comments
|
Comments
As a huge Pedroia fan,
I am worried about the lack of power. I think the work here is really interesting, and you got it pretty well spot on at the end.
The commercial was particularly funny, because on several notable occations Pedroia killed high and tight pitches for HRs in the ALCS against Tampa Bay, or at least high pitches. I think his HR power is pretty limited to high fastballs, where as he hits line drives at a high percentage all over the zone. Pitchers saw his big swing early on and given his small stature, they figured they would try to beat him high, doubting that he could make them pay. After 19 HRs they started to pitch him more down and away where he might hit well, but he is less likely to jack them. No one wants that little guy going yard against them, now do they!
by Slugger O'Toole on Jul 9, 2009 9:24 AM EDT reply actions 0 recs
ISO
Very interesting, Graham. I worry about basing conclusions on ISO for a particular part of the strike zone, though. Because ISO is a combined statistic (that is to say, one value subtracted from another), it stabilizes over a longer period of time even than either of its components (SLG and BA, or TB and H, whichever you prefer).
I’m not sure how many pitches you had total to work with, but after throwing out the ones that were outside the zone (with good reason) and splitting it up into nine zones, I can’t imagine there were too many left in any one zone. Since XBH are such lumpy outcomes, it seems just as likely that the results for individual parts of the strike zone are just as likely to be (un)luck.
Is there a way to test this null hypothesis?
by Tommy Bennett on Jul 9, 2009 9:35 AM EDT reply actions 0 recs
This
isn’t true, is it?
Because ISO is a combined statistic (that is to say, one value subtracted from another), it stabilizes over a longer period of time even than either of its components (SLG and BA, or TB and H, whichever you prefer).
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 9, 2009 9:39 AM EDT up reply actions 0 recs
Well
Isolated power is determined by subtracting one value from another. Since batting average has a (relatively) high rate of error, the significant figures are low. When adding or subtracting two values, you must use the number of significant figures for the number with fewer. So with only a few hits in each zone, we have very low sig. figs.
I may be wrong, but I believe I am right about that. It’s also why OPS can fluctuate so much over small samples (it is an added statistic).
by Tommy Bennett on Jul 9, 2009 9:57 AM EDT up reply actions 0 recs
Makes sense, I guess
Slugging includes batting average (sort of), and we all know that batting average can be very fluky. Then again, ISO subtracts the batting average part of Slugging, so it may actually be the opposite.
Derosa.
by vivaelpujols on Jul 9, 2009 10:10 AM EDT up reply actions 0 recs
I don't think this is true
ISO removes the effect of batting average from slugging.
by Graham on Jul 9, 2009 10:41 AM EDT up reply actions 0 recs
Ok, I understand my mistake
ISO = (2B+2*3B+3*HR)/AB and therefore does not involve any subtraction. Apologies.
by Tommy Bennett on Jul 9, 2009 11:11 AM EDT up reply actions 0 recs
Bleh.
SigFigs.
OverTheMonster - ALLERGEN WARNING: May contain peanut butter.
by bs.uf15bosox9bears23 on Jul 9, 2009 4:30 PM EDT up reply actions 0 recs
Now that I think about it
It stabilizes as quickly as BA. I think.
by Tommy Bennett on Jul 9, 2009 10:06 AM EDT up reply actions 0 recs
We have ~70 pitches in each area of the strike zone
That amounts to something like 20-30 at-bats for Pedroia in each area, a tiny sample to be sure. However, since we already know that something is up with his power on a much larger sample size (i.e. the season to date), I don’t think the sample sizes caused too many problems – or at least did not lead to any erroneous conclusions.
Going into the rest of the year, I’d still expect a more normal zone profile from Pedroia, but we don’t have to worry about sample size when we are only looking at what has happened to date.
by Graham on Jul 9, 2009 10:46 AM EDT up reply actions 0 recs
I was not being clear
we don’t have to worry about sample size when we are only looking at what has happened to date.
My thought is that his struggles might be due more to luck than to anything else. Is that what you’re saying?
by Tommy Bennett on Jul 9, 2009 11:13 AM EDT up reply actions 0 recs
I'm not sure
Gun to my head, I’d say that yeah, it’s probably luck, just because I’d expect his number on pitches down the middle to be way higher.
by Graham on Jul 9, 2009 11:21 AM EDT up reply actions 0 recs
To expand on reliability, there's always Pizza Cutter's famous two studies:
Number of PAs for a .7 split-half correlation (note that if a stat isn’t listed it either wasn’t tested or didn’t achieve .7 before 650 PAs:
Hitters
50 PA: Swing %
100 PA: Contact Rate
150 PA: Strikeout Rate, Line Drive Rate, Pitches/PA
200 PA: Walk Rate, Groundball Rate, GB/FB
250 PA: Flyball Rate
300 PA: Home Run Rate, HR/FB
500 PA: OBP, SLG, OPS, 1B Rate, Popup Rate
550 PA: ISO
Pitchers
150 BF – K/PA, grounder rate, line drive rate
200 BF – flyball rate, GB/FB
500 BF – K/BB, pop up rate
550 BF – BB/PA
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 9, 2009 11:49 AM EDT up reply actions 0 recs
Wow, I didn't know LD rate for pitchers stabalized so quickly
Derosa.
by vivaelpujols on Jul 9, 2009 11:51 AM EDT up reply actions 0 recs
Yeah, kinda weird, especially given the subjectivity in classifying LDs.
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 9, 2009 11:58 AM EDT up reply actions 0 recs
I have it as much closer to popup rate than ground ball rate
But then again I prefer LD/ball in air etc
by Graham on Jul 9, 2009 12:03 PM EDT up reply actions 0 recs
Yeah I think
My conclusion was correct but my reasoning was not. Thanks for that link.
by Tommy Bennett on Jul 9, 2009 11:52 AM EDT up reply actions 0 recs
So in doing the commerical, Pedroia was hoping to use some reverse psychology on pitchers, which has failed?
One thing I’m curious about is Pedroia’s stats by zone relative to league averages. I’m guessing most hitters hit high pitches for power, for example. So does Pedroia handle them better than average? You could even compare how Pedroia’s splits by zone compare to league average splits by zone (which is what B-Ref calls their “t” split, like tOPS, I believe.)
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 9, 2009 9:37 AM EDT reply actions 0 recs
I would also be very intersting in seeing this
Derosa.
by vivaelpujols on Jul 9, 2009 10:10 AM EDT up reply actions 0 recs
I have to remember than this blog is set for East Coast time
by Graham on Jul 9, 2009 10:23 AM EDT reply actions 0 recs
No, you have to remember that I have the power to reschedule posts ; )
We had a hole, so I moved this up.
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 9, 2009 10:25 AM EDT up reply actions 0 recs
I, for one
like coming in to work and having something ready to distract me immediately.
by jwiscarson on Jul 9, 2009 11:12 AM EDT up reply actions 0 recs
It seems to me that Dusty swings a lot at pitches out of the zone, so maybe the pitches you avoided should be looked at.
ZiPS has his ISO coming back closer to last years for the rest of the season, projecting at .146
OverTheMonster - ALLERGEN WARNING: May contain peanut butter.
by bs.uf15bosox9bears23 on Jul 9, 2009 4:42 PM EDT reply actions 0 recs
.
First, I discard everything not in the strike zone. This seems like a pretty big deal since it removes a fair amount of hits, including XBH, as well as taking out most of his swinging strikes, but when I did bother looking at it it didn’t change my conclusions much.
by Graham on Jul 9, 2009 5:06 PM EDT up reply actions 0 recs
You expect me to read things?
OverTheMonster - ALLERGEN WARNING: May contain peanut butter.
by bs.uf15bosox9bears23 on Jul 10, 2009 12:12 AM EDT up reply actions 0 recs
Observation
I am an avid Pedroia Fan and watch the Sox on a regular basis. An observation I have noticed is that Pedroia is streaky. He leads the Majors in 3 hit games. He will get into hitting streaks where no one can get him out. With Big Papi hitting so poorly for most of the year the pitchers were not worried about giving a single to Pedroia because they knew Papi would get out.
It looked like when the pitchers knew Pedroia was hitting well they would pitch him away and hold him to bloop singles down the left field line with the occasional double. I mean he is the reigning MVP, if you could hold the previous MVP to a single you would be happy too.
by drabidea on Jul 10, 2009 1:30 PM EDT reply actions 0 recs
Yes
last year Pedroia had more than 1/3 of his total HRs in August. Probably to a degree 17 HRs were a fluke, but he is streaky in all aspects of his hitting.
by Buzzy on Jul 13, 2009 11:01 AM EDT reply actions 0 recs

by 










BtB on Facebook

















