Jason Varitek’s doubles and GIDPs
Posted by Andy on January 15, 2011
This is an oddly-specific post based on something a reader commented on in Steve's recent post about most doubles at Fenway Park.
Let's see how often Varitek grounded into double plays in games in which he also doubled vs games in which he did not.
I'm going to show you how I did this analysis in case you want to follow along at home. From the main PI page, I put Jason Varitek in the "Player Specific Tools" search box, clicked the bubble for Batting Game Finder, to get to a PI search page for Varitek's career games.
By setting 2B>=1 as the search criteria, and picking "Find Players with Most Matching Games in Multiple Years", you get the following summary stat line for Varitek:
| Rk | #Matching | PA | AB | H | 2B | 3B | HR | RBI | BB | SO | SH | SF | IBB | HBP | GDP | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 272 | Ind. Games | 1129 | 1020 | 459 | 296 | 0 | 34 | 238 | 90 | 157 | .450 | .495 | .840 | 1.335 | 6 | 6 | 5 | 7 | 15 |
So there are his totals for games in which he doubled. (Keep in mind that these are total stats for home and away while Steve's post was looking just at what happened at Fenway Park.)
So Varitek had 15 GIDPs in 1129 plate appearances, a rate of 1.3%. However, if we ignore the 296 doubles, then he had 15 GIDPs in the other 833 plate appearances, a rate of 1.8%.
To find out what Varitek did in the rest of his games, we could subtract the numbers above from his career totals, or just go back and do another Game Finder search and set 2B=0. Obviously, for every game in his career, he either hit no doubles or one or more doubles, so between these two searches we'll have every game in his career to date.
| Rk | #Matching | PA | AB | H | 2B | 3B | HR | RBI | BB | SO | SH | SF | IBB | HBP | GDP | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 1206 | Ind. Games | 4460 | 3857 | 799 | 0 | 13 | 148 | 483 | 503 | 992 | .207 | .304 | .329 | .633 | 13 | 37 | 54 | 50 | 111 |
So that's 111 GIDPs in 4460 PAs, or 2.5%.
Why the big disparity I wonder? My first thought is that if Varitek isn't doubling in a game, then maybe he's facing a tougher pitcher, and therefore he's less likely to hit line drives and more likely to hit a weak grounder suitable for a GIDP. But my gut tells me that this effect wouldn't be large enough to account for the difference.
In the games in which he doubled, Varitek had 4.15 plate appearances per game. In the ones in which he did not, he had 3.70 plate appearances per game. It stands to reason, perhaps, than in a game in which he GIDP, he was more likely to be removed for a PH or defensive substitution, and therefore less likely to get a double in that same game.
I'm sure there's got to be some simple reason that I just can't come up with...answers, please?
January 15th, 2011 at 9:09 am
If he has doubled in a game then there are fewer remaining plate appearances in which he will gidp. If he doubles once in a game there are 3.15 pa's in which he can gidp. If he doesn't double there are 3.7 opportunities to gidp hence the greater rate.
January 15th, 2011 at 9:37 am
Is there a way to get the PI to total up Varitek's stats from just May 20th of each year? He had a bunch of years where he hit one or more homers on May 20, but I'm not PI-savvy enough.
January 15th, 2011 at 10:17 am
Hahahahaha!!!!
This made me so happy this morning, I can't even tell you. Hilarious. As I was looking more into it, I realized my estimate of how many I was expecting was too high. However, I maintain he should have had more than he did - and I'm glad to see that you thought the same! 🙂
One thing that comes to mind is that the doubles themselves subtract from the ratio of everything, because they are taking up plate appearances. But that doesn't really explain the second thing you've said. Perhaps Varitek (maybe all or most players?) are less likely to GiDP if they're hitting the ball well in a given game. Or perhaps it has something to do with the nature of the double. To hit doubles, you must be driving the ball well, rather than hitting weak grounders into the infield.
On the other hand, maybe it had more to do with the fact that the people in front of him weren't usually on base in those particular games (as a fluke thing). There's a pretty wide disparity between 2B and RBI so that may be true (also true for a couple of other guys on the list in the Fenway-only post, like Yaz - but he was in a much lower-scoring offensive era, so perhaps that's to be expected).
Most likely, though, it's just chance, and I happened to find it odd, so I said something. But thanks for the follow-up.
January 15th, 2011 at 2:17 pm
As with most phenomena, there are probably multiple contributing factors. Some potential factors have already been pointed out in the comments above, such as:
*PA ratios
*men-on-base situations (to piggyback on this: a pitcher is probably more likely to give up a pitch that can be driven well when there's no runner on 1st)
*games where Varitek is hitting well (to piggyback on this: according to http://baseball-reference.com/blog/archives/9677, Varitek hits .456 in home games in which he doubles)
Another potential factor that comes to mind:
*For a set of games in which Varitek averages over 10% fewer PA than another set of games, it is probable that the whole lineup is averaging fewer PA in that set of games, which probably means that the pitching is better in that set of games, which makes GIDP more likely and 2B less likely.
January 15th, 2011 at 4:02 pm
I think this is all random. Varitek never grounded into that many DPs. The struck out quite a bit and batted pretty low in the order.
January 15th, 2011 at 8:40 pm
Every scientific experiment needs a control. Unfortunately, I'm not sure if there's any easy to way to compare all players in games where they doubled and when they didn't, but I'll look at the guys with the most doubles this century.
Bobby Abreu: GIDP 1.9% of non-double PA in games when he doubles, 1.8% in others
Todd Helton: 1.6%, 2.3%
Albert Pujols: 3.1%, 3.2%
Orlando Cabrera (surprise): 2.0%, 2.5%
David Ortiz: 1.7%, 2.1%
So in games in which they did not double, these guys GIDP about 15% more often than in non-2B PA in games when they did double. Varitek did it 38% more often. I have a small sample, so I question how meaningful that difference is. I'd guess Varitek does have one of the bigger differences among all players, but I bet a lot of that is chance.
January 15th, 2011 at 9:12 pm
While there may be a good explanation, I don't think the removal for a PH or defensive substitution are it. He was more likely to be a PH and/or defensive replacement on the games he didn't catch than he would be pulled out of the games he did catch.
January 16th, 2011 at 1:24 am
Following Josh's last point. Could the pitchers against whom he doubled be relatively more fly ball pitchers? Shouldn't we be able to divide out the PAs in which there was a runner on 1st with < 2outs. We want to be comparing like situations. It may be that in the games where Varitek doubled he just had fewer potential DP PAs. Randomly or for unknown reason we haven't yet seen. My bet, I'm not giving long oddsis on random.
January 17th, 2011 at 11:59 am
I agree with #1 above. That was my first reaction to the PA stats provided.
January 17th, 2011 at 1:42 pm
Varitek hit 121 games doubles that were his only hit in the game. That lowers his other at bats to three or four. Less chances.
Varitek hit 117 games with doubles where he had two hits for the game. That lowers his other at bats to two or three times. Less chances.
Varitek had 32 games with doubles where he had three hits in the game. That lowers his other at bats to one or two. Less chances.
Varitek had 2 games with doubles where he had four hits in the game. That lowers his other at bats to zero or one.
If we do the math, then there were about 629 plate appearances in games where he doubles where he didn't get a hit. Multiply that by 2.5 percent (what you said was his career average), that comes to 15.75 GIDP. There you go.
January 17th, 2011 at 5:39 pm
William, you are comparing bruised oranges to mushed apples, that is you are making many errors while comparing unlike things. The 2.5% figure is GiDP/PA in games where Varitek does not hit a double. This was properly compared to 1.8% of PA's where he did not double in games in which he doubled. You are comparing this to PA's in which he did not get a hit, (which you misfigured, it is 670, not 629), in games in which he did get a double. A better comparison would be to double play situations, (man on first, < 2 out), in games in which he did or did not double.