GIDP: The Underrated Production Killer
Earlier in the week, I wrote a post about how grounding into double plays erased 16.5% of Jim Rice's career offensive value. Of course, Rice isn't the only player I looked into, but the story really evolved to go into Rice further. Today, I'll share a few graphs of some other players of interest.
One player who was similar to Rice is Joe Torre. Torre actually was worth more offensively than Rice (300 runs above average to 279) and his double plays were just a tad less damaging (-43 runs compared with Rice's -45). As a result, the double plays cost him 14.3% of his career value, by far the highest total of any player with 300+ batting runs above average (Rocky Colavito was second at 8.5%).
Among the players with the worst GIDP runs in history, George Scott also stood out to me because he had far fewer batting runs than Rice and Torre. As a result, 36.4% of his offense was erased.
The highest active player (though he did announce he's now retired) is Mike Lowell.
Think that's bad? How about poor Eric Karros? A full 40% of his offense was wiped away by the ol' 6-4-3.
Lastly, this one was even tough to visualize in the same manner as the others. Brooks Robinson, of course, is a Hall of Famer with his glove. But he still collected 2848 hits and 268 home runs, which would make you think he was a pretty darn good offensive player. The problem is, it took him forever to do this, and he didn't have much plate discipline while doing it either. It all adds up to a marginally above average offensive player, according to both batting runs (+20) and OPS+ (104).
Sadly, all of it (and then some!) is wiped away by double plays. The GIDP knocks Brooks Robinson into a slightly below average offensive performer.
Here's the table that I used (from Rally's WAR database, then hand-updating the active players from Baseball-Reference). I used players with -20 GIDP runs or worse, then sorted by percentage of offensive value erased.
| Name | PlApp | Bat | DP | WAR | Pct |
|---|---|---|---|---|---|
| Gus Triandos | 4368 | 2 | -21 | 12.9 | 1050.00 |
| Deron Johnson | 6546 | 11 | -21 | 5.1 | 190.91 |
| Brooks Robinson | 11567 | 20 | -35 | 69.1 | 175.00 |
| Larry Parrish | 7716 | 16 | -23 | 35.7 | 143.75 |
| Frank Thomas | 6820 | 46 | -26 | 16.7 | 56.52 |
| Lance Parrish | 7363 | 42 | -20 | 13.9 | 47.62 |
| Eric Karros | 7024 | 63 | -25 | 9.0 | 39.68 |
| George Scott | 8185 | 118 | -43 | 30.9 | 36.44 |
| Lou Piniella | 6266 | 58 | -21 | 11.4 | 36.21 |
| Mike Lowell | 6498 | 67 | -21 | 29.1 | 31.34 |
| Ivan Rodriguez | 10133 | 71 | -21 | 67.7 | 29.58 |
| Dick Stuart | 4320 | 68 | -20 | 6.1 | 29.41 |
| Bob Bailey | 6951 | 98 | -25 | 24.9 | 25.51 |
| Lee May | 8145 | 118 | -26 | 22.7 | 22.03 |
| Cecil Fielder | 5893 | 123 | -26 | 15.3 | 21.14 |
| Willie Horton | 7976 | 151 | -28 | 24.3 | 18.54 |
| Jim Rice | 8959 | 279 | -46 | 41.5 | 16.49 |
| Julio Franco | 9632 | 169 | -27 | 40.6 | 15.98 |
| Paul Konerko | 6831 | 194 | -30 | 22.1 | 15.46 |
| Joe Torre | 8738 | 300 | -43 | 55.6 | 14.33 |
| Tony Perez | 10746 | 255 | -36 | 50.5 | 14.12 |
| Cal Ripken | 12746 | 181 | -24 | 89.9 | 13.26 |
| Ted Simmons | 9574 | 208 | -27 | 50.4 | 12.98 |
| Rico Carty | 6261 | 251 | -30 | 31.4 | 11.95 |
| Joe Adcock | 7217 | 204 | -24 | 34.2 | 11.76 |
14 comments
|
2 recs |
Do you like this story?
Comments
this is great work, adam.
Robinson’s made me cringe a little, though.
The baseball season doesn't have to end! Create your own players, coach your own teams, and join your friends in THE premier baseball MMO. Two Out Rally opens October 25th!
Two Out Rally, BASEBALL MMORPG | Facebook | @2OutRally
Those figures for Robinson are way out of line with...
…almost anyone else’s. I come up with 80-some RAA without adjusting for park, Fangraphs does as well. And he played in mostly pitcher’s parks in his career. I can’t explain +20 at all.
I'm looking at rWAR vs fWAR right now...
And Brooks’ difference between the two is a staggering 25.5 wins.
We may have a problem.
On Twitter: @baseballtwit
Well, lesse.
If you add in DP/ROE runs, rWAR says Robinson was -17 RAA in his career with the bat. Fangraphs says he was 133.3. That gives you 15 wins right there. Defense is the same (both are using TotalZone).
Hey a Gus Triandos sighting
Herc’s man-crush on “The Wire”.
Awesome visuals, you get a rec.
by James Kannengieser on Nov 26, 2010 10:56 AM EST reply actions
I don't understand this.
If you were to say that Brooks Robinson would have had +55 runs of offensive value without grounding into any double plays, then the net result with GIDP’s would be a reduction of 64%.
by nathaniel dawson on Nov 27, 2010 10:27 AM EST reply actions
Okay, I see what you're doing now
Those batting runs above average didn’t include GIDP’s. Once you include them, I see what you’re saying. Like rcbuss, I think the percentages are misleading because you’re comparing offensive production to the average, rather than a replacement baseline.
by nathaniel dawson on Nov 28, 2010 4:43 PM EST up reply actions
I’d imagine these numbers wouldn’t look so bad if they were compared to Batting Runs Above Replacement, or total Batting Runs (wouldn’t part of your offensive value be from being better than replacement, as well as being better than average?).
I think you could do either
I do like using above average when talking about things from a historical perspective. Replacement level is awesome when you’re looking at things like filling a roster, producing value against a contract, and—as the name replies—replacing a player. I focus more historical career value, where I want to see how much better a guy was “than everyone else”, or the average.
On Twitter: @baseballtwit
Interesting, but...
Not exactly fair to the players involved, unless we’re claiming GB% with a runner at first is some kind of skill beyond just plain GB%?
Over the Monster -- SB Nation's Resident Red Sox Site
USG
According to Rally, here's where the GIDP numbers come from:
GIDP – Runs from hitting into double plays, above or below average considering a player’s GIDP opportunities. A GIDP opportunity is a groundball fielded by an infielder with less than two out and a runner on first base.
On Twitter: @baseballtwit
It does seem that if you're going to penalize a high GB guy for hitting behind high OBP guys...
you should reward him for positive contributions based on base-out states, too. Good call. Or maybe the ol’ RE24/LI_RE24 approach (like WPA/LI)?

by 





































