The Three Worst Double Plays Ever
…or at least since 1974, since that’s how far the play-by-play database goes back at the moment.
All non-masochistic fans hate watching a hitter from their team ground into a double play. It is almost always (that’s right, almost, as we’ll see in a post later this week) devastating for your team’s chances. In terms of linear weights, the average double play in modern baseball is about .37 runs worse than a normal out because it costs another out and takes a runner off of the bases.
Of course, the actual effect of a double play depends on the game situation in which it occurs, it’s place in the story. While Win Probability Added (WPA) isn’t a good way to value individual players, it is a good “story stat,” as it gives a quantitative sense of the ebbs and flows of the game play-by-play by seeing what the teams chances of winning before and after each event are. So let’s take a look at the worst (from the perspective of the hitter’s team) three double plays (just grounded into double plays, as things like lining into a double play are a different sort of beast) according to WPA since 1974.
We have tie for third:
3. Russ Johnson, -.573876 WPA, May 13, 2000, Reds (8) versus Astros (7). It was a seesaw battle, with the lead going back and forth. Going in to the ninth, the Astros were leading 7-6, but then Ken Griffey, Jr. hit a two run homer off of Billy Wagner to put the Reds up 8-7 (+.655 WPA). The bottom of the inning featured slightly less legendary players in its crucial moment. Scott Williamson, who bounced between the rotation and the ‘pen that season, came into close for Cincinnati. After getting 1996 NL MVP Ken Caminiti to ground out, Williamson walked Roger Cedeno, then gave up a hit to Craig Biggio to put runners on second and third with one out. The Reds decided to walked Jefef Bagwell to set up the double play — a smart move, as even without taking into account Bagwell’s ability and the fact that the next spot up was Wagner’s, the Astros WPA went down by .020. Up came the pinch-hitter: veteran utility man Russ Johnson. Yes, it was the Johnson-Williamson match up that people had been waiting for all day. When Johnson’s plate appearance started, Houston’s win expectancy stood at 57.4%. He made two outs, and the game went to the Reds. You win this round, Williamson!
3. Richard Hidalgo, -.573876 WPA, May 27, 2000, Braves (6) versus Astros (5). Well, I think we’re getting at part of the reason the Astros missed their Pythagorean Expectation by 9 games in 2000… For those of you who don’t remember Richard Hidalgo, he had a couple of monster seasons scattered among a few mediocre ones with the Astros, and the Mets and the Rangers both took chances on him hoping to recapture the magic. I remember he was in very high demand at my fantasy league’s auction prior to the 2005 season: “put him in the Rangers’ park and he’ll hit 30 jacks.” Well, the first half came true, he did get the park, and put up .221/.289/.416 (.303 wOBA) in 88 games for the 2005 Rangers. He never played in the majors again.
Whatever else happened, his 2003 (6.3 WAR) and his 2000 (7.5 WAR) seasons with the Astros were awesome. In 2000 Jeff Bagwell was still in his prime, but it was Hidalgo who led all Astros with 7.5 WAR. TotalZone loved his defense (+13, mostly in center field), but his bat was incredible: 149 wRC+, .314/.391/.636, and 44 home runs. Given his prior performance, it was somewhat like Jose-Bautista-in-2010, except the follow-up didn’t happen for three years. I’ll stop there for the sake of Blue Jays fans (for the record, I don’t think they are very similar players).
Just two weeks after Russ Johnson’s Big Play, the Astros found themselves in a similar situation. While Hidalgo was a flyball hitter, on this day in 2000 (a down season sandwiched by two division championships for the franchise) he didn’t manage to get it nearly high enough when it counted. The Astros had spent the day falling behind and coming back to within one run. In the bottom of the ninth, everyone’s favorite closer of the late90s/early 00s, John Rocker, game on. He promptly walked two batters, then threw a wild pitch to advance them to second and third. He then was ordered to give Ken Caminiti (what a wonderful cast of characters) a free pass and was removed for Kerry Ligtenberg. Within none out, Ligtenberg got Daryle Ward (wow) to fly out, but not deep enough to score a runner. With the bases loaded (sensing the theme?) one out, and the Astros down by one, up came Richard Hidalgo, the man who would finish the season with a .323 ISO. Well, given the title of this piece, you know what happened next. He ended this particular game much like he ended his career.
1. Eric Young, -.5781 WPA, August 27, 1996, Reds (4) vs. Rockies (2). Near the beginning of the expansion Rockies’ impressive (?) run of mediocrity in the mid-1990, they were locked into this fierce struggle with the equally mediocre Cincinnati Reds (the Rockies finished 1996 at 83-79, the Reds at 81-81). Ah 1996, when a .369 wOBA (.324/.393/.421) was only slightly above league average (101 wRC+) when adjusting for Coors Field. Young was decent offensively player for a second baseman over the years (career 97 wRC+). Young was the Rockies primary lead-off man in 1996, and scored 113 runs.
This play probably isn’t memorialized over the Young family fireplace. The Rockies got down 0-4 after an Eric Davis (!) three-run homer in the fourth. However, a Dante Bichette home run in the sixth brought them back to within one run (3-4). In bottom of the ninth, the legendary Jeff Brantley came in to finish off the Rockies. He promptly allowed a single to Vinny Castilla and walked John Vander Wal. Walt Weiss bunted the runners over (a rare case of a “smart” bunt actually being smart: +.018 WPA). Brantley’s then walked Eric Anthony to load the bases, which is surprisingly listed as a negative WPA play for the Rockies. It makes a little more sense when you take into account the base, out, score, and inning, however, not to mention, the possibility of what happened next. With the Rockies down by one, the bases loaded, and one out, WPA had their chances of winning at 57.8%. Young, their speedy leadoff man who would end the season with almost a .400 on-base percentage, was at the plate… and he sent a grounder to the shortstop for a game-ending double play.
Not exactly Casey at the Bat, but devastating nonetheless.
As hinted at the beginning of the piece, check back later this week for the three best ground-into-double-plays of all time (since 1974).
Matt Klaassen reads and writes obituaries in the Greater Toronto Area. If you can't get enough of him, follow him on Twitter.
Jeff Brantley was so clutch!
Eric Davis (!)
Before there was Ichiro!, there was Eric Davis (!)
Why is WPA greater for 1 than for the other 2? They were all in bottom 9, bases loaded, 1 out, down by 1. Does WPA adjust for run environment?
Bingo
The linear weights might have been updated for season.
Looked at from that perspective, the premise of this article is moronic. Obviously the biggest negative WPA in a season (on a DP) comes from down by 1, bases loaded, 1 out DPs. So then this article asks, in which years do the linear weights make that play even worse than in other years.
It has nothing to do with what is going on in that game, just has something to do with how the batter vs pitcher averages were shaking out that year.
You will notice that there are in the middle of the ‘roid era, so presumably that is when having bases loaded with one out had the highest value. Because of this, there were probably lots of team mates who had a hard time controlling their rage afterwords.
Well, as Mike S. says below, I suppose it’s interesting that apparently no since 1974 has lined into a double play with men on 2d and 3d, down 1, bottom 9th.
You are correct about the event weights, but I’m not sure how that makes this any more “moronic” than (analogously) two hitters in different seasons having the exact same line but having different batting runs above average because of different seasonal run environments and park adjustments.
Thanks for the comment.
If there’s a League Championship version of WPA, i.e. how does a given play affect the chances of a team winning their Division or the Pennant or whatever, then this discussion might be more interesting, I think. A critical GDP to end a team’s season would be the sort of thing I would classify as the “worst ever”.
@Matt Klassen, the thing is that nobody (who is qualified) hits only homers. Here, we know the play (bottom of the ninth…)
The linear weights assume that they are facing a league average pitcher, but one might have been facing Mo and another might have been facing some wash-up who was the only pitcher left in the second game of a double header. The latter case would be a worse GDP than the first, even if pitchers *in general* were better in the year the first one occurred.
The point is you are adjusting a single game for the year it is in. This makes sense for a year’s worth of performances (as your example points out), but not for a single game.
… and this is true for a lot of other stats, too. LWPA gives a lot of context, but not all of it. As is often said, WPA is a “story stat.” I don’t think that makes it irrelevant or useless for “story” posts like this. Maybe not even “moronic.”
@Matt Klaassen, in this case the stories are all the same–bottom of the last inning, down by one, bases loaded.
I think you could have looked at all such situations and picked your favorites–that would have been interesting. But I don’t get why I care what year it happened in or why that trumps the exact pitcher the batters were facing.
The implication of linear weights by year is that Pedro Martinez was a pretty batter friendly pitcher in 2000 while some AAAA player was very hard to face in 2010–it just doesn’t make sense.
To each his or her own, I suppose. I guess I would think I’m looking less for individual players, and more for situations about which to tell stories involving players. Thanks for the discussion.
I find it hard to believe that Wilson Valdez wasn’t the GIDP protagonist of at least one of these scenarios.
Man, I forgot out crazy that 2000 season from Hidalgo was. 40+ homers and doubles.
Well, there were those two double plays in Game 2 of the Phillies-Yankees 2009 World Series. Those were worse.
I callz ’em as I seez ’em.
By the way, SI.com has never looked so good. Love the new orange font. Keep up the good work Nat Klasson!
You have two examples of walks being good for the defense to set up the double play – which they got. Therefore, it is possible to have a larger negative. If a hitter lined into a double play with runners on second and third with one out to end a one run game that would presumable give a slightly more negative WPA. I’m sort of surprised that hasn’t happened since 1974.
He said he’s excluding those, since those aren’t really fair.
I really should learn to read one of these days. Sorry.
No worries! I think that would also be an interesting series of dp’s to look at.
Technically shouldn’t the tie be for second?
i was about to say that.
Yes
It depends on which direction you are counting from. I prefer to count from +? down to 1.
Hmmm… that did not publish correctly. The ‘?’ should have been an infinity symbol.
I used to do it that way, and got corrected for THAT. I’ve seen it both ways. In the Olympics (at least years ago), I remember there being a “tie for second” but those performers both got bronze…
it’s not surprising that an IBB to load the bases with 1 out in the bottom of the ninth increases WPA to the road (pitching team). this happened in 2 of the 3 examples in the list
In his 2000 and 2003 season it’s widely speculated that Richard Hidalgo was all hopped up on Mountain Dew.
When you say the batters just grounded out, are you saying for certain they just hit a routine dribbler to an infielder or did any of them make a great defensive play to turn the dp?
The way I’ve done this (well, the way WPA “does” it) that doesn’t matter — it takes account of the game stats of the play, not the granular details of of it.
Thanks for the comment.
(what a wonderful cast of characters)
Wow… Daryle Ward, John Rocker, Richard Hidalgo, Kerry Ligtenberg, and Ken Caminiti mentioned. Looking at the play log, I also see a Quilvio Veras, Brian Jordan, John Burkett, Reggie Sanders, and something called a Wally Joyner. That truly is a remarkable cast of characters.