The Gods and Dogs of Garbage Timeby Chris Jaffe
February 18, 2008
For as long as I can remember, clutch hitting has been one of the great debates among statheads. Does it exist? Can it be proven? If so, who does the best in clutch situations? And so on. Me? Well, I've stayed the hell out of that whole mess. I think there's something to it, but I can't prove it.
These discussions always look at how players do in the most important situations—with runners on, runners in scoring position, and of course the all-important close and late at-bats. That's how you dole out labels of clutch and choker.
It occurred to me that the entire debate can be flipped around. How about looking at how guys do when it doesn't matter at all? Instead, tabulate career numbers for a slew of players and see who really stepped up when it didn't matter because the ballgame had been decided, and who slacked off taking it easy.
The Set Up
Due to the wonder and glory that is Baseball-Reference, anyone can check on this. This year the website included player splits. One bunch tells how any player from 1957 onward did ranging from when the score was tied, to when there was more than a four-run difference.
That last one, performance in blowouts, is the key to this study. Games where a team is up or down by five or more runs are rarely competitive. It's just garbage time. A player who really performs better in garbage time—a Garbage Time God, frankly isn't as valuable to his team as his stat line indicates. Whether it be a result of how a player responds to pressure or random happenstance, he did less to help his teams win games than the back of his baseball card indicates.
By and large this study consciously intends to skirt the big debate in clutch hitting: is it skill or is it luck? Instead, my main focus will be on value. We can all agree that hitting a homer when your team is down by 10 runs is fundamentally less valuable than doing it in a tie game.
Still, while the focus is on value, I think these results, at least with the Garbage Time Dogs, are less likely to be purely a product of luck than traditional clutch hitting examinations. While it's an open question whether or not a player can really step it up a notch or two for the big time situations, I can't imagine anyone would think that a player could lower it a notch or two. If a player doesn't think he has to concentrate as hard, his results may diminish.
Alternately, a Garbage Time God might be more mentally comfortable in the less pressure-filled environment. I can't prove anything, but for me that idea passes the smell test anyway.
Plan of action
I found every player with at least 1,000 games played from 1957 to 2007, and compare their overall numbers to their numbers when the game is out of reach and see who really picked it up when it didn't matter and who fell back some.
Actually, one minor tweak to that. I'm only going to look at guys whose entire careers are since 1957, so I'm sure the numbers I'm using are their full info. I don't know if anyone noticeably got better or worse at this over time, but just in case I'll only use full sample sizes.
There's over 700 players in my sample size. Let's start by comparing OPS. For what it's worth, these guys have a combined career OPS of 764 in over 4.25 million plate appearances. Over a half-million of these plate appearances came in blowouts, where they had an aggregate OPS of 772. Teams will rarely have their top relievers in games in Garbage Time, and you'll definitely have your share of mop-up men so their numbers should improve.
The batters closest to the average are Tony Phillips (763/768) and Amos Otis (768/773).
But now for the main event: the Gods and Dogs. My formula is (Normal OPS/Garbage OPS). And remember—you want to be a Garbage Time Dog. These guys provided better value. Here are the 20 greatest Dogs since 1957:
Name allOPS GarbOPS Gar. Score Gates Brown 750 570 1.316 Jerry Adair 639 522 1.224 Willie McCovey 889 743 1.197 John Lowenstein 740 621 1.192 Ron Hassey 722 609 1.186 Gary Sutherland 599 513 1.168 Wayne Garrett 691 593 1.165 Jerry Remy 655 565 1.159 Tommy Helms 642 558 1.151 Mark McLemore 690 601 1.148 Phil Garner 712 622 1.145 Jim Dwyer 751 661 1.136 Dean Palmer 796 701 1.136 Leon Durham 831 736 1.129 Mike Devereaux 709 628 1.129 Wally Backman 688 610 1.128 Hubie Brooks 718 637 1.127 Buck Martinez 627 557 1.126 Frank Howard 851 757 1.124 Tino Martinez 815 725 1.124
Tino Martinez just makes it, which is fitting given his reputation as a Clutch God. Since he was at his worst when it mattered least, his career was more valuable that his stat lines appear, aiding his clutch reputation.
It's perfect that Gates Brown vaults to the summit. The man was a pinch hitter who was supposed to deliver in the clutch, not when the game was out of reach. And he is blowing the field away! In part it's because he's got fewer plate appearances as a one-swing-a-game guy, but his score is amazing. Here's what it looks like if you take his total numbers, Garbage Time figures, and non-GT appearances (all minus garbage) and normalize them all to 600 plate appearances (with his actual rate stats, so they won't perfectly match his prorated at-bats, hits, etc):
PA AB H 2B 3B HR W K AVG OBP SLG Total 600 533 137 18 5 20 57 65 0.257 0.330 0.420 Garbage 600 546 115 14 4 8 44 62 0.211 0.275 0.295 Not G. 600 532 140 19 5 22 59 65 0.264 0.337 0.443
Also notable is the only Hall of Famer on the list: Willie McCovey. He had two to three times as many plate appearances as the guys around him. Amazingly, among the other Hall of Famers in this pile, the second-best (Paul Molitor) scores at a mere 1.049 (817 regular OPS, 779 in Garbage Time). McCovey's murdering his fellow immortals. Here's the pride and joy of the Bay Area prorated to 600 plate appearances:
PA AB H 2B 3B HR W K AVG OBP SLG Total 600 507 137 22 3 32 83 96 0.27 0.375 0.515 Garbage 600 538 130 19 2 25 57 105 0.242 0.320 0.423 Not G. 600 504 138 22 3 33 86 95 0.273 0.381 0.524
With many of these guys, the issue might just be sample size, but that's less likely with McCovey, who had two seasons in his Garbage Time splits alone. Look at those lines and what do you see? It's a guy who isn't taking his time. His walk rate's plunging and as he's swinging at more bad pitches, he falls behind and can do less damage. Maybe he was less interested when the game meant less.
Them's the dogs. Now how about those gods?
Name allOPS GarbOPS Score Davey Johnson 744 917 0.811 Sid Bream 756 929 0.814 Larry Brown 613 734 0.835 Tony Kubek 667 798 0.836 Donn Clendenon 770 897 0.858 Bill Almon 648 754 0.859 Moises Alou 886 1028 0.862 Jerry Grote 642 743 0.864 Charles Johnson 763 879 0.868 Tony Oliva 829 955 0.868 Wes Parker 726 835 0.869 Milt Thompson 707 812 0.871 Joe Pepitone 733 841 0.872 Tommy Davis 734 842 0.872 Gary Carter 774 885 0.875 Leon Wagner 795 909 0.875 Jody Davis 710 808 0.879 Roy Smalley Jr. 740 842 0.879 John Mayberry 799 909 0.879 Bob Horner 839 952 0.881
Well, speaking as a Cubs fan, it makes sense to see Moises Alou on that list. He always seemed a little too sensitive. Here he's shining when there's the least pressure. Gary Carter's obviously the best player there, but I think the most important discovery on this list, though, has got to be Tony Oliva. After all, he has a viable shot to make it into Cooperstown as a Veterans Committee pick (assuming that committee ever deems anyone else worthy ever again). Here's Oliva at 600 plate appearances:
PA AB H 2B 3B HR W K AVG OBP SLG Total 600 556 169 29 4 19 39 56 0.304 0.353 0.476 Garbage 600 558 159 26 5 33 35 55 0.334 0.378 0.577 Not G. 600 549 165 29 4 18 40 56 0.301 0.350 0.464
Oliva had just enough of that to be a reasonable borderline candidate. That's precisely why this is so damaging. If you're only a borderliner, then you can't be raising your game to such an incredible degree when it doesn't matter. Personally, I've supported his Cooperstown candidacy. Until now.
Maybe the most interesting line of the bunch, though, belongs to Sid Bream:
PA AB H 2B 3B HR W K AVG OBP SLG Total 600 528 139 32 2 15 60 76 0.264 0.336 0.420 Garbage 600 556 189 47 3 22 34 70 0.339 0.377 0.552 Not G. 600 525 133 31 2 14 63 77 0.254 0.331 0.395
I mean, what the hell? That can't be the same guy. It's like having Clark Kent around whenever Lex Luthor is pitching for the Yankees, while Superman only shows up to go bowling. He's crushing the ball for average and power when it doesn't mean a damn thing. Interestingly he declines in one area. Like McCovey, his walks drop considerably in Garbage Time.
My hunch? When the game got closer Bream tightened up, and took more pitches. He wasn't taking them to wait on one, but because he was a bit more passive the more crucial the situation. Then again, he also homered in three consecutive NLCS, so maybe I'm just blowin' smoke. (Looks back at his numbers). Naw, no way. The difference is too insane. If anything, his NLCS performance is the flukey one.
Long as I'm looking at these guys, I really ought to look at the most extreme one of the bunch, future managerial great Davey Johnson:
PA AB H 2B 3B HR W K AVG OBP SLG Total 600 527 137 27 2 15 61 74 0.261 0.340 0.404 Garbage 600 545 175 31 3 26 51 56 0.320 0.383 0.534 Not G. 600 525 133 26 2 14 63 76 0.254 0.335 0.388
Like Bream, he made better contact and did more with it while walking less. But good God Lord Almighty did he ever make better contact.
Tony Oliva makes me wonder about other possible enshrines. I took 50 of the players not currently in—both current and retired who I felt had the best shot at induction and sorted them. I'll admit my in/out line for this list was arbitrary, but with 60 guys, I'm sure I got everyone worth looking at. Here are the Top Dogs among those up for grabs:
Name allOPS GarbOPS Score Will Clark 881 795 1.108 Barry Bonds 1051 950 1.106 Vlad Guerrero 970 877 1.106 Roberto Alomar 814 752 1.082 Mark McGwire 982 915 1.073
Just missing is Captain Fist Pump himself, Derek Jeter, who has always had a very nice clutch reputation. His 850 lifetime OPS includes a 782 garbage time performance. Near the top of the list, however, Barry Bonds has often fought a reputation as a shirker when it mattered.
You know how valuable Barry Bonds' stats look when you see them on the page? Well, it turns out they underestimate how much he helped his teams. Sweet Jesus. I gotta see how he looks prorated at 600 plate appearances. Because it's Barry, I'll throw in an extra column—intentional walks.
PA AB H 2B 3B HR W IW K AVG OBP SLG Total 600 469 140 29 4 36 122 33 73 0.298 0.444 0.607 Garbage 600 495 139 23 3 35 95 14 78 0.281 0.399 0.551 Not G. 600 466 140 29 4 36 125 35 73 0.300 0.449 0.612
It's an across the board decline. Sure, intentional walks are down, but so are regular walks. You'd think if they pitched around him as often his natural walk rate would maintain itself. Since that's not the case that should mean they're pitching him to him more. While it looks like he homers as often, please note he has notably fewer at-bats in normal situations, yet actually has one more homer when it matters in the prorated lines. He has over a team season's worth of data for the Dog Days, so sample size really shouldn't be an issue.
Well, along with Oliva, who are the HoF-hopeful Garbage Time Gods?
Name allOPS GarbOPS Score Tony Oliva 829 955 0.868 Steve Garvey 775 863 0.898 Jeff Kent 861 951 0.905 Joe Carter 770 843 0.913 Bobby Grich 795 870 0.914
Just missing the list is sabermetric darling Jeff Bagwell (overall 948 OPS and a garbage time score of 1029), who has also had some rather well-known postseason problems. He's had a great career, but his overall value isn't as impressive as his numbers at Baseball-Reference make it appear. The shock is Steve Garvey. Maybe it's memories of Game Four of the 1984 NLCS, but I always considered him to be a clutch God, not a garbage man.
A few other players really got my attention when looking up this information. In particular, drastic changes in walk and whiff rates are important, I believe. It indicates a drastically different approach to the plate when the game no longer matters. Some had their whiff rate go up, indicating they were going all out. One such hitter is Dr. Stangeglove, Dick Stuart.
PA AB H 2B 3B HR W K AVG OBP SLG Total 600 550 145 22 4 31 41 132 0.264 0.316 0.489 Garbage 600 562 158 24 5 24 25 105 0.280 0.310 0.467 Not G. 600 548 143 21 4 32 44 136 0.262 0.317 0.492
The key is his walk and strikeout rates. In Garbage time, he made contact in almost four-fifths of his plate appearances. In other times, that only happened in 70% of plate appearances. Though his contact rate dropped, his isolated power dropped a whopping 38 points. When it mattered, he wanted to inflict the maximum damage with each swing. When it didn't matter, he was content to get a hit.
Another fun one is Jerry Remy.
PA AB H 2B 3B HR W K AVG OBP SLG Total 600 539 148 17 5 1 43 49 0.275 0.327 0.328 Garbage 600 543 125 13 3 0 53 57 0.230 0.299 0.266 Not G. 600 538 151 17 5 1 42 48 0.281 0.330 0.336
A lousy hitter in general, but when the game was all but over, he just seemed passive to the point of apathy. His walks and strikeouts were up, indicating more taken pitches. Plus it looks like he took weaker swings. Like I said at the top, one can always debate whether or not a player rises up to a higher level in the clutch, but it's always possible for him to take it down a gear in these situations. Remy? He was parked in neutral.
In terms of value, for some players career numbers obscure as much as they illuminate. Find a player who win shares or linear weights says was as good as Willie McCovey, and I'll guarantee the former Giant did more to help his teams with the stick.
And while this study mostly concerns itself with value, some of the splits are so wild that I have trouble believing that some guys either didn't try as hard when the game was out of reach or thrived in the lack of pressure. This has an implication for more standard clutch hitting ideas. It could be that a clutch hitter got his reputation at least in part because of how he took it easier when the game didn't demand his utmost.
History instructor by day, statnerd by night, Chris Jaffe leads one of the most exciting double lives imaginable; with the exception of every other double life possible to imagine. Despite his lack of comic-book-hero-worthiness, Chris enjoys farting around with this stuff. His new book, Evaluating Baseball's Managers is available for order. Chris welcomes responses to his articles via e-mail. Oh, and now he's on twitter.