Judgmental Stats: The Win
|Joe Posnanski||Feb 13, 2017|
A few years ago, I wrote something about judgmental baseball stats. I've had numerous people email and ask me to revisit the concept. So, why not? Let's revisit with a little Judgmental Stats Series.
Let me say up front: I'm not interested in actually CHANGING the traditional baseball stats. They've been around so long, and have so much history, that to change them and make them "more logical" would probably do more harm than good. Cy Young's 511 wins, Ty Cobb's .366 average, Rickey Henderson's 130 stolen bases, these are ingrained in the game. We have in our mind what a 20-game winner, a .300 hitter, a sub 3.00 ERA feels like, and there's no real reason to mess with that. Even errors, a generally dumb concept, have such a place in the game's story that doing away with them would leave a void.
But, even so, it's still worth pointing out that, unlike what I suspect many baseball fans think, most historic baseball stats are not counting numbers or calculated with simple math. They are permeated with shaky math and ethical judgments about what players DESERVE to have credited and debited to their baseball accounts. We like to quote these stats like they are simple facts. They are not. If you see 10 oranges in a bucket it is a fact to say "There are 10 oranges in a bucket."
But a baseball counting of that would say, "Well, one of those oranges is too small to qualify, and one of them is not exactly orange -- more like vermillion -- so that only counts as half an orange, and two of the oranges have seeds which make them different ..."
The Pitcher Win is the most obvious of the judgmental stats. The Win has been talked about, complained about, rebelled against for years now -- Brian Kenny, Keith Law, many others have written about how ridiculous it is to judge pitchers by a statistic that is so team oriented AND so weird. There's no point going through the complicated rules revolving around the pitcher win.*
*"Winning and Losing Pitcher" requires THREE PAGES in the Baseball Rulebook, and this includes this baffling "comment" on Rule 9-17.
"It is the intent of Rule 9.17(b) (Rule 10.17(b)) that a relief pitcher pitch at least one complete inning or pitch when a crucial out is made, within the context of the game (including the score), in order to be credited as the winning pitcher. If the first relief pitcher pitches effectively, the official scorer should not presumptively credit that pitcher with the win, because the rule requires that the win be credited to the pitcher who was the most effective, and a subsequent relief pitcher may have been most effective. The official scorer, in determining which relief pitcher was the most effective, should consider the number of runs, earned runs and base runners given up by each relief pitcher and the context of the game at the time of each relief pitcher’s appearance. If two or more relief pitchers were similarly effective, the official scorer should give the presumption to the earlier pitcher as the winning pitcher."
Have fun with that one, kids. Last year, Baltimore relief pitcher Brad Brach won 10 games. Now, you tell me: How did Brad Brach win 10 games? Glad you asked:
-- On April 10, he came into a game with two outs in the fifth. Baltimore was leading 4-3, but Tampa Bay had runners on first and third. He got Brad Miller on a deep fly ball to end the inning. Brach pitched a scoreless sixth, the Orioles held on to their lead and Brach got the W.
-- Next day, April 11, he came into a tie game in the eighth inning against Boston. He pitched a scoreless inning. The Orioles scored three runs in the ninth on Chris Davis' home run. Brach now 2-0.
-- On April 29, he came into the game in the seventh with Baltimore leading Chicago 3-2. He gave up a game-tying homer to Brad Lawrie before getting out of the inning. The Orioles scored three runs in the bottom of the inning on Nolan Reimold's three-run homer. Brach now 3-0.
-- On May 21, Brach came into a game against the Angels with two outs in the seventh. He got Albert Pujols to ground out and pitched a scoreless eighth. The Orles scored three runs in the top of the ninth on Matt Wieters home run. Brach now 4-0.
-- On June 1, Brach came into a game against Boston in the sixth with the score tied. He got out of the inning when the Xander Boegarts single he gave up did not score Dustin Pedroia from second. The Orioles scored two runs in the bottom of the inning. Brach pitched a scoreless seventh. Brach now 5-0.
-- July 9, Angels again, tie game, Brach pitches a scoreless eighth. Orioles score the winning run in the bottom of the inning. Brach now 6-1.
-- On August 14, at San Francisco, Brach pitches a scoreless eighth with the Giants ahead 7-5. The Orioles score three runs in the ninth on a Jonathan Schoop home run. Brach now 7-1.
-- On August 30, Brach finishes off the seventh with the score tied against Toronto, and he pitches a scoreless eighth. Matt Wieters hits a two-run homer in the bottom of the inning and Brach is 8-2.
-- On September 16, Brach pitches a scoreless eighth with Tampa Bay ahead 4-3. Orioles score two in the bottom of the eighth. Brach is 9-3.
-- Two days later, again against Tampa Bay, Brach pitched a scoreless eighth in a tie game, Mark Trumbo homers in the bottom of the inning and that's Brach's 10th victory.
Now, I went through all 10 of those to make the point: Are any of those wins what you think about when you consider the pitcher win? Brad Brach got 10 wins, more than numerous very good starters like Julio Teheran, by pitching an inning or so at the right time. That is and always has been ridiculous.
Then, even people who defend the pitcher win as a stat generally believe that relief pitcher wins are useless things. They will say, "Yeah, I never look at relief pitcher wins -- but starter wins are still useful." OK, well, are you ready for this? I'm going to give you a statistic that will blow your mind -- you ready for this?
Here you go:
Number of games in 2016 where a pitcher got a win while pitching two innings or less: 724.
OK, you got that? That's 724. Now get ready for this.
Number of games in 2016 where a pitcher got a win while pitching seven innings or more: 689.
Bchloo! Yeah, that's the sound of my mind getting blown. It was more common in 2016 for pitchers to get wins by pitching a measly inning or two at the right time then it was for starters to get a win going at least seven innings. We are in a different time, a time when the win is outdated. Like Kevin Bacon so wisely said in Footloose, "And there was a time for this law, but not anymore."
In the past, I've suggested -- and I'll suggest it here again -- that only starters should get wins and losses. When I say this, people immediately point out that it's stupid to give a starter credit for a win or loss if he doesn't actually play any role in when the game is won or lost.
To which I reply: YES! IT IS STUPID! OF COURSE IT IS STUPID!
BECAUSE GIVING PITCHERS "Wins" IS STUPID.
But, at least if you do it this way, you get a real, live counting stat. You are actually counting the team's record when that pitcher starts. That's a real thing, .not some bizarre judgmental, outdated, dartboard-in-the-dark sort of stat.
Here were your TRUE win-loss leaders in 2016:
Rick Porcello, 25-8
Jon Lester 24-8
J.A. Happ 24-8
Cole Hamels 24-8
Max Scherzer 24-10
Masahiro Tanaka 23-8
Johnny Cueto 23-9
Chris Tillman 22-8
Adam Wainwright 22-11
Chad Bettis 21-11
Collin McHugh 21-12
By the way, when using career true win stats, Jack Morris moves up to 302 wins (Hall of Fame!). Six pitchers since Deadball have more than true 400 wins (Clemens, Maddux, Sutton, Ryan, Carlton, Spahn).
And it's actually fascinating to look at true win percentages ... here's a question you probably have not thought much about: Is Tim Hudson a Hall of Famer? Unless you are specifically a Tim Hudson fan, your immediate reaction is probably: "No." Or "What?" Hudson won 222 games with a 3.49 ERA, barely 2,000 strikeouts, never won the Cy Young or even got a first place Cy Young vote. Even if you look at advanced stats, his 57.2 WAR and 48.5 JAWS fall below Hall of Fame median numbers.
BUT what if we had counted wins this true counting way rather than the old way.
If we did: Hudson's career record would be 297-182, which is really impressive. First of all: Did you have ANY IDEA that Tim Hudson started 297 winning games? And that .624 win percentage would rank in the Top 10 since Deadball. The only pitchers with that many wins and that high a winning percentage are Whitey Ford, Lefty Grove and Jim Palmer. I'm not saying this makes him a Hall of Famer. I am saying that if we counted wins in this truer and more logical way, his Hall of Fame case would certainly resonate more.*
*And how much better would Mike Mussina look with a 325-210 record?
Now, does this true win-loss record tell you anything more than the old win-loss record? Well, in a small way, yes: At least the true win-loss tells you something concrete, what the team's record was when that pitcher started. This is how we count quarterback win-loss records, coach win-loss records, manager win-loss records, etc. Nobody even knows what the old win-loss record is supposed to count, exactly.
But in a larger way, no, they're probably about the same level of meaningful, blurring the pitcher's individual performance with the team's offensive performance with general timing in a way that gives us a number that offers a badly obstructed view of the pitcher. There's no way around that when you want to credit pitchers with wins or losses.
But, I would say, hey, if you want pitcher wins, drop the judgment and really count pitcher wins.