Redefining the Win

A while back now, I saw a panel at the MIT Sloan Sports Analytics Conference featuring Bill James, John Walsh, John Thorn, John Dewan, and Dean Oliver. The topic on hand was re-working box scores to more accurately reflect reality while at the same maintaining the casual fan-friendliness of the existing boxscore. Specifically, we now have the ability to actually change the win statistic for pitchers into something meaningful, that can maybe help bridge the paradigm gap between sabr-friendly and old-school fans.

The ability to evolve any given statistic that the panel referred to has to do with the vast and ever-widening social facets of the internet. Basically, Bill James and the rest of the panel argued that if we were really serious about re-defining the pitcher win, we could viral market a replacement for it. Because of the democratic nature of the internet, any such evolution would naturally undergo many different mutations until it became something that everyone could agree upon.

The idea of improving the win stat might seem trivial, but I really don't think it is. The sabrmetrics-based paradigm gap I referenced is perhaps the biggest issue facing baseball writers and broadcasters today, as the more stat-headed among us among us end up rolling our eyes, shouting at the television or computer, or just tuning out completely when the writers that either can't or don't care to understand the theories advanced statistics expose. Meanwhile, the less stat-headed among us inevitability do the exact same thing when some nerd starts talking about Bayesian Expressions and Markov Models and all kinds of algorithms. There needs to be a bridge here.

So, consider this my initial concept of an improved pitcher win, and please take this ball and roll with it in whichever direction you think it ought to go next, even if that is perhaps backwards in my little opinion.

The current rules governing pitcher wins are as follows, according to rule 10.17 of MLB's official rules:

10.17 WINNING AND LOSING PITCHER

(a) The official scorer shall credit as the winning pitcher that pitcher whose team

assumes a lead while such pitcher is in the game, or during the inning on offense in

which such pitcher is removed from the game, and does not relinquish such lead,

unless

(1) such pitcher is a starting pitcher and Rule 10.17(b) applies; or

(2) Rule 10.17(c) applies.

Rule 10.17(a) Comment: Whenever the score is tied, the game becomes a new contest insofar as

the winning pitcher is concerned. Once the opposing team assumes the lead, all pitchers who have

pitched up to that point and have been replaced are excluded from being credited with the victory. If the

pitcher against whose pitching the opposing team gained the lead continues to pitch until his team

regains the lead, which it holds to the finish of the game, that pitcher shall be the winning pitcher.

(b) If the pitcher whose team assumes a lead while such pitcher is in the game, or

during the inning on offense in which such pitcher is removed from the game, and

does not relinquish such lead, is a starting pitcher who has not completed

(1) five innings of a game that lasts six or more innings on defense, or

(2) four innings of a game that lasts five innings on defense,

then the official scorer shall credit as the winning pitcher the relief pitcher, if there

is only one relief pitcher, or the relief pitcher who, in the official scorer’s judgment

was the most effective, if there is more than one relief pitcher.

Rule 10.17(b) Comment: It is the intent of 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.

(c) The official scorer shall not credit as the winning pitcher a relief pitcher who is

ineffective in a brief appearance, when at least one succeeding relief pitcher pitches

effectively in helping his team maintain its lead. In such a case, the official scorer

shall credit as the winning pitcher the succeeding relief pitcher who was most

effective, in the judgment of the official scorer.

Rule 10.17(c) Comment: The official scorer generally should, but is not required to, consider the

appearance of a relief pitcher to be ineffective and brief if such relief pitcher pitches less than one inning

and allows two or more earned runs to score (even if such runs are charged to a previous pitcher).

Rule 10.17(b) Comment provides guidance on choosing the winning pitcher from among several succeeding

relief pitchers.

(d) A losing pitcher is a pitcher who is responsible for the run that gives the winning

team a lead that the winning team does not relinquish.

Rule 10.17(d) Comment: Whenever the score is tied, the game becomes a new contest insofar as

the losing pitcher is concerned.

(e) A league may designate a non-championship game (for example, the Major League

All-Star Game) for which Rules 10.17(a)(1) and 10.17(b) do not apply. In such

games, the official scorer shall credit as the winning pitcher that pitcher whose team

assumes a lead while such pitcher is in the game, or during the inning on offense in

which such pitcher is removed from the game, and does not relinquish such lead,

unless such pitcher is knocked out after the winning team has attained a

commanding lead and the official scorer concludes that a subsequent pitcher is

entitled to credit as the winning pitcher.

In a word...gross. Okay, a starting pitcher must go at least 5 full innings (or 4 innings if the game is called after 5 innings) and he must leave the game with his team ahead and his team must never stop being ahead once he has left the game. Relief pitchers must pitch "effectively" and have pitched the most recent half-inning before their team took the lead for good in the game. And none of this matter if the league decides to change the rules on a given day.

First of all let's forget about the pitcher loss; that's another show.

Consider what the win means from a practical point of view. When a fan or analyst cites a pitcher's win total, he is arguing that that pitcher held the primary responsibility for those games being won. Roy Halladay is the Phillies' ace because when he goes out there and pitches, he makes sure the Phillies win way more often than not. The Orioles' pitching staff is terrible because they don't have anybody who can reliably go out there and make sure the Orioles win on that day. The credit for the team's victory is going to the pitcher who gets the win.

When does a pitcher really deserve that credit? I propose a slightly modified version of the quality start to be our guideline. To earn a win, a pitcher must (a) pitch for the winning team obviously, (b) pitch at least five innings*, and (c) give up at most: 3 runs (earned or unearned) for 5 or 6 inning outings, 4 runs for 7, 8, or 9 inning outings. Any pitcher that pitches more than 9 innings in a winning effort automatically gets the win, regardless of the runs scored during the game.

*I'll assume we're only talking about full nine inning games (or games with extra innings)

We don't need a team's total pitcher wins to equal their total team wins, because there are always games that are won by the offense. The typical 10-9 slugfest obviously does not belong on any pitcher's resume. We also do not need the bulk of relievers to ever get a win. Does anybody even look at Jim Johnson's win total? What would it possibly tell you even if you did (he won 6 games in 2011, by the way)? This amended system cancels out those extreme yet common problems, as well as awarding the starter even if the bullpen blows the lead.

This is just a start, but it is an improvement. And here are how the Oriole pitchers would have been merited in 2011 under these adjusted rules for pitcher wins (with a hearty thanks to baseball-reference):

What else could be done to improve the win?