Wins Aren't a Good Stat for Evaluating Pitchers

     The most well-known statistic in baseball for pitchers is the win-loss record.  It's not hard to see why.  The stat shows how many times a team wins or loses a game with that pitcher being on the mound.  On ESPN or any other sports programming network, it's the first stat used when previewing the pitchers of an upcoming baseball game, usually listed before a pitcher's ERA (earned run average).  Any casual observer to the game would typically assume that this stat would be the best indicator of how good a pitcher is, similarly to how a record is used to show how good any team is.  However, the stat is misleading, and should be noted, but not taken as the and all be all of stats when determining how good a pitcher is.

     The reasoning behind it isn't that difficult to understand.  A player wins a game if their team isn't winning when they begin pitching, their team is winning when they stop pitching, and then the team goes on to hold the lead and win the game.  The issue with this is that while the pitcher can do his best to allow as little runs as possible, he has no control over how the opposing pitcher performs, especially in the AL which uses the designated hitter, or DH.  So for example, let's say both pitchers play the full game.  One allows two runs, a great outing.  However, the other pitcher only allows one. Despite having a great game, the first pitcher takes a loss.  Despite how good a pitcher might be, if he's on a team that doesn't score, or is up against a good pitcher, or just gets unlucky, he just doesn't always come up with a record that suits his skill as a pitcher.   

     That was just an example.  However, similar situations take place all the time in baseball.  Here are just a few real life examples.  For people who aren't familiar with baseball, a pitcher's ERA is the number of earned runs he allows for every nine innings he pitches.  The average is a little over four, and an elite pitcher has an ERA of under three.

     First off is Jeff Samardzija.  In his time with the Cubs this season before being traded to the A's, he went 2-7.  Ouch.  The casual observer would write him off as a bad pitcher who should be begging for a job.  But then look at his ERA.  It was 2.83, or what I would call elite.    In Samardzija's first eight starts, he had an 0-3 record and a 1.62 ERA.  However, the Cubs had only given him 1.88 runs of run support per game, and that had left him winless.  

     Next is the case of Wily Peralta, a Brewers pitcher.  I actually saw this on SportsCenter a few days back and it's part of what inspired me to write this.  Peralta went 4-5 in April and May, with an ERA of 2.73.  However, since then he's gone 11-2, with only a 3.80 ERA, over a full point worse.  Why?  In April and May he received an average of 2.6 runs of support per game, where since he's received 6.1 runs per game.  He had nothing to do with that, but it's making his record appear like he's a star.

     However, the best example of why wins are deceiving comes from today, when David Price threw a one-hitter, allowed no earned runs, and lost.  How is that possible?  Well, a batter reached first on an error, and then went all the way home to score off the only hit that Price allowed, so the run wasn't earned.  The Rays lost 1-0, and since Price was pitching when the Rays' run was scored, he gets a loss, despite throwing a one-hitter.

     So now hopefully having established how inconsistent records can be, now the question becomes what is the best stat for evaluating pitchers.  Some people like stats like strikeouts or strikeouts over walks.  I disagree.  While I admit that throwing a high percentage of strikes is essential for good pitchers, it really doesn't make much difference how a pitcher gets his outs.  Therefore if a pitcher throws a strikeout for every out, but also has an ERA of 7.00, he isn't a good pitcher, no matter how many strikeouts he has.

     If you want to know what some actual MLB pitchers think is the best stat for evaluating pitchers, you should read this article as I found it to be pretty interesting.  http://www.fangraphs.com/blogs/players-view-the-best-stat-to-evaluate-pitchers/.  One thing I noticed is how 1/4 of the players asked answered by saying innings pitched.  This intrigues me.  I agree that a pitcher's ability to pitch deep into a game is important, as it keeps the bullpen rested and allows pitchers to really assert themselves.  However, I wouldn't go so far as to call it the most important stat.      

     Now by this point it might seem obvious to you that I like ERA.  I do.  I think that it's pretty difficult to come up with a better stat for measuring how good a pitcher is than how many runs he allows.  However, there might be one, and that's WHIP.  Again, for non-baseball people, this stands for walks + hits allowed per inning pitched.  This basically means that besides batters who are hit by a pitch, WHIP is the amount of batters who get on base per inning a pitcher pitches.  

     It's a tough decision as to which is better and they both have pros and cons.  ERA shows how many runs a pitcher allows, but doesn't show guys who are left on base (LOB).  WHIP is the opposite, showing all guys who reach base, but not a total of runs.  Either way, I say these are the best stats for pitchers, and are both much better than wins.

     Thanks for reading,

     Connor