In discussing Major League Baseball's most relevant statistics, whether it's individual player records, player performance, or team results there's still a big institutional gap between traditional measuring sticks and advanced concepts.
Baseball's new metrics haven't been "new" for at least twenty years. but the mainstream sports media and much of the MLB establishment continue to miss their value and meaning.
And it's not about using the magic computer box to ferret out a series of obscure statistics from esoteric data. It's about an idea. To use quantitative methods to identify those results that measure the specific performance of individual players, not influenced or skewed by the performance of other players.
Bill James, the maven of sabermetrics, characterized it as "objective knowledge".
Only because professional baseball moves slower than an R. A. Dickey fastball do these metrics still feel "new" even though they were introduced in the 1980s.
Matt Cain's 2013 season is one that on the surface he would like to forget. Cain went 8-10 with a 4.00 ERA and did not have a shutout or complete game in 30 starts.
The two questions that should pop up are, 1) did Matt Cain actually have a "bad" year in 2013; and, 2) if he didn't have a bad year, why the 8-10 result?
Keeping in mind that pitcher wins and losses, like hitter RBI, are statistical measures dramatically dependent on the performances of a player's teammates throughout the season.
As a pitcher, you can't get wins if your hitters don't score enough runs, or your bullpen is shaky, or if the defense behind you can't control balls in play or make double plays.
As a hitter, you can't create runs if the guys ahead of you don't get on base, or if you are surrounded by a slow-running offense that needs extra hits to move around the diamond, or if you don't have a multiplicity of ways to get on base yourself.
Looking at Matt Cain's season, he did very well in the statistical categories that isolate his performance: a 1.16 WHIP, a 7.71 strikeout average per 9 innings (33rd among the top 80 MLB starters), and 7.7 hits allowed per 9 innings (8th best among NL starters). Of Cain's 30 starts, 20 were quality starts (at least 6 IP with 3 ER or less) which tied him at 16th among NL starters and 30th of all MLB starters.
So what happened on the way to an 8-10 record?
Comparing Matt Cain to several other starters in 2013 provides some clues:
|Matt Cain, SFG||1.16||184.1||20||2.89||3.40||2||23||8-10|
|Anibal Sanchez, DET||1.15||182.0||20||2.86||5.31||10||9||14-8|
|Julio Teheran, ATL||1.17||185.2||18||3.02||4.03||6||22||14-8|
|Ubaldo Jimenez, CLE||1.33||182.2||16||3.17||4.50||12||16||13-9|
|Bartolo Colon, OAK||1.17||190.1||23||2.90||5.43||21||14||18-6|
Two things immediately jump out: run support and double plays. San Francisco Giant hitters provided Cain with an average of 3.40 runs per start, which is terrible. How terrible? Cain was 71st out of the top 80 starters in the Majors in run support per start.
Of course the Giants offense skunked the entire pitching staff, coming in 21st of 30 MLB teams with 629 runs and 29th in HRs with 107. MLB teams scored an average of 4.16 runs per game in 2013 and NL teams hit an average of 144 home runs.
Cain wasn't helped by San Francisco's shaky defense, whose 107 errors placed them 24th among 30 MLB teams. But the killer here was the inability of the Giants infield to turn the double play-- they were 28th in baseball in average DPs turned per game (0.79) and it killed Matt Cain's starts.
Contrast that with Bartolo Colon, who not only got 21 double plays turned behind him, his run support was 1.27 runs above the MLB average. There are certainly many other variables here, including League and park factors, but there's one inescapable conclusion.
The combination of a poor defense and the inability to produce runs is what led to Matt Cain's mediocre win-loss record. Look what Julio Teheran did with a similar WHIP, only 3 times the double plays, and run support much closer to the MLB average. And Teheran has two less QS than Cain, a higher opposition OBP and gave up only one less HR than Cain.
This is all good news because Matt Cain's 2013 performance was stellar and there's no reason to doubt that 2014 will be the same or better. Only in those categories where Cain depended on the performance of his teammates did things fall short.
There's no doubt that baseball is a team sport, but traditional measures of player performance mask what individual pitchers and hitters actually accomplish during any given season.
I have to be honest here. This article is incredibly short-sighted. While I agree with the overall idea that Cain pitched much better than the numbers indicate, I really don't feel like you solved anything with your analysis. In fact, you actually did the same thing in your argument that you mocked MLB for in the first two paragraphs. The idea of Cain, or any Giants pitcher for that matter, not getting great run support is nowhere close to being a new concept. We're talking about a guy (Cain) who compiled a 15-30 record between 07-08 while logging ERA's in the mid-3's. There's a reason they call it being "Cain'd" when a Giants pitcher throws a gem but doesn't get a victory. The guy has a career 67% quality start rate, yet his overall record is only 93-88. In a perfect world, it should be much, much better than that.
Cain was a great pitcher in 2013... for 25 starts. For the other 5 starts, he pitched very poorly. In those 5 starts (StL, Mil, Col, StL, LAD), he allowed 37 earned runs. That's 45% of his total earned runs for the entire year (82), in 1/6 of his starts! In the two starts against St Louis, he gave in 16 ER. That's 19.5% of his yearly total! You're right. Cain was a better pitcher than his win/loss record shows. But it's not because he got poor run support or his defense didn't turn enough double plays (he's a fly-ball pitcher anyway)... it's because he was extremely hittable in 1/6 of his starts.
I don't mean to be offensive on your site. But if you're going to make a stink about people being reluctant to look outside the box in analyzing statistics, then I'd expect you to look a little deeper than run support and W/L record.
The "case" has been solved, Richard. Thank you very much. Matt Cain's superficially mediocre year had me worried, that is, until now. In actuality, Matt Cain appears to be perpetually unlucky, lack of run support, even less than what his colleagues on the Giants seem to receive. At some point, do you think you could investigate his whole career record. I wonder whether I'm right, I mean, about lack of run support and perhaps then fielding support (lack of double plays, errors).
CC- You're not offensive at all. But I think you missed my point here. You said "...I'd expect you to look a little deeper than run support and W/L record."
My whole piece was about the importance of not paying attention to a starting pitcher's W/L record. The majority of fans, players, and MLB front offices still look at that obsolete stat as the ultimate measure of a starting pitcher.
Also, reread my first three paragraphs-- my point was that run support is not "a new concept", it's just a concept that is routinely ignored by most of the MLB establishment.
The breakthrough came in 2010 when Seattle's Felix Hernandez won the AL Cy Young Award with a 13-12 record. But the majority of the MLB establishment still believes in the traditional numbers-- they don't get that Mike Trout was the real AL MVP the past two seasons (not Detroit's Miguel Cabrera).
Awareness of isolating and evaluating individual player performance is slowly making headway in the game. But only slowly.
Two more points.
You can pick out four, five, or six bad starts by every starting pitcher in the game each season to rationalize that, without those starts, they would have had a "great" season. But you know what? Those starts count-- just like the good ones.
In 2012 Zack Greinke of the Dodgers started 34 games (last season Cain started 30). Greinke had six games in which he gave up 32 earned runs. So for 28 starts Greinke pitched great.
Greinke's 1.20 WHIP is slightly higher than Cain's 2013 number, but Greinke went 15-5 in 2012. Which demonstrates what run support can add to the equation.
And Matt Cain didn't get poor run support and sub-par defense in just his four worst starts of 2013-- he, and virtually every other Giants' starter, got terrible run support all season long. It's the #1 reason the Giants finished 16 games out of 1st place last season.
[For the record, Matt Cain's career quality start rate is 66%-- 175 QS, 265 GS].