Have the Mets cost Santana 20 wins?

Six times so far this season Johan Santana has pitched at least seven innings and allowed two ER or fewer, yet his record in those games is 1-0. It’s poor support from his teammates like that which prompted Fox Sports columnist Bob Klapisch to write:

“You could be conservative and say the Mets have cost Santana a dozen victories since 2008. A more realistic estimate would be 15 or even 20.”

It is unfathomable to me that in just over two years with the Mets that Santana has been denied 20 wins. His record in New York is 33-18. Does Klapisch really think that Santana should be 53-18? As it stands currently, Santana’s winning percentage with the Mets is .647, better than Tom Seaver’s and just percentage points behind Doc Gooden for the best all time among pitchers with at least 50 decisions.

A 53-18 record would be good for a .746 winning percentage. Of course, some of these new wins might be games that Santana actually lost, which would make his winning percentage even greater. Perhaps Klapisch envisions Santana being 53-13 (.779) or 53-8 (.869). Those numbers would be nothing short of amazing.

Of course, no pitcher is going to win every time he pitches a Quality Start. The question becomes: At what point is the outing good enough that we should be willing to credit the pitcher a win? There is no definitive answer to this question but we need to put some guidelines out there to better examine the claim by Klapisch that Santana has lost between 12-20 victories.

For this piece, I am going to assume that if a pitcher goes 7 innings and gives up 0 or 1 ER, that is a performance that should result in a win. Furthermore, if he goes 8 innings and give up 0-2 runs, that qualifies as well. Finally, if he pitches a complete game and gives up 0-3 runs, that should get him a win, too.

I do not pretend this method is perfect. But by setting these standards, we have some type of objective way to measure how many wins Santana has lost out on in his time with the Mets. While it is not perfect, it is an improvement over picking a number out of thin air and parading it as fact.

Since he joined the Mets, here is how Santana has fared under the qualifications listed above. Any start with a fraction of an inning is essentially rounded down. So, if Santana pitched 7.2 innings, he would need to give up 0 or 1 run to “earn” a win. The chart below lists the actual results.

Innings 	        Wins	Losses	No-Decisions
7	                10	2	        4
8	                2	2	        1
9	                3	0	        0

So, in games with the Mets where Santana has pitched seven innings and given up 0 or 1 run, he is 10-2 with four no-decisions. If we use this definition of “earned” wins, Santana has been deprived of nine wins in his career with New York. That is a significant total, but short of what Klapisch called “conservative” and far short over what he called “realistic.”

Another way to look at deprived wins for Santana would be to see what his record would be with average run support. Baseball Prospectus has a stat they call Support Neutral Wins (SNW), which aims to do this very thing. Here are Santana’s year-by –year win totals with the Mets, first with actual results and then the SNW total.

2008	16	21.4
2009	13	14.2
2010	4	7.8
Total	33	43.4

This shows a little over 10 wins, which again falls under the total that Klapisch called “conservative.”

So, Santana has been unlucky while with the Mets, even if not to the degree that Klapsich insisted. Still, he has been an extremely valuable pitcher in his time with the club. And if he had won all of the games under the system I proposed, his record with New York would be 42-14 (.750), nearly identical to the 42-13 (.764) mark that Gooden enjoyed his first two years with the Mets and better than the 43-19 (.694) record that Seaver posted in 1969-70.