Saturday, 4 July 2015

Susceptible Stats

While the C-LAY-ton Kershaw system is currently taking the world of baseball betting by storm, it's worth taking a moment to explain to non-baseball aficionados that a pitcher's Win - Loss record is not actually a very meaningful statistic.

It's actually a very good example of how context is important with your statistics, as the insightful Shapeshifter has previously mentioned.

A pitcher's Win - Loss record is very much a secondary statistic and one of those statistics more susceptible to luck. From Nate Silver's The Signal And The Noise:
If you want to predict a pitcher's win-loss record, looking at the number of strikeouts he recorded and the number of walks he yielded is more informative than looking at his W's and L's from the previous season, because the former statistics are much more consistent from year to year.
Strikeouts are important because they stop a batter from reaching base, and if he fails to get on base, he's not going to score a run. If he doesn't score a run, the chances of his team winning the game are reduced.

He goes on to explain how a pitcher's win-loss record is also affected by how many runs his team's offence score - something he has little control over, and in the American League, no control over at all. Poor Mr. Kershaw actually suffered from this last night, although he didn't take a loss as he was out of the game by the time it became a loss. Kershaw conceded just one run in seven innings, while striking out seven.

In other words, Kershaw is better than his recent Win-Loss record might suggest. A lesser pitcher, which is most pitchers, might well give up several runs, but if his team can score several runs plus one, he gets credited for the win.

Kershaw's average Strikeout per innings is up a little on last season at 1.28 (2014:1.20) but his Bases on Balls total is down a little 27 from 114 innings (2014: 31 from 198).
Eleven years ago today, Greece won Euro 2004. Today, they'd be happy to win 2004 Euros 

No comments: