Monday, November 29, 2010

Why Counting Stats Just Don't Cut It

The casual baseball fan has a few stats they like to look at, this is mostly because they are easily understandable and readily available from well known sources. For batters, these sort of stats would be: batting average (AVG), homeruns (HR), runs batted in (RBI) and the various other hits (1B, 2B, 3B). For pitchers, these stats would be: wins (W), losses (L), earned run average (ERA) and strikeouts (K).

WSU starting pitcher, Chad Arnold.
Arnold, who was drafted in last years
MLB Amateur Draft, did not come to
terms with the Dodgers in time and
will return to WSU this Spring.
With the exception of AVG, a simple percentage stat, we call these statistics "counting stats". It's a fairly self-explanatory name, these stats are simply counted up along the course of a season and used to evaluate the worth of a certain player. Well, this post is designed to begin hacking away at the belief widely held that these counting stats are a sufficient means of evaluating a player.

Unfortunately one stat has to be the one to get attacked for the greater good. I'm sorry "wins" (W) but today, you go down.

For most casual fans, as well as a good amount of baseball writers and analysts, wins are seen as one of the two best gauges for measuring a pitchers value. I'll start out by describing how a pitcher goes about earning a "win."

According to MLB, the official definition for the winning pitcher is: the pitcher who last pitched prior to the half-inning when the winning team took the lead for the last time.

So, what does a pitcher need to do to earn a win? Be the pitcher who pitched last before his team took the lead, without ever losing that lead later in the game. (Also, as an attachment, there is a minimum requirement of 5 inning pitched for a pitcher to get a W)

So with that in mind, I'll set up the following situation: In the 7th inning of a game, the starting pitcher is pulled with his team leading 12-11. This pitchers team will go on to win 12-11, never surrendering the lead after the starter left the game. If you go by the official rules, this pitcher should be awarded a W. Good job, right? Absolutely not.

This pitcher gave up 11 runs. 11! It's another flawed statistic, but his ERA for this game (assuming all runs were earned) would be 14.14. Just because his offense was able to out-perform the opposing team, this pitcher gets a win for giving up 11 runs, a start that should never be awarded with a positive statistical boost.

The reason this statistic is so flawed is that it takes into account many, many factors that are not in a pitchers control. The pitchers team offense/defense for example. I'll illustrate another game. Pitcher throws a 4-hit, one run complete game but his team losses the game 1-0. The run was scored on a fly into left field that was dropped by a sub-par fielder. Is that the pitchers fault? No it most definitely is not, yet it will be reflected in his line as a "loss."

If this stat was largely ignored, I would have no problem with it, but the problem is that this is not the case. Wins are looked at as a tell all stat by far too many fans and even some writers and analysts. In fact up until this year wins were arguably the biggest deciding factor for most writers when voting for the AL and NL Cy Young awards. 

Now this is absolutely not the only stat I have qualms with, this is just the best example for me to use. I'm sure wins will come up again, and in the near future I'm going to be attacking other stats. I'm sorry, traditionalists, but the statistical side of this sport is changing, fast. 

No comments:

Post a Comment