Wednesday, November 23, 2011

Sabermetrics is for Squares

I was watching Clubhouse Confidential last night with Brian Kenny and they were discussing the NL MVP vote were Ryan Braun and Matt Kemp finished 1-2. One could make a case for either player since they both had great years. Kenny had some guest on his show that was pretty much inventing stats. Braun and Kemps numbers were "adjusted" for their ballparks, and a few other factors. Then they were compared by using some sabermetric stats I've never heard of. The actual stats that happened on the field were barely discussed, I didn't see them talk about Avg., Runs, HR, RBI once and their OBP and SLG were adjusted somehow. Billy Ripken made a joke about making his own adjustments at the end of the show when Kenny said Curtis Granderson leading the league in runs and RBIs was just because of his teammates.

There was an article by Dave Shoenfield on ESPN that said Justin Verlander was kind of lucky this year because his Batting Avg on balls in play was only .236 compared to CC Sabathia's .316. My reaction is, isn't that a good thing on Verlander's part? For the most part, award voters are not into sabermetrics. However, there are a few who are like Keith Law. This is the same guy who voted Javier Vazquez #2 ahead of Adam Wainwright and left Chris Carpenter off the ballot in 2009 because of the WAR and FIP(a stat that doesn't consider outs made by defense)stats. This year he placed Clayton Kershaw third despite the fact Kershaw led the NL in Wins, ERA, Strikeouts, and WHIP. He voted Halladay number 1 because of WAR, and said Lee and Kershaw was a toss up because Lee had tougher competitition. Law didn't mention that Lee and Halladay also had better team defense and a better offense to back them up. Law also left 21 game winner Ian Kennedy off his 5 spot ballot.

WAR is a joke of a stat, its an attempt to put a one number total value ranking on a player but it often has funny results. It stands for wins above replacement, but can a complex formula really determine how many wins a player brings to a team? Pitching is calculated using FIP and BA on balls is play instead of regular pitching stats, and defense is ranked by UZR an unreliable defensive measure. I'm sure there is numerous adjustments for hitters as well. There's also handicaps and penalties depending on the player's position, league, and home ballpark. There is at least 3 different versions of the stat, and its formula is constantly changing. Tony Phillips has a higher career WAR than Lou Brock, and Ben Zobrist ranked higher on one version of WAR than Albert Pujols in 2009. Garbage stat in my opinion.

People who are into sabermetrics are more like accountants than baseball fans. I'm not sure they even like watching baseball and seem to think the game is played by robots not humans. Just like corporations and governments twist numbers around to suit their agenda, baseball stats can be used to do the same. There was alot of talk about how revolutionary Billy Beane's Moneyball was, but without scouts and player development coaches it wouldn't of worked. The movie downplayed scouting, but without scouts the A's would of never had Eric Chavez, Miguel Tejada, Jason Giambi, Barry Zito, Tim Hudson, or Mark Mulder. That was the key to their success more than sabermetrics. However, the Athletics have had 5 straight losing seasons, and haven't made the playoffs since 2006. Maybe Beane should hire some more scouts and baseball people instead of number crunchers.

No comments:

Post a Comment