Mean Regression is Encouraging for Detroit Tigers Fans
By Chris Hannum
Usually mean regression is something to worry about in the back of your mind, when you root for a team that is presumed to be good. Take the Tigers at this point last offseason as an example: in the back of our mind we were worrying about whether that was really a 95 win team and whether it was really fair to expect that kind of production from Jhonny Peralta and Alex Avila (to name two).
Mean regression refers to the tendency for a guy who overachieves to crash back to earth and for those who slump to rebound, and for the 2012 Tigers mean regression bit down hard. It wasn’t actually a 95 win team. It wasn’t fair to expect repeats from Peralta and Avila, etc… Mean regression also can refer – at the team level – for the tendency for a whole team that overachieves to crash back to earth and for a whole team that stinks it up to rebound. There are two reasons for that: one is that the team may have actually played better (or worse) than their record indicated – like the 2012 Baltimore Orioles – and the second is that terrible teams can get better through attrition alone as terrible players are replaced with other warm bodies who aren’t quite as bad. It’s hard to improve a 2 WAR position, but pretty darn easy to improve a -2 WAR position.
Sep 11, 2012; Chicago, IL, USA; Detroit Tigers right fielderBrennan Boesch
(26) makes a catch against the Chicago White Sox in the fourth inning at US Cellular Field. Mandatory Credit: Mike DiNovo-USA TODAY Sports
Now, the Tigers were just about as good as their record in 2012 and the Tigers have at least as many cogs who underachieved in 2012 as overachieved. We aren’t expecting them to crash earthward on either count, despite the fact that they were a good team – they were a good team that underachieved and made the playoffs by virtue of a lack of competition. On the last count, though… typically it is terrible teams that are able to improve just by cutting out the dead wood. Dead wood (which I count as guys who give you below replacement level production) is a big part of what makes a bad team bad and it is hard to have a good team with such embarrassments dragging you down. But… the 2012 Tigers managed to do it, at least on the positional side. “Dead weight” on the Tigers positional roster subtracted 5.7 wins (by Fangraphs WAR) from the team’s total in 2012 – and while just about every team winds up with some guys getting some playing time at below replacement level that 5.7 wins was good for 4th most/worst in baseball. More stink than the Indians and more stink than the Royals. The least? Anaheim, with 0.6 wins lost due to stink.
Anybody worth keeping on a roster should be capable of replacement level production as at least their expected baseline – that is how the term is defined. If some guy scrubs out and shows that he isn’t capable of that, it should be easy to replace him with a AAAA warm body who can. For those of you who have forgotten the 2012 season already, here’s a short list of the Tiger offenders:
1. Ryan Raburn -1.5 wins
2. Brennan Boesch -1.3 wins
3. Don Kelly -0.7 wins
4. Delmon Young -0.7 wins
5. Ramon Santiago -0.3 wins
6. Brandon Inge -0.3 wins
7. Omir Santos -0.3 wins
8. Jeff Baker -0.2 wins
9. Danny Worth -0.2 wins
Any guys that you expect to have a real role on the 2013 team? Ramon Santiago probably will (or Danny Worth will usurp his spot) and Brennan Boesch might – though I figure that if he doesn’t get traded he’ll probably sit in AAA until somebody gets hurt. Don Kelly will be at Toledo without any real hope of a 25-man spot. The rest are just gone. And good riddance. Their at bats and innings in the field might not go to stars, but they will go to players we expect to play at or above replacement level! So… if we get a whole season from Omar Infante, remember that he isn’t just going to be a 2 win guy – he’s going to displace 1.8 losses from the horrible, horrible play of 2012 Inge and 2012 Raburn. Torii Hunter may regress significantly from his 2012 season, but he’s going to displace 1.3 losses from Boesch on top of whatever else he adds – same for Victor Martinez. We should expect a positive mean regression, just from getting rid of this dead wood, of about 3 extra wins – which is something that not many playoff-caliber teams can say. After all, the 2011 Tigers finished with only 2.2 negative WAR from their below-replacement-level guys.