As you probably heard, the first batch of the 2011 PECOTAs were released on Monday. The first wave includes a spreadsheet that lists each player and their weighted mean projection for the upcoming season.
A quick word about PECOTA and projections: They’re fun. Not gospel. Just because PECOTA says Kila Ka’aihue will bash 25 home runs, that doesn’t mean it’s going to happen. And it’s not a failure of the system if he doesn’t hit 25 home runs. One of the great things about PECOTA is, it breaks down their projections by percentile. It’s kind of a best case versus worst case scenario for each player. The percentiles will be on the Baseball Prospectus website in a few weeks. For now, all we have to play with are the weighted means.
I’m partial to PECOTA because they’ve been around forever (and continually fine tuned)and because I’m an employee of the company. (Full disclosure and all that.) I had nothing to do with the projections because I have nowhere near the brain power required to crunch the numbers… Let alone launch Excel. Just because I like PECOTA doesn’t mean I ignore the other systems that are out there. They all have their strengths and they all have their weaknesses. Probably the best thing to do is throw all the projections into a pot, stir ‘em up and see where the numbers fall.
I posted his home run and on base projections to Twitter the other day and got quite the response. One of the weaknesses of PECOTA I think is found in it’s projections of players who don’t have a ton of major league experience.
When I saw these numbers, I immediately thought of the percentiles, because this seems awfully optimistic to me. The funny thing is, Bill James shares PECOTAs optimism, projecting 22 home runs and a .375 OBP for the Hawaiian Punch. Marcel… Not so much. Just 10 HR (although in about half the plate appearances) and a meager .325 OBP. (That’s one of my issues with Marcel. I don’t want to extrapolate projections for an entire season. Just assume each player with projections will play a full compliment of games. Projecting playing time is even more speculative than projecting performance. Especially when it comes to a player like Kila.)
This gives me an opportunity to mention his slider bat speed.
53% Improve Rate
According to Baseball Prospectus, Improve Rate is the chance a hitter will improve at all based on his three previous seasons. An improve rate of 50% means he will perform the same as in the past. That doesn’t mean that Cabrera only has a 3% chance of improvement. Rather, it means he’s more likely to build on his performance from the previous three years. I know… Semantics.
The Melk Man’s Improve Rate is the highest on the team.
One of his comparable players is Gregg Jeffries. This delights me, because I really disliked Jeffries. I have the feeling I’m going to feel the same about Melky.
Maybe I just have an irrational affinity for short left-handers, but I’m really excited to see what Collins can do at the big league level. While I mentioned PECOTA struggles with players with not much major league experience, it’s probably a little easier to come closer for projections with relief pitchers. Collins’ strikeout rate is projected a tad on the high side for my taste, but I don’t think it’s way out of line. If he gets 60 innings or so, there’s no reason to think he won’t top 60 punch outs.
I’m extremely hopeful he opens the season in Kansas City.
PECOTA (and other projection systems I’m sure) nailed this trend that will develop throughout the summer in Kansas City – the Royals just don’t have the horses in the starting rotation to rack up the strikeouts. This is going to be a problem.
Thankfully, the Yunigma is gone and Lorenzo Cain (if he gets playing time in CF) mean the defense up the middle will be stronger than the last couple of seasons.
Believe me, the defense is going to play a huge role on this team.
Comparable Players: Steve Kemp, Barry Bonds, Roger Maris
This is where PECOTA gets some heat… And deservedly so. Ignore for a moment that Alex Gordon was listed in the same breath as Barry Bonds. He was mentioned with Bonds and Steve Kemp? Seriously? PECOTA must have been on some sort of 80’s acid flashback.
The comps are a known problem. Before you flip out (or decide that Gordon has the potential to hold the single season record for home runs… Wait… That would be flipping out) just realize that this is an area that is continually being fine tuned. And there’s a lot of work to be done.
The moral of this story: Enjoy the projections, but take them with a grain of salt. They’re something fun to look at to pass the time before pitchers and catchers report and the games start. Nothing more, nothing less. Sometimes, they’re crazy accurate. And sometimes they’re so far off the mark it’s like the numbers were run on an overworked Commodore 64.
Either way, it’s just another sign that the new season is almost here. Thank god.