Wednesday, January 26, 2005

Cellar Dwellin' with PECOTA?

My ticket snafu hasn't been resolved yet. I haven't heard anything since Monday night other than, "sit tight". But, in the meantime, I got inspired to post again because Baseball Prospectus released their PECOTA projections for the 2005 season. Since it's a subscription service, you won't see me regurgitating all the raw data.

Before I look at their projections, though, I'd like to say thanks for the nod this blog got in the Washington Post, while I was off pouting about my ticket situation.

Their projections are a useful way to compare teams, except that their projections didn't capture the last week or so of transactions. Here are the problems I see with their projections in the NL East:
  • Loaiza and Osuna are not considered in the analysis of the Nationals.
  • Delgado is not considered in the analysis of the Marlins.
  • Worse, Delgado is credited to the Mets!
  • Smoltz' projection is befitting a closer, not a starter, to the detriment of the Braves.

Without adjusting for those people, here is how PECOTA projects the division (remember, VORP is a relative indication of total value towards winning):
  1. Mets (596.8)
  2. Phillies (538.6)
  3. Braves (448.7)
  4. Nationals (437.4)
  5. Marlins (434.9)
Adding in Delgado to the Marlins would seem to bump them up to about 60 points (Delgado himself is worth about a 45, then he'd probably replace a corresponding value of about -15). If we can just trade Delgado's VORP for Mientkiewicz, the Mets only drop about 30, and the number is probably less considering that he'll play a full year for NY, whereas he'd be sharing duties should he remain in Boston. Let's give them a -25.

PECOTA actually thinks our entire pitching staff is "better than replacement level", which is good, but adding in Osuna's 22.9 and Loaiza's 17.9 doesn't offset any negative numbers. In fact, Loaiza's projected value probably balances out a nearly similar effort from other starters. I think we would gain no more than 30 points from these two additions. A +30 for us.

Smoltz is a bit of a wild card. Just in a closer role, he'd rack up a 20.8 VORP. I get the feeling that PECOTA would have projected him above a 40 if he'd been considered a starter. Let's say +20 for ATL.

Adjusting for this, we get:
  1. Mets (571.8)
  2. Phillies (538.6)
  3. Marlins (494.9)
  4. Braves (468.7)
  5. Nationals (467.4)

One way or another, we're clearly going to have to fight hard not to end up in the basement.

Not only are these numbers just projections, I do think these numbers are really only valid for projecting possible 1st half performance. They can't take into account what moves are going to be made in the middle of the summer, which can have a big impact.

Interestingly enough, while PECOTA doesn't think our rotation has a superstar, it's in line with everyone except the Mets. And, our bullpen gets a lot of credit in comparison... it's the thing balancing out the fact that PECOTA thinks we have a horrible offense. Let's look a bit closer.

Without adjusting for recent additions, here's what the system thought about overall pitching staffs:
  1. Mets (256.4)
  2. Phillies (231.7)
  3. Nationals (224.6)
  4. Marlins (218.9)
  5. Braves (178.7)
Once you add in Osuna and Loaiza, we look to have the #2 pitching staff overall. Just looking at the core set of starters, we rate #3, behind NY and FLA (though Smoltz brings ATL up above us, if I'm right about him). But, PECOTA doesn't think Armas is going to pitch very much, so if he's healthy and effective, we might not just have the second best pitching staff and bullpen, we might have the second-best rotation.

The offensive picture is bleak. Here are the rankings, before adding in Delgado:
  1. Mets (340.4)
  2. Phillies (306.9)
  3. Braves (270.0)
  4. Marlins (218.9)
  5. Nationals (212.8)
Once you credit Florida for Delgado, we're way behind everyone else in terms of offense. Clearly, PECOTA doesn't like the Guzman and Castilla signings any more than we do (Castilla is projected to be the bigger bust for 2005, having a projected VORP just barely above 0).

And, to think, some people on the Ballpark Guys forum took offense to my calling our offense anemic! At least I have the best statistical projection model to date on my side!


At 9:45 PM, Blogger Yuda said...

Our offense is definitely mediocre at best, but I think they're lowballing Wilkerson.

Not that it boosts us out of the basement or anything. Just sayin', I guess.

At 10:19 AM, Blogger Yuda said...

Upon further consideration, I think last night's comment of mine was fueled more on beer and homerism than any reason to believe we've got a better offense than PECOTA says. ;)

At 11:17 AM, Blogger John said...

Heh. I think you're right that the projections are wrong on Wilkerson. There are also a few players giving us negative VORPs that won't ever see any playing time. But, other teams seem to have similar issues, so it's likely to all wash out.

I think you'd have to stretch to see our offense as anything other than the worst in the division, even if there are some things that PECOTA is clearly getting wrong.

At 12:55 PM, Blogger Basil said...

Thanks for all that effort, John. PECOTA has always fascinated me, if only for the sort of mysterious introduction of it in BP2001 (I believe).

As I've discussed with Yuda and Ryan, I'm thinking of joining the BP subscriber ranks. It would be helpful, if only for access to the full stats pack. What's your thoughts on subscribing?

At 4:56 PM, Blogger John said...

I'd say it's more useful than the book, even, if only because I can't carry the book around with me everywhere. I also enjoy many of the premium articles. I think it's worth the money, but it's not like I live on a college student's budget or anything.


Post a Comment

<< Home