2009 MLB Preview: Divisions

Totals Contest is still open until 8pm EST, but I'm seriously not accepting anything after that closed. I'll have a full post up about everybody's picks later this week, but for now here are mine: (10) Dodgers O, (9) Angels U, (8) Twins U, (7) Blue Jays U, (6) Rays O, (5) White Sox U, (4) Mariners O, (3) Astros U, (2) Marlins U, (1) Giants U, (Tiebreaker) 3.95 ERA.

If I was better at this the following would've been six posts spread out over the course of last week, but that didn't happen, and now the season starts in a few hours, so here goes.

Key:
PECOTA: Projected wins in Nate Silver's system.
CHONE: Projected wins in Sean Smith's system.
Sim: Projected win total used in 20,000 simulations; weighted 66% PECOTA, 29% CHONE, 5% regression to 81. Because PECOTA and CHONE already take schedule into account I had to manually figure out the strength of each team before scheduling; that number is not shown. The simulation also asks for the standard deviation of each team's "true talent", which is important because these projections are of course estimates rather than facts. After consulting with MGL I used .030, although that number itself is an estimate.
Div%, P'Off%: Obvious.
Div, Lg, WS: Best odds on each team to win their division, league, and the World Series, along with where they can be found. The ROI column is just for divisional odds.

AL East
Best odds: Rays, for the second year in a row. I was hoping the Orioles at 100-1 would come out looking good, but sadly that was not the case. I think, if you're looking at it without actually running the numbers, it's hard to figure how likely a team in the Rays' position is to win the division, just because it's rare that a 92-win team is projected to finish third. The +500 available at Bookmaker is a good price, but nothing to go crazy about.

Toughest team to predict: Rays, but only because everybody pretty much agrees on the AL East. PECOTA's 94 is about the most optimistic win projection you'll find on them, although the Yahoo guys are pretty bullish. Nothing here looks unreasonable to me; most other systems have them between 89 and 91 wins.

AL Central
Best odds: Indians, by a wide margin. Yes, a team with Scott Lewis, Anthony Reyes, and Carl Pavano in their rotation (however briefly) is a good bet to win its division. And that +275 that Carib was initially offering (back when Carib used to post worthwhile odds) is really looking good.

How is an obviously flawed team a significant favorite over four other obviously flawed teams? To further investigate, I made up five categories and threw together a table with each team's projected VORP in each category. This is flawed in various ways (including the lack of a distinct category for fielding), but:

There are a few interesting things about this, even beyond the color scheme.

While the Indians' rotation isn't very good, they're still middle of the pack in that category, since none of the five teams have a particularly impressive collection of SPs. This will happen, I suppose, when three teams hog all the good pitchers. You'd think the Twins would do better here, but they don't just because Blackburn and Perkins are unlikely to even approach last year's performances. Somewhat absurdly, the Royals come out on top in this category, with Meche, Greinke and three replacement level guys.

Cleveland ends up on top because they're in the top three in each category, they gain a lot of ground on all the non-Detroit teams with their lineup, and they jump ahead of all four rivals in the bullpen category. The top five in the Indians' pen (Wood, Lewis, Perez, Smith, Betancourt) is one of the best in all of baseball (I guess Boston's is probably better), with Lewis having the highest PECOTA ERA in the group at 4.09.

So, yeah, it's easy to discount the Indians' chances by bringing up the back end of their rotation, or even just "Pavano"; I do it all the time. But the lineup is strong, especially if Victor Martinez returns to his '04-'07 form, and while they may only have two recognizable names in the rotation they do have some pretty impressive pitching depth.

Toughest team to predict: Twins, mostly because of the big gap between PECOTA (76) and the O/U (83). Their .305 BA w/RISP last year probably has something to do with that, and, as noted above, I think their rotation is somewhat overrated. Their lineup after Mauer (if healthy) and Morneau is also pretty ugly, although I'm pretty sure that's no secret.

Perhaps not coincidentally, there's also a big split with the Twins between the more "traditional" Yahoo quartet (average prediction: 84.5 wins) and the non-Marcel sabermetric projections, which all have Minnesota winning between 76 and 80 games.

AL West
Best odds: Mariners. In a slight departure from last year, Seattle's to win division odds look to be the best on the board. Part of this is their pathetic performance last year; the Yahoo guys average out at just 68 wins on Seattle. Part of it is the team sitting there at -155, which will be discussed below.
The Ichiro injury news isn't encouraging, but he's only expected to miss eight games, and and remember PECOTA hated him in the first place (.291/.337/.358), so its Mariners projection was never really dependent on 750 PAs from the career .331/.377/.430 hitter. Seattle is not going to score a ton of runs, but their rotation's combined VORP of 94.2 puts the entire AL Central to shame, and could be enough to keep them in the race out west.

Toughest team to predict: Angels. If you look at the Pinnacle win totals, you'll notice that the juice on all the odds indicates that most of these numbers have been bet toward each team's PECOTA projection since they were first posted. As of about a week ago, though, the Angels were still sitting out there at 89 wins, with the under listed at just -115.

The increasingly negative injury news about their rotation may have ended that, as they're currently off the board at Pinny and +125/-145 at 5Dimes. But for awhile it seemed like the Angels were the one team for which the odds were holding a large margin away from CHONE and (especially) PECOTA. Some of that may be explained by people betting on the over just based on last year's win total, and even the more sabermetrically inclined having a tougher time siding with PECOTA because of its history of missing low on LAA.  Regardless, now that Joe Saunders is their opening day starter it looks like just about everyone thinks the West is going to be a tight race.

NL East
Best odds: Nationals, who if you believe these numbers are the second best divisional bet out there. There's a 10-win difference between the Yahoo consensus and PECOTA/CHONE here, and I'll admit that I didn't expect anyone to have them winning 77 games back in February. It would take some fluky peformances/good luck for Washington to come out of the East, but that'll happen every once in awhile. Olsen-Lannan-Cabrera-Zimmerman isn't going to remind anybody of the '54 Indians, but that's not a total disaster, and the lineup is pretty good, especially if Nick Johnson stays healthy (basically a lock) and Anderson ".223/.273/.305" Hernandez doesn't get 288 PAs.
Toughest team to predict: Mets. The big gap here is actually between PECOTA (92) and CHONE (86), with the O/U splitting the difference. The Yahoo guys are also all over the map, coming in anywhere between 86 and 93. I'm sorry, but if they only win 86 games in the National League with that core, somebody should be fired. Again.

NL Central
Best odds: Reds, but there's really not much to see here. The Cubs are big favorites but, unlike the Angels, not unreasonably so. It's possible that with all their young guys one could make a case that the standard deviation of Cincinnati's performance should be larger, but while that would help I doubt it would make a huge difference.

Toughest team to predict: Pirates. PECOTA says 64, while CHONE is high on them at 73. Almost everyone else is in the middle (it'd be hard not to be), although the Yahoo guys are in the low end of that range. I don't have much to add beyond the fact that if you're giving Nyjer Morgan 323 PAs, you're doing it wrong. Not that I've given a ton of thought to this, but it seems like moving McLouth to left and plugging McCutchen into center would help.  They're in good hands now, at least.

NL West
Best odds: Dodgers. While LA does seem to be the consensus favorite, I don't think the gap is generally perceived to be a large as PECOTA has it. That shows in the odds across the board, as the Dodgers' average division odds are +125, while the D'Backs come in only slightly higher at +145. Despite the park effects in Arizona's favor, PECOTA has the Dodgers scoring five more runs, and one through eight (or, at least, 1-6 and 8) that lineup is indeed pretty impressive. The D'Backs don't have any obvious holes offensively either, but they also don't have any guys with a projected VORP of over 30; LA has four such hitters (Ramirez, Martin, Kemp, Ethier). Arizona's rotation is tough to beat, but the Dodgers do make up a bit of that ground in the bullpen.

Toughest team to predict: All of them. The NL West is the anti-AL East, and in more ways than just the quality of teams. There's at least a six game gap between PECOTA and CHONE on each team except the Giants, and even if CHONE knew they had Ramirez I doubt that'd get them above 87 wins. PECOTA is unusually low on the Rockies (really hates their run prevention, has Franklin Morlaes pegged at at 7.01 ERA), while CHONE is uncommonly bullish on the Padres.