March 29, 2007
Lies, Damned Lies
PECOTA Projected Standings, AL and NL West
We've reached the conclusion of our three-part PECOTA miniseries. This last time around I'll take a look at the teams in the Western part of the country, and make some predictions that could buy you a lot of baseball cards if they come true.
The ground rules are as follows: for each club, I'm going to identify one player for whom I think the PECOTA forecast looks low ("Take the Over") based on statistical, injury, or scouting information that isn't picked up by the system (or where that doesn't work, based on general gut-feel). We'll do the same thing for a projection that looks too optimistic ("Take the Under"). Finally, I'll present an adjusted overall W-L forecast based on these sorts of factors, as well as 'meta' variables like the strength of a team's management, its depth, and its disposition toward or against making trades ("The Verdict"). One exception to this rule: I'm also going to point out one PECOTA projection that might be out of line with the expectations of other forecasters, but where I think the system is getting it right ("Sticking to My Guns").
Los Angeles Angels of Anaheim (86-76)
Take the Over: PECOTA is penalizing Casey Kotchman for an extremely poor performance in limited playing time in 2006 that was caused in part by his mononucleosis. I maintain that he's an overrated player-even his 90th percentile forecast limits him to .293/.360/.439, which is about average for a first baseman. But he also won't be as bad as PECOTA's .249/.308/.348 forecast line, and if he is, there's no way he'll get 350 plate appearances with the Angels.
Take the Under: You can't spell Gary Nathaniel Matthews Jr. without HGH.
Sticking to My Guns: PECOTA is slightly more optimistic on Howie Kendrick than some of the other forecasting systems, but he has such an extraordinarily quick bat that I have little doubt he'll match the .305 batting average PECOTA is projecting for him. I also like him to beat his PECOTAs in 2008-2011 as he learns how to take a few more walks, something he never really had to do when he was hitting .360 every year in the minors.
The Verdict: The Angels' starting pitching had a few health scares in spring training, but Jered Weaver and Bartolo Colon are progressing nicely, and this rotation should be very, very good. In fact, the Angels would be candidates to be the best team in baseball if they weren't wasting key offensive spots on players like Kotchman, Garret Anderson, and Shea Hillenbrand. I think Bill Stoneman takes at least one opportunity to leverage his deep minor league system and upgrade the middle of his batting order, which has the potential to produce huge gains at the margins. We're going to add two wins for 88-74. Keep in mind that the Angels project to be a very good secret sauce team if they reach the playoffs. They're my pick to represent the A.L. in the World Series.
Oakland A's (80-82)
Take the Over: I wasn't a big fan of the Mike Piazza acquisition, but in the absence of evidence to the contrary, I think we have to go with the intuitive conclusion that moving out from behind the plate will be a boon to his offense. Look for something in between last year's .283/.342/.501 and PECOTA's .264/.324/.443.
Take the Under: PECOTA doesn't get to use 2004 to assist it in projecting Mark Ellis, since he missed that year due to injury. If Ellis had played in 2004, I suspect that his performance would have looked a lot more like his mediocre 2003 and 2006 than his flukish 2005, and that would take some points off his forecast.
Sticking to My Guns: Save for Piazza and Ellis, most of these forecasts are pretty well in line with consensus expectations. If you want to go with an obscure one, Jay Witasick has nine lives, and I don't think he'll have any trouble meeting PECOTA's 4.09 ERA after an awful 2006.
The Verdict: We have to give them one win for Billy Beane's tendency to make optimal personnel decisions, and another win for Rich Harden's wonderful spring, even though he's disappointed us before. That gets us to 82-80. There's some upside beyond that on the pitching staff, but not much on the offense. The A's are under a lot of pressure to perform well in the first half, because if the playoff races are as close as PECOTA is projecting them, it's going to be a real seller's market at the trade deadline, and the A's could easily decide to call it a rebuilding year and take the arbitrage gains.
Texas Rangers (80-82)
Take the Over: I don't necessarily expect Sammy Sosa to beat his extremely pessimistic projection (.219/.293/.385), but I also don't think it will take 339 PA for the Rangers to dump him if his bat isn't more rejuvenated than that.
Take the Under: PECOTA has trouble projecting negative outcomes for Eric Gagne-his 25th percentile ERA is 2.89-because his performance was quite good in the limited amount of time that he was actually able to pitch in 2005 and 2006. He's clearly riskier than PECOTA lets on.
The Verdict: The Rangers have quite a history of performing worse in the second half of the season since moving into the Ballpark at Arlington, one that's been especially profound in recent seasons (see the chart below). My hunch is that this is something systematic, probably related to the burdens of playing 81 games in a park that both burns out the bullpen and burns the mercury. Given the uncertainties in the back end of the pitching staff, I don't expect that trend to abate this year. Subtract 3 wins for a 77-85 finish.
Year 1st Half 2nd Half 1994 42-45 10-17 1995 39-30 35-40 1996 51-36 39-36 1997 43-42 34-43 1998 48-39 40-35 1999 48-39 47-28 2000 42-43 29-48 2001 35-52 38-37 2002 39-47 33-43 2003 38-55 33-36 2004 49-37 40-46 2005 46-40 33-43 2006 45-43 35-39 TOTAL 565-548 446-491 .508 .476
Seattle Mariners (73-89)
Take the Over: PECOTA has been picking on Ichiro Suzuki since he was in diapers. It actually turned out to be right in 2005, but it's been wrong every other time. Nevertheless, there are real reasons for concern-Ichiro is turning 33 this year, and while he never hit many home runs, he was also limited to just 20 doubles last season. That said, he's a unique enough player-you really have to go back to the Deadball Era to find appropriate comparables-that I don't think PECOTA is the best way to make the case.
Take the Under: We talk a lot about big spikes in workloads, but what about big spikes in leverage? This has to be a concern with J.J. Putz after his injury problems this spring.
Sticking to My Guns: Jose Vidro isn't going to be any better than the other 29 major league GMs are anticipating.
The Verdict: This is an old offense without much upside, and the team's management has developed a nasty case of making poor personnel decisions. Still, PECOTA's underrating of Ichiro counts for something. Call it even, and figure they'll be 73-89.
Arizona Diamondbacks (88-74)
Take the Over: Jose Valverde improved a lot in the second half, and as I pointed out in the Sports Illustrated baseball preview, he should be able to consolidate those gains by making better use of his splitter.
Take the Under: Even though the team projection comes off as optimistic at first glance, not many of the individual forecasts do. I suppose that if I had to pick between my children, I'd say that Chad Tracy is going to wind up a bit closer to his 2006 numbers (.281/.343/.451) than PECOTA is anticipating.
The Verdict: You probably have to go back to the 1994 Indians to find a team with this much position-player talent coming up through the system together. We have the Diamondbacks projected higher than most other outlets, but that's because they're a young club, and as I mentioned in Monday's chat, I think the other forecasting systems have some work to do before they catch up with the quality of Clay's translations.
It's worth mentioning that PECOTA is a self-correcting forecasting system-it works by means of a two-stage process. First it creates a baseline forecast for each player: both the 'test player' (the guy whose stats we are projecting) and the comparables. Then it sees how the comparable players performed against that baseline, and makes adjustments to the test player's forecast accordingly. If the DTs were systematically overestimating young talent, for example, then PECOTA would catch that by noticing that the comparables were failing to live up to their baselines, and would downgrade the test player's projection as a result. Being able to hedge against its own assumptions is a big part of PECOTA's accuracy.
In any event, I'm sticking with the 88-74 forecast and going one step further: the Arizona Diamondbacks are my favorites to represent the National League in the World Series.
San Diego Padres (86-76)
Take the Over: Adrian Gonzalez had to bear the burden of being a #1 draft pick early in his career, and may have been promoted too aggressively as a result. Then he got stuck in a bad situation in the Texas system, where he was buried behind Mark Teixeira. Given that backstory, I think he'll hold on to more of his 2006 breakout than PECOTA indicates.
Take the Under: Not that John Schuerholz' judgment is flawless, but you have to wonder why the Braves dumped Marcus Giles for none-of-the-above. Especially after reading articles like this one. I could see the Giles acquisition working out very well, but I could also see it working out very badly.
Sticking to My Guns: Trevor Hoffman's PERA was 3.45 last year as a result of a declining strikeout rate. At age 39, there isn't much chance of a reversal in the peripherals, and the ERA will probably catch up. If Hoffman does regress this year, it will be interesting to see whether the Padres' remain deferential to their Hall of Famer, or give the closer's job to Scott Linebrink or Cla Meredith.
The Verdict: Solid baseball team. Boring baseball team. Solid projections. Boring projections. That adds up to 86-76, just like PECOTA is guessing.
Los Angeles Dodgers (80-82)
Take the Over: Takashi Saito's numbers last year were much better than his Japanese translations anticipated. There are a couple of schools of thought about this. The first is that Japanese translations are inherently speculative, and we shouldn't give them much weight now that we have the domestic alternative on hand. The second is that Saito, like Shingo Takastu, benefited a lot from being new to his league, something which might be evidenced by the fact that his ERA was much higher against NL West clubs (3.64) than it was against the rest of the league (0.25!), which didn't get to see him as often. The second theory is a better theory, but in practice, a 3.70 ERA looks awfully high for a pitcher in Dodger Stadium who struck out more than 12 batters per nine innings last season.
Take the Under: Brett Tomko had a fairly large starter/reliever split last year, and was given the fifth starter's job in spite of being way outpitched by Chad Billingsley in spring training. I expect the Dodgers to pay the price for that decision.
Sticking to My Guns: Jason Schmidt's whiff rate has declined from 10.0 K/9 in 2004 to 7.6 K/9 last season. That's not terribly abnormal for a 34-year-old pitcher, but Schmidt does have lot of mileage on his arm, and it's not very likely that he'll get back to his 2002-2004 performance levels.
The Verdict: This is one of the harder teams in baseball to forecast because of the huge amount of injury risk, exemplified in guys like Jeff Kent, Nomar Garciaparra, Randy Wolf, and now apparently Rafael Furcal. I'm not sure that they're the best team in the division even if they do stay relatively healthy; the NL West has gone from being relatively weak to relatively strong, and the Dodgers lack any kind of true superstar talent. Still, it's a deep club, especially considering their minor league system, and that counts for something. We'll add two wins for 82-80, but we'd add more if their GM wasn't as veteran-happy as Ned Colletti.
Colorado Rockies (79-83)
Take the Over: None that I'm terrifically comfortable with, especially given the ambiguities of the Humidor. I suppose you give this to Troy Tulowitzki on the basis of his scouting reports.
Sticking to My Guns: I'm a pretty big believer in Chris Iannetta.
The Verdict: PECOTA overprojected the Rockies by eight games in 2003, five games in 2004, and six games in 2005. It was almost dead-on last year, but I have to wonder if the system is failing to account for the inherent disadvantages of playing in Coors Field. Talent-wise, this is the best Rockies team in years, especially as they've been able to patch their holes at catcher and shortstop with rookie solutions, but I'm going to have to see them play .500 ball to believe it. Subtract three wins for a final tally of 76-86.
San Francisco Giants (79-83)
Take the Over: I suspect that PECOTA has Barry Bonds' rate performance about right (.267/.441/.535), but that's still enough to make him one of the five most productive hitters in the league. It's all going to boil down to how much he plays. Right now, we have Bonds projected to accumulate exactly half of the Giants' playing time in left field, which is looking conservative after a productive and quiet spring. There are a whole host of scenarios where things could go very, very wrong with Bonds, but the initial assumption is that he takes a more businesslike approach this season, and beats his playing time forecast.
Take the Under: We understand PECOTA's appreciation for broad-based skill sets, but the Ray Durham forecast (.304/.373/.500) looks a little zany.
Sticking to My Guns: AT&T Park is very tough on left-handed power, but it's more forgiving in the other dimensions of offense, particularly the triples that Dave Roberts loves to hit. Nice signing by the Giants, and I expect him to defy his age and match PECOTA's .295/.365/.407 projection.
The Verdict: All kinds of weird forecasts here. Durham, Randy Winn, and Omar Vizquel look high, but Barry Zito and Matt Cain look a little low. Tim Lincecum could sneak in and be one of the best pitchers in the league after the All-Star break, or contribute absolutely nothing to the Giants this year. And there's a ton of injury risk on this club. If you had to pick one team in baseball that could either win or lose 95 games this season, this would be it. But if my assumption is correct that Barry Bonds is going to play a little more often than we currently have him projected, that has to be worth at least a +2, putting the Giants at 81-81.
We've cycled through three different iterations of the depth charts in the process of running this series, and a couple of the team projections have fluctuated as a result. I'm also going to make three slight adjustments to my previous projections based on late-breaking developments, which are as follows:
That leaves things to settle out as follows:
AL East PECOTA (3/27/07) Nate ----------------------------------------- New York Yankees 93-69 95-67 Boston Red Sox 92-70 92-70 Toronto Blue Jays 80-82 82-80 Tampa Bay Devil Rays 78-84 75-87 Baltimore Orioles 75-87 74-88 AL Central PECOTA (3/27/07) Nate ----------------------------------------- Minnesota Twins 91-71 88-74 Cleveland Indians 90-72 89-73 Detroit Tigers 85-77 88-74 Chicago White Sox 73-89 77-85 Kansas City Royals 66-96 66-96 AL West PECOTA (3/27/07) Nate ----------------------------------------- Los Angeles Angels 86-76 88-74 Oakland Athletics 80-82 82-80 Texas Rangers 80-82 77-85 Seattle Mariners 73-89 73-89 NL East PECOTA (3/27/07) Nate ----------------------------------------- Philadelphia Phillies 87-75 86-76 New York Mets 86-76 88-74 Atlanta Braves 82-80 83-79 Florida Marlins 79-83 77-85 Washington Nationals 66-96 59-103 NL Central PECOTA (3/27/07) Nate ----------------------------------------- Chicago Cubs 85-77 87-75 Milwaukee Brewers 85-77 83-79 St. Louis Cardinals 81-81 81-81 Houston Astros 80-82 81-81 Pittsburgh Pirates 76-86 74-88 Cincinnati Reds 71-91 72-90 NL West PECOTA (3/27/07) Nate ----------------------------------------- Arizona Diamondbacks 88-74 88-74 San Diego Padres 86-76 86-76 Los Angeles Dodgers 80-82 82-80 Colorado Rockies 79-83 76-86 San Francisco Giants 79-83 81-81
Finally, let's talk about my playoff picks. These are not terribly scientific, but rely to some extent on the secret sauce method.
ALCS: Angels defeat Red Sox, 4-2
NLCS: Diamondbacks defeat Padres, 4-1
World Series: Diamondbacks defeat Angels, 4-3
A little goofy? Sure, but the one thing I'm almost certain about is that this season is going to be a lot of fun.