March 29, 2005
Lies, Damned Lies
PECOTA Projects the NL
You should all know how this works by now, so let's jump right in.
W L PCT RS RA Phillies 90 72 .554 833 744 Braves 82 80 .506 758 749 Marlins 81 81 .502 724 720 Mets 81 81 .497 736 741 Nationals 74 88 .459 685 749One of the systemic flaws in offseason coverage is that activity tends to be mistaken for progress; a lot of people seem to think that the Angels had a really good offseason, when it looks to me as though they've added two big, problematic contracts. The Phillies, conversely, made just one significant this move this winter, signing Jon Lieber for a reasonable amount, and the consensus is that they've lost ground. What that opinion fails to recognize is that this is a better team than their record of a year ago suggests.
RS RA 2003 Actual 791 697 2004 Actual 840 781 2005 Projection 833 744The Phillies will hold steady in their offensive output while their run prevention ends up about halfway between where it finished in 2003 and 2004. Are either of those assumptions unreasonable? Phillies hitters didn't have any particularly fluky years a year ago, and they have a reasonable expectation for getting more production out of center field and perhaps left field. Meanwhile, all five of last year's primary starting pitchers significantly underperformed their PECOTA projections. While it's easy to come up with "there's something in the water" explanations (and if that something was Larry Bowa, he's been dumped in the Delaware), more often than not that's going to involve a significant element of bad luck. The pitching is deeper this year--the team has seven viable starters, counting Ryan Madson--and folks like Brett Myers ought to be at least a little bit better. We're not expecting any miracles; shaving 30 or 40 runs off the runs allowed total seems like a pretty reasonable guess.
What's harder to explain is why the Braves and the Marlins don't come out a little bit better. Let's reverse the process a little bit and say all the bad things that we can about these clubs. First, the Braves:
Counterweighting that is the genius of John Schuerholz and Leo Mazzone, the wild card of Andy Marte, and my long, sad history of prematurely projecting the Braves' demise. I think they'll beat this projection, but I don't think they're the favorites to win the division. You can only get away with things like having Brian Jordan in your Opening Day lineup for so long.
Now on to the Marlins:
This is probably a harder case to make than the previous one. Several of the individual Marlin PECOTA projections look low to me, like those for Pierre and Miguel Cabrera (.291/.356/.519), something that I don't see with the Braves. Still, I think we're talking about a couple of wins' worth of underestimation, rather than the Marlins being some sort of silver-speared juggernaut.
I don't feel the same need to defend the Mets' projection, since it's clear that expectations for the team drop off radically once you cross the Hudson River, but I'll confess to being disappointed that the team didn't come out a little bit better. I like the approach that Omar Minaya took this winter, and both Carlos Beltran and Pedro Martinez are projected to have excellent seasons. The problem is that the Mets are a good Playstation club: If you're living in a world where the grass is always fluorescent green and the regulars like Cliff Floyd and Mike Cameron and Jose Reyes and Mike Piazza always have healthy, productive seasons, then the Mets are big playoff contenders. In real life, the odds are against that. Needless to say, missing out on Delgado and settling for Doug Mientkiewicz also hurts a lot, and reminds me of the time that my friend Geoff was too spineless to sneak into Madonna: Truth or Dare, prompting an afternoon with Doc Holliday instead.
As for the Nationals, I'll be looking forward to seeing them come through Wrigley in July, but purely for the novelty effect: this is almost certainly the most boring club in baseball, and its having a GM who is rumored to be trading for Neifi Perez doesn't exactly inspire confidence.
W L PCT RS RA Cardinals 92 70 .569 818 708 Cubs 89 73 .550 773 696 Astros 80 82 .494 724 734 Reds 76 86 .472 789 836 Brewers 73 89 .453 707 781 Pirates 72 90 .444 700 788The version of the projections that I'm working from has Mark Prior pitching 170 innings and Kerry Wood pitching 155 innings. Those are guesstimates at best; the Cubs run one of the tightest PR ships in baseball. It's entirely possible that they are in fact downplaying the seriousness of the injuries, just as it's also entirely possible that the media is overplaying the seriousness of the injuries in response to the organization's relative silence on the matter.
Perhaps the better question is: just how much difference does it make? Let's look at a variety of different scenarios and see the corresponding impact on the team's projected runs allowed and won-loss record:
Prior IP Wood IP RA W-L GB Best Case 200 200 689 90-72 2 Base Case 170 155 696 89-73 3 Bad Case 110 100 707 88-74 4 Worst Case 0 0 732 85-77 8The difference between Prior and Wood pitching, say, 200 innings apiece and 100 innings apiece is not nearly as great as you might be led to think. In either case, the Cubs would rate a couple of games behind the Cardinals, but well within striking distance and in prime position to pick up the wild card. Even if we were to take Prior and Wood out of the mix entirely, PECOTA expects that they'd remain a competitive club, finishing at 85-77 and having a strong chance at a playoff spot in the parity-laden National League.
There are two reasons for this. First of all, PECOTA is already building some injury risk into the performance and playing-time forecasts. Prior's ERA is projected at 3.32, and Wood's at 3.58, which are good marks but would be better if the pitchers hadn't been injured last year, thereby depressing their peripheral stats. More importantly, the Cubs have very good organizational pitching depth. Glendon Rusch and Ryan Dempster can pick up some additional innings without hurting the team too much, PECOTA thinks that Sergio Mitre has sleeper potential, and prospects like Angel Guzman and Bobby Brownlie are pretty close to being ready. A rotation of, say, Zambrano/Maddux/Rusch/Mitre/Guzman would be far from the worst in the league.
The Cubs would have a wider margin for error, of course, if they had picked up a better replacement for Sammy Sosa. On the other hand, there are a couple of trends that work in their favor:
We're living in an era of raised expectations for Cubs fans. The Cubs are going to be competitive even if Prior and Wood continue to struggle with injuries. Unfortunately, a lot of people are going to be unhappy with anything less than ninetysomething wins and a division title.
The Cardinals, as it happens, are not in an entirely dissimilar position from the Cubs. Mark Mulder and Matt Morris have injury concerns of their own, and while they are not as prolific as Wood and Prior, St. Louis doesn't have the pitching depth that the Cubs have, especially with Rick Ankiel's retirement from pitching. We're also projecting the offense to drop about 37 runs, mostly as a result of losing Edgar Renteria's contributions at the shortstop position. Finally, it's worth remembering the Cards outperformed their Pythagenport projection by five games last season. If it weren't for the soft bottom in the Central, it could very easily be the case that the team winning the division will do so with fewer than 90 wins.
The Astros are going to be a poor team this year. They're two starting pitchers short of a rotation, and virtually every regular in the starting lineup comes with a significant age, injury or performance disclaimer. Frankly, I think 80 wins is on the optimistic side; the team's defense is not going to be good, and there are more spots with downside risk than upside risk. The Reds might have been competitive if the pitching problem had been addressed more effectively, but when your projected opening day starter (Paul Wilson) is forecast to have 5.01 ERA, you have a lot of ground to make up. It's hard to construct an argument for the Pirates; PECOTA expects regression from folks like Oliver Perez and Craig Wilson.
I will confess to having a funny feeling about the Brewers: there's a world in which J.J. Hardy and Rickie Weeks make a lot of progress, the bottom end of the rotation sorts itself out, and the team hangs around the fringes of the wild-card race before hitting the wall at some point in August. I'd feel better about them still if they didn't have Brady Clark projected as their starting center fielder. Just for kicks, if Doug Melvin were somehow able to pick off Mike Cameron from the Mets and we inserted him into Clark's spot in the lineup, then PECOTA expects that they'd pick up about 25 additional runs at the plate and another 10 or so in the field, which might be just enough to make them interesting.
W L PCT RS RA Giants 85 77 .526 766 725 Padres 84 78 .518 733 706 Dodgers 83 79 .513 752 732 D'backs 75 87 .463 746 806 Rockies 73 89 .451 864 952In contrast to the Cubs' situation, the Giants' pennant hopes are nearly entirely dependent on the health of Barry Bonds' knees. The scenario above assumes that he'll receive about two-thirds of a season's worth of playing, or about 460 PAs, accounting for both his starting the year on the shelf and sporadic rest and relapses throughout the year. Here's how things might look if we made some different assumptions:
Bonds PT Bonds PA RS W-L GB Best Case 85% 611 785 87-73 +3 Base Case 65% 464 766 85-77 +1 Bad Case 45% 319 751 83-79 -1 Worst Case 0% 0 732 78-84 -7None of this should come as a surprise: Bonds has been worth between 11 and 15 wins above a replacement-level hitter over the course of the past four seasons, and while Michael Tucker and Todd Linden are somewhat better than waiver-wire players, they're a hell of a lot closer to replacement level than they are to Barry Bonds.
While it's clear that the Giants are Bonds or Bust, it's less clear who their replacements would be at the top of the NL West heap. The Padres' offensive core is nearly as old as the Giants', and the eternally grouchy PECOTA regards Jake Peavy's big campaign in 2004 as a career year, rather than a preview of coming attractions. The Dodgers, though a hair behind the Padres here, appear to have a little bit more upside, as well as a little bit more flexibility as the positional and pitching backups are better. They've also got probably the best manager in the game, and a general manager who at the very least is unafraid of fine-tuning his roster in the heat of a pennant race. That said, they're going to need to make up a lot of ground between having exceeded last year's Pythagenport projection by four games, and losing Adrian Beltre and his 11.5 WARP.
Indeed, the division might seem ripe for a dark horse, but neither the Diamondbacks or Rockies look to up to the task. Arizona will be improved, but from a 111-loss benchmark, and a number of their PECOTAs seem on the optimistic side, especially those of Troy Glaus (.281/.391/.579) and Brandon Webb (3.98). Even if the Rockies get a miracle season from Jeff Francis, there is no second fiddle to Todd Helton, and the roster is comprised of organizational filler, rather than the sorts of young players who could emerge with breakout seasons. Top to bottom, this is the worst division in baseball.
Player BA R HR RBI OBP SLG VORP WARP Pujols, SLN* .334 126 43 124 .419 .633 91.3 12.1 Edmonds, SLN* .287 102 41 122 .395 .592 71.4 10.7 Rolen, SLN* .293 104 32 113 .386 .545 58.9 10.4 Bonds, SFN* .339 89 28 93 .532 .687 86.0 10.3 Helton, COL .331 124 33 108 .437 .584 59.2 9.9 Beltran, NYN .280 102 29 93 .377 .508 57.1 9.4 Abreu, PHI* .301 114 28 99 .414 .527 60.8 9.2 Glaus, ARI .281 102 40 112 .391 .579 59.5 8.5 Durham, SFN* .272 110 18 68 .362 .439 50.8 8.5 Dunn, CIN .270 112 42 122 .395 .562 51.1 7.8 Giles, SDN .282 93 23 89 .392 .486 50.8 7.8 Thome. PHI* .274 110 44 124 .395 .582 57.2 7.0 * Plays for projected playoff team.This is everyone with a projected VORP of 50.0 or higher, a threshold that perhaps unfairly excludes Miguel Cabrera, Andruw Jones, and Nomar Garciaparra.
It's safe to say that Barry Bonds is not a particularly good choice for MVP this year. He's already got the steroid cloud working against him, as well as the Willie Mays problem (repeat winners don't make for good headlines, unless they're Roger Clemens), so losing a big chunk of his playing time could all but doom his chances. Bonds might have to hit .400 this year to be an odds-on favorite to repeat.
That said, it would be misguided to preemptively blame the media for fumbling this one. PECOTA reckons that Bonds will only be the fourth-best position player in the league this year, behind the three big Cardinals boppers. Albert Pujols is not only the leading MVP candidate, he's a legitimate Triple Crown threat. PECOTA projects Pujols to have the second best batting average in the league, behind Bonds--but we're guessing that Bonds doesn't get quite enough plate appearances to qualify for the batting crown. It also projects Pujols just one home run of off Jim Thome's lead, and tied with Thome atop the RBI standings. Hell, it expects him to lead the league in runs scored too, producing a potential LeBron James-caliber quadruple double.
The second-best candidate here might be Thome, even though his poor defense hurts his WARP ranking. If I'm right that the media are underrating the Phillies this spring, there could very easily be some overcompensation this fall, and Thome makes for a better representative than Bobby Abreu since he has a higher profile and will have some nice home run and RBI totals. There will also be a lot of momentum behind Carlos Beltran if the Mets manage to reach the playoffs, but we don't expect that to happen.
NL Cy Young
W-L IP ERA BB K VORP WARP Schmidt, SFN* 17-9 215 2.98 67 214 57.4 8.1 Sheets, MIL 16-11 220 3.22 43 220 56.1 8.3 Martinez, NYN 16-8 200 2.93 56 211 54.8 7.8 Zambrano, CHN* 15-10 205 3.51 83 176 44.1 6.8 Oswalt, HOU 13-12 210 3.73 56 171 43.5 6.8 Prior, CHN* 13-8 170 3.32 54 183 43.0 6.3 Clemens, HOU 13-11 200 3.79 66 182 38.3 6.2These are PECOTA's favorite candidates; guys like Mark Mulder, Tim Hudson, John Smoltz and Josh Beckett are also going to be in the running. What we've got here is something of a disconnect between the best starting pitchers and the teams that are going to put a lot of runs on the board, which could lend a significant advantage to any pitcher that managed to reach 20 wins. Jason Schmidt is my pick; he's probably the least recognized really great player in baseball.
NL Rookie of the Year
Player BA R HR RBI OBP SLG VORP WARP Barmes, COL .272 95 16 98 .315 .424 11.1 5.4 Atkins, COL .287 80 16 81 .354 .461 15.7 5.0 Burke, HOU .270 28 3 29 .344 .396 14.9 3.4 Dubois, CHN .252 32 12 34 .344 .487 14.3 3.0 Hardy, MIL .268 23 7 27 .328 .437 11.9 2.5 Marte, ATL .263 24 9 26 .348 .479 10.3 2.1 Pitcher W-L IP ERA BB K VORP WARP Francis, COL 9-9 150 5.04 53 127 23.9 4.2 Floyd, PHI 7-7 120 4.84 57 87 8.9 2.2This is an ugly list. Talent-wise, the best player who we're projecting to get a full dose of playing time is Francis, who has the misfortune of being a Colorado pitcher, though Jason Jennings sets a positive precedent in that department. If Francis balks, or the voters balk on him, that would open the door for one of his teammates on the offensive side; it's hard to imagine a less inspiring ROY selection than Clint Barmes, but these numbers would likely be enough to get the voters' attention. Hardy and, especially, Marte will be viable candidates if they get the playing time; I could also see a surprise winner like the Pirates' Zach Duke.