Mid-season Prescriptio... (07/15)
the archives are now free.
18 comments have been left for this article. BP Comment Quick Links
| |
|
|
July 15, 2010 Prospectus Hit and RunResetting The Races
With the All-Star break upon us, it's a convenient time to re-evaluate the playoff races relative to our pre-season PECOTA projections. Unlike last year, when three teams the system tabbed as potentially playoff bound had more or less flatlined by the break, none of our initial picks are totally out of the running. That said, injuries to several Phillies and Red Sox have turned both teams from favorites into underdogs with regards to reaching October, and no less than 11 teams have seen their chances of making the playoffs fall below 2 percent. That's the conclusion when we compare the plain vanilla version of our Playoff Odds report against our pre-season projections. In that report, each team's current record and third-order Pythagorean record— their record after adjusting for scoring environment, run elements, and quality of opposition—are factored into a Monte Carlo simulation of the rest of the season, with their records regressing not to .500 but to their third-order winning percentages. Run differentials are the key here; from a predictive standpoint, the sun tends to shine on teams that outscore their opponents by the widest margin. Dividing the teams into three batches, here are the biggest losers:
The above columns represent each team's actual winning percentage (Win%), third-order winning percentage (3Ord%), estimated chances of winning their division (Div) and wild card (WC), total chance of reaching the postseason, their pre-season odds of reaching the postseason (Proj.), and the change in those cumulative odds over the course of the first half (+/-). Where the Phillies initially figured to have a 3-in-5 chances at October, they're down to about a 1-in-5 shot. Having swapped out Cliff Lee for Roy Halladay over the winter—oh, what might have been had they kept both—they were favored to take the NL East for the third straight time, but instead they find themselves in third place, 4 ½ games out. Halladay has been among the league's top starters, but no other member of their rotation has a Support Neutral Winning Percentage above .500. The bullpen has suffered greatly with the absences of Ryan Madson and a functional Brad Lidge, while the offense has been merely middling, at one point averaging 2.5 runs per game over an 8-17 stretch. Note that the numbers don't reflect the impact of the absences of Chase Utley and Placido Polanco, both possibly out until September due to thumb and elbow injuries. Forecast to have the game's best record, Boston's chances have fallen below 50 percent. The Red Sox have soldiered on gamely amid injuries, producing the majors' highest-scoring offense while missing two-thirds of their starting outfield as well as Josh Beckett, and while they may soon be whole on those fronts, Dustin Pedroia and their catching tandem of Victor Martinez and Jason Varitek are now on the shelf. Elsewhere, the Mariners put up an overdue white flag with Friday's trade of Lee, though the blame lies in their inept offense and not their excellent rotation. The Diamondbacks' hopes died a cruel death when Brandon Webb failed to return from his shoulder injury in a timely fashion; he's yet to throw a pitch, but the team has already celebrated Snake Whacking Day. On to the biggest gainers:
Among the happy stories, none is happier than the Padres. Projected for 73 wins during a rebuilding season, they instead have a two-game lead in the NL West and a better-than-even shot at making the playoffs. Their offense is nothing to write home about even after adjusting for the hitters' graveyard that is Petco Park; their .252 True Average ranks 15th in the league, and Adrian Gonzalez (.322) is their only premium hitter. On the other hand, both their rotation and bullpen have been the league's elite according to our win expectancy-based stats, with 22-year-old Mat Latos emerging as an ace in the making, Clayton Richard and Wade LeBlanc putting up surprisingly strong first halves, and Heath Bell ranking as the league's most valuable reliever, with set-up man Mike Adams fifth. Similarly, the Reds, who were forecast for 77 wins, have risen up to snatch first place in the NL Central, where the gap between the Cardinals and the rest of the pack appeared to be larger than that of any other division. Joey Votto has emerged as an MVP candidate for the Reds, Scott Rolen is enjoying a strong rebound, Jay Bruce is finally delivering on his blue-chip promise, and youngsters Johnny Cueto and straight-to-the-majors rookie Mike Leake have anchored the rotation. Elsewhere, the Rangers have put themselves in a commanding position in an AL West race with no overwhelming favorite as of opening day, with Josh Hamilton and Vladimir Guerrero rebounding impressively from subpar 2009 showings. Their acquisition of Lee to prop up an underperforming rotation is as much about playing deep into October as it is getting there. The Yankees, who failed to land Lee just when they appeared to have sewn up a deal, have overtaken the Red Sox and Rays to finish the first half with the majors' top record. Andy Pettitte, CC Sabathia and Phil Hughes have formed a strong big three at the top of the rotation, and Robinson Cano and Nick Swisher have picked up the slack for an offense elsewhere affected by injuries, age, and subpar performances by Mark Teixeira and Curtis Granderson. The Braves have filled the void atop the NL East thanks to timely comebacks by Troy Glaus, Tim Hudson and Billy Wagner, not to mention the emergence of rookie Jason Heyward. Finally, the teams which have more or less held their ground:
Aside from the Angels and the Cardinals, these teams fit into two categories: contenders in crowded fields who haven't lost hope yet (the Dodgers and Twins) and non-contenders whose status as such has been confirmed. The Halos, whose unimpressive pre-season forecast drew plenty of ire given that they'd won four of five previous AL West flags, climbed back into the race with an 18-9 June but still have a negative run differential. The Cards entered the year with the highest odds of any NL team due to the projected gap between them and the rest of the NL Central field. Their odds stood at 85 percent on June 5, and while they've lost considerable ground, they still rate a better-than-even shot. Remember that teams can move the needle by more than 20 percent with a good or a bad week. Projections are not destiny, and for most of these teams, it's not too late to rewrite the future.
A version of this story originally appeared on ESPN Insider
|
Again with the 'Cliff Lee will really help them come playoff time!' shtick.
Nate Silver's research showed there's nothing/nothing/nothing special about starting pitching in the postseason. At all. No different than adding a big bat to the team, or filling any position player hole.
Really, if your very own research hadn't debunked this, I'd stop haranguing about it. Which I'll stop now anyway. The end.
If you're referring to the Secret Sauce, remember that it's much closer to a rule of thumb than a bulletproof formula.
Anyway, the three relevant things Nate found to be relevant were staff strikeout rate (EqSO9), Closer WXRL and team defense via FRAA. Add Lee (6.5 EqSO9 in 2010) to the Phillies rotation in place of Kyle Kendrick (3.7 EqSO9) over the course of a full season and that's going to significantly improve the team's standing in that department and thus in the Secret Sauce formula.
I am still curious why you think Secret Sauce says Cliff Lee wouldn't be helpful in a playoff series.
Richie,
I am confused as to where you come up with the idea that starting pitching is not a predictor of success in the playoffs. In the "Secret Sauce" article, Nate Silver states that there are 3 predicotrs of success:
"A power pitching staff, as measured by normalized strikeout rate.
A good closer, as measured by WXRL.
A good defense, as measured by FRAA"
Lee would classify as a power pitcher, with an above average strikout rate. He would certainly improve the Rangers chances come playoff time, because he is likely replacing a pitcher with a lower normalized strikout rate. The difference may be small, but there is still a difference.
Don't forget that right now, Lee replaces Omar Beltre in the rotation. Beltre is a nice story, but his 3 innings per start schtick got old fast. likely Beltre was done anyway, with Harrison going back into the rotation, but, like figureing out an unearned run from reconstructing the inning, deciding which replacement level pitcher they replaced with one of the 5 best starters in baseball is a huge upgrade to the team. Even if its not a specific upgrade for the playoffs, it still makes the team better, and that will help in the playoffs.
If Rich Harden gets healthy and effective or Derek Holland can come back and be effective, then you push Scott Feldman out and your in even better shape. A playoff rotation of Lee, Harden, Wilson, Lewis, and Hunter should scare a lot of teams, especially with a pen which is chock full of guys who strike out more than a batter per inning and allow much less than a hit per inning (like Oliver: 25 hits, 43 k's, 39.2 innings, or Feliz: 27 hits, 42 k's, 37.2 innings)
Lee is a definite improvement atop the rotation, and certainly helps their cause towards reaching October. But but as I said all over Twitter last Friday, the Rangers' playoff rotation doesn't exactly scare me. He's the obvious #1, and while I like Lewis, I'm not sure he's a legitimate #2 and worry about his innings jump (176 in Japan last year, 204 pace this year prior to postseason), and I don't see anybody else in their pool who is a #2 based upon current health and performance.
The guys in that pool who don't miss bats (Hunter, Feldman) aren't great bets - just look at the latter's regression to the mean this year after his big 2009, or the fact that the former had a 2.00 ERA through six starts last year and a 5.09 mark the rest of the way.
Meanwhile, the ones who do miss bats are the ones with the high walk rates that have caused them to depart early and force the Texas bullpen to throw more innings than any other AL team. Wilson's going to run into an innings problem, Harden's got a 4.59 ERA and 4.8 BB/9 while averaging 5.3 innings per start over the last two seasons AND missing about 1/3 of his starts.
There's plenty of upside in the Rangers' rotation from a long-term standpoint. I'm just not at all convinced they can tap into it in time for October 2010.
Sometimes it is best to just look at odds of winning a baseball game. How much does having a solid starter increase your odds of winning a baseball game? Answer that question, compute the percentages, and then move the Secret Sauce stuff around at the margin to bump things up or down by a couple percentage points. Being better at baseball than one's opponents doesn't need a study to confirm it. Nate found 3 interesting things that shined above the noise, but that doesn't mean that the same things that make one team have a more than 50% chance of winning a game won't be important, too. You still need to assume the better team wins a baseball game more often than not.