Brandon Webb is off to another great start- 8-0, with a 2.41 ERA. His insane ground ball rate (63.2%) has allowed him to give up only 2 homers over 56 innings.On the surface, this is very similar to 2006, when he started off 8-0 with a 2.18 ERA. In terms of his chances of winning some ridiculous number of games though, two things are different this year.
The first is that this year he's gone 8-0 in 8 starts, while two years ago it took him 11 to reach that mark. This means, assuming he makes 34 starts, he's "on pace" to go 34-0, rather than 25-0. More relevantly, he has 8 wins with 26 starts left, rather than with 23.
The second is Arizona's offense. In '06, they averaged 4.77 R/G. So far this year they've scored 216 runs in 39 games, which works out to 5.54 R/G. They are unlikely to keep up that pace, but the offense is certainly better than it has been in years past. PECOTA had them at 5.07 R/G before the season started, and AEQR has them at 5.03 R/G so far. So, going forward, PECOTA's 5.07 is probably as good a guess as any.
The preseason projections has Webb's ERA at between 3.01 and 3.49, with an average of about 3.35. Adding in 56 innings of 2.41 ERA and a 2.92 FIP so far this year, and 3.25 is probably about what we should expect the rest of the year.
We also have to factor in the bullpen. PECOTA had Arizona's bullpen at a 4.52 ERA- they've exceeded that expectation so far, with a 3.30 ERA in 120 innings. So, we'll put them at 4.40. Webb averages 7.0 innings per start, so that means the Diamondbacks will give up an average of 3.51 runs in Webb's starts. If they score 5.07 R/G, they'll have a winning percentage of .662 in his starts.
He won't get decisions in all of those though. Over the past couple years, he's gotten decisions in 85% of his team's wins. So, finally, we come to what percentage of starts he'll get the W in- 56.3%.
Assuming he makes 26 more starts, here's what we should expect his win output to be:
This is incredible. The most likely scenario is 23 wins. That is a lot. Here are the odds that he hits various milestones:
An 89% chance to win 20 games is pretty nuts. Before the season started, he had about a 45% chance of winning 20 games, so the 8-0 start really helped.25 games is also a realistic possibility, which should come as no surprise. 30 is noticeably less so. If he gets decisions in 85% of his starts, we'd expect him to get 22 more decisions- to get 22 more wins would be pretty incredible. It is possible, though.
With two more wins than anyone else in baseball, Webb has already jumped out to the lead in the NL Cy Young race. Zambrano or Hudson could give him a run for his money, and Peavy's always a threat. Sheets if he stays healthy and the Brewers remember how to score runs. It would be cool if Volquez managed to stay in the race- having a low enough ERA is tough in that park, but he's got a head start with a Gibsonesque 1.12 ERA this far. I'd love to see Lincecum win it but, unfortunately, he plays for the Giants.
Also: BREAKING: Kazmir Signs [BPro Unfiltered]
Photo: AZ Sports Hub.



9 comments:
I mark for any Ben Sheets reference. Thanks.
Also, it is posts like this that keep me coming back three times a day.
There is no way that any pitcher has a 45% chance to win 20 games before the season starts. I urge you to re-check your math if you came to that conclusion.
n=34, p=.563. Prob (X=>20) = 0.45374
Run it through a binomial calculator yourself, that's what you get. Only slightly unrealistic assumption is that he doesn't get hurt.
One of the best pitchers in the league combined with one of the best offenses. Not sure why you think a median of 19 wins is so unrealistic.
The assumption that he doesn't get hurt is more than "slightly unrealistic". Nearly 50% of MLB pitchers had at least one stint on the DL last year (sample size caveat, of course). The point is that pitchers get hurt. A lot. Even relatively durable ones like Webb. At BBTF you mentioned that injury risk was left out, I'd love to see some new calculations including that factor if you have time. Nice work.
65% chance he makes 34 starts, 5% 32, 5% 30, 5% 28, 5% 26, 5% 24, 10% <20 gives 32.4%.
(That's for before the season.)
Where'd you get the 65% from? Seems about right, just wondering if you are basing it on anything. I'm not trying to be critical, I'm genuinely interested.
Thin air.
I considered your previous anecdote, and then gave him some extra credit for averaging 33.75 starts per year over the last four years. And poof, 65%.
9-0 is so much cooler than 8-0
Post a Comment