Friday, October 03, 2008

And The Cherry-Picker Of The Year Is...

This guy.

If you're going to define a prediction as "accurate" based on the results, I think it should have to fit a couple of simple criteria: it suggests a higher chance of a team winning than the public thinks, and that team goes on to win.

Here's what I don't call accurate: Forecast five different things, then gloat when only one of them comes to fruition.

Take a look at this prediction, and see if you can spot the part where they say the Phillies are a good bet.

Did you miss it? I'm not surprised. It's tucked away at the end, long after they pimp the Brewers not one but four times:

"For game one, we said the Phillies were big favorites, having won 676 of our 1,000 sim runs. We just finished 1,000 runs of game 2 on our Diamond Mind simulation, and The Sim expects different results today. With C.C. Sabathia on the mound, that's no surprise, really. But the key might be whether repeatedly pitching on 3-days' rest might start to catch up to him, and he threw 122 pitches on Sunday. With that in mind, we ran our simulation 3 ways:

1. Sabathia is not significantly affected by the short rest (something he showed over the past couple of weeks).
In this scenario, the Brewers win 59.5% of the 1,000 simulation runs we did, scoring an average of 5.0 runs compared to 3.8 for the Phillies.

2. He is affected by the short rest. We ran the simulation 2 ways, one where the fatigue factor assumed is mild and one we'd project as normal for him. Using "mild" fatigue, Milwaukee wins 7 fewer games out of 1,000 sim runs of today's game, reducing their projected odds of victory to 58.8%. However, when we used "normal" fatigue, based on Sabathia's pitcher durability ratings in the 2008 version of our Diamond Mind Online game, the Brew Crew won only 528 times out of 1,000 sim runs.

Keys to the Game: Sabathia's durability.

And, just as getting on base was the key for Philadelphia in game one (as The Sim predicted), not getting on base looks like it will be the Phils' downfall in game 2. Sabathia's control is critical here - if he is tired and does not have his typical excellent control, the Brewers could be in trouble.

If you're looking for likely hitting stars, the sim points to surprisingly good odds that Milwaukee infielders Craig Counsel and Ray Durham will have a good day, each scoring a number of multi-hit games, while the Phillies might pitch around Prince Fielder, who shows a high number of walks.
Philadelphia starter Brett Myers does not fare well in many of the sim runs, and his bullpen mates fare even worse, indicating that this could turn into a Brewer rout by the end.

In any case, the key point of leverage here is Sabathia - if he's on, the Brewers will likely pull even. But if the heavy pitching load he's been carrying lately starts to catch up with him, this odds becomes much closer. The key might be his pitch count - watch his control, especially in the middle to late innings. And if the Phillies get to the Brewer bullpen early, they become the favorites, according to The Sim.*

Oh, and did we say Sabathia's durability is a key factor?"

*If you couldn't figure out on your own that the Phillies were the favorites if Sabathia got knocked out of the game early, go unsubscribe from this blog right now.

If you're counting at home, that's five predictions, four of them inconsistent with the game's outcome. Guess which one they highlighted on the summary page?


They needed five predictions to get one right; I bet I can do the same for Game 3 of this series:

1. Philadelphia will win a blowout
2. Philadelphia will win a close game
3. Milwaukee will win a blowout
4. Milwaukee will win a close game
5. Bud Selig declares the game a tie

Hooray for accuracy!

No comments: