Using a logistic regression model and adjusting for the fact that Yankees and Phillies will not use some of their regular season pitchers the Yankees have a 65% chance to win series and Phillies a 35% chance. Here are more details:
- 9.6% of time Yankees in 4
- 14.5% of time Yankees in 5
- 21.7% of time Yankees in 6
- 19.2% of time Yankees in 7
- 3.4% of time Phillies in 4
- 9.8% of time Phillies in 5
- 10.4 of time Phillies in 6
- 11.3% of time Phillies in 7
The simulations were done with Palisade’s Monte Carlo Excel -add-in @RISK.
Interesting stuff. Any background on how you came up with these numbers?
Comment by Jeremy — October 26, 2009 @ 12:56 pm
Rated teams based on a logistic regression model using all games. then played out world series 10,000 times factoring in home edge using @RISK.
Comment by wayne winston — October 26, 2009 @ 4:01 pm
So your logistic regression model gives you a probabilty of a team winning a game based on a bunch of factors (the X’s)? How many predictor variables do you have in that model? (I guess starting pitching is one variable?) I guess the point is it’s more accurate than just calculating a flat winnng probability that could be applied for all games in the World Series and using a binomila distribution to predict outcomes. Interesting stuff.
Comment by Steve — October 31, 2009 @ 12:05 pm
chance of winning is different each game because of home edge if nothing else. basically i use chess system adjusting for improvement in starting pitching due to shortening of rotation
Comment by wwinston — November 4, 2009 @ 9:41 pm