Pythagorean expectation is a formula invented by Bill James to estimate how many games a baseball team "should" have won based on the number of runs they scored and allowed.

I don’t want to get into details of the formula or it’s validity, but I thought it would be interesting to apply the same principle to Hockey in order to predict the playoff winners.

Using goals for and against instead of runs scored and allowed, Here are the EXP W%, (Expected winning percentage) for each team in the league for 2007. Beside it is the ACT W% (Actual winning percentage). Finally the last column compares the results of the season with the expectations of the formula, by substracting the ACT W% to the EXP W%.

From these assumptions, it could be debated that Ottawa, Calgary, NY Rangers are better then their record indicates. Also that Dallas, Vancouver, New Jersey and Tampa Bay overacheived or got more breaks than they derserved.

While any team can win or get far in the playoffs, reasearch show that teams within the top EXP W% have a much greater chance of capturing the final prize.

For example, I haven’t done the exercise for hockey, but in NFL Football, only one team out of the top 5 EXP W% was won the Superbowl in the least ten years.

BTW, I'm not commenting on the firing of Claude Julien, but if you look at the EXP W% of New Jersey, they are 13th in the league. Also they have the highest differential between EXP W% and ACT W% (or LUCK)

it's 2 for baseball, but that's an empirical result, not a theorical one. It could very well be different for hockey.

Thanks for the wikipedia link, it is a pretty good read.

Good point, I used 2 as a general guideline. I tried quite a few others but 2 seemed to give quite realistic results. I know for baseball the 1.81 is more precise, but if you saw how much trouble they went into calculating that, you probably would do like me and stick to 2, although something other then 2 might very well be more accurate.

Pythagorean expectation is a formula invented by Bill James to estimate how many games a baseball team "should" have won based on the number of runs they scored and allowed.

I don’t want to get into details of the formula or it’s validity, but I thought it would be interesting to apply the same principle to Hockey in order to predict the playoff winners.

Using goals for and against instead of runs scored and allowed, Here are the EXP W%, (Expected winning percentage) for each team in the league for 2007. Beside it is the ACT W% (Actual winning percentage). Finally the last column compares the results of the season with the expectations of the formula, by substracting the ACT W% to the EXP W%.

From these assumptions, it could be debated that Ottawa, Calgary, NY Rangers are better then their record indicates. Also that Dallas, Vancouver, New Jersey and Tampa Bay overacheived or got more breaks that they derserved.

While any team can win or get far in the playoffs, reasearch show that teams within the top EXP W% have a much greater chance of capturing the final prize.

For example, I haven’t done the exercise for hockey, but in NFL Football, only one team out of the top 5 EXP W% was won the Superbowl in the least ten years.

For your entertainment only. Let the games begin.

Cheers.

Wait a minute! Isn't pythagorus a lazy Russian?? You can't post that here.

Vancouver and the Devils are "lucky"... you know why?

Could it be because Luongo and Brodeur can steal some games now and again?

Yes because of the goalies, and also because their defensive styles do not give very high GF - GA differentials, which are the basis of these assumptions.

On the other hand, Tampa Bay have a luck factor around the same as Vancouver & NJ (higher than Vancouver, lower than NJ). Definitely not because of goaltending or defensive systems there.

You should've lengthened the 'excuse me'. Steve Martin started that one on SNL didn't he ? My Mom still has no idea why whenver she gives me something to drink, I say 'Good Tang, Mrs. Lubner'. Then again,she stopped wondering about a lot of things many years ago.