The basics of the model are below in table form:
Now this is a little complicated but we can all understand a simple table. As before please note I capped out win probabilities at no better than 90% and no worse than 10%. This was done as the best team ever record wise still managed to lose 9 of 82 games (and the worst team ever still won 11 of 82):
For team A and team B playing in a seven game playoff series ,with A being the higher seed first you need to know the win probability margin of team A beating team B at a neutral site.How you work this out is up to you but the simplest possible model would be using winning percentage. For example,
Team A: .750 Winning Pct.
Team B: .650 Winning Pct.
Margin A-B = .100 (now go look it up)
%TeamA beats TeamB @ neutral site = .100 + .500=.600
Probability of Team A winning a seven game playoff series as the home team= 74.4%
Probability of Team A winning a seven game playoff series as the road team= 68.4%
See, It’s simple.
One of the items that came up as I built the model was the varying significance of Homecourt advantage (and the point of this post actually). You can see this here:
So the math is clear. The more even two teams are facing each other in a seven game series, the more important home court advantage becomes . For dead even teams home court is worth almost half a win. I now feel a little better about the NBA finals.
Note: Short post, I know but I’m working on the Build v1. The minute allocation model may force me to crack open vb or C.