April 5, 2008

Playoff matchups - probabilities.

vs.68%32%
vs.68%32%
vs.68%32%
vs.100%
vs.100%
vs.68%32%
vs.100%
vs.100%
vs.100%
vs.100%
vs.62%38%
vs.62%38%
vs.100%
vs.100%
vs.62%38%
vs.62%38%


I hope this is reasonably self explanatory. The team in the left most column has a list of teams it will probably play in the playoffs. For example Boston has a 32% chance of playing Montreal in the first round. There is still a lot undecided in the East, but the west is pretty close to decided...

















4 comments:

Anonymous said...

cool little blog post JG.


question - I realize San Jose has been on quite a run, but they seem to be way off their expected GF lately. Did they get a new (possibly blind) play-by-play data recorder, or have they really been that unlucky?

JavaGeek said...

You tell me...

S.J
ROAD: GF: 99, EGF: 121 (+22)
HOME: GF: 105 EGF: 142 (+37)

They can't score on the road either...

There are a few strange games such as 9.6 expected goals vs. Anahiem in a 2-1 win. [+7.6]. 7.1 EG on Roloson with 49 shots [+6.1]

A huge chunk of the +37 came from two games [+15]

They haven't had a regulation loss since Feb. 20...

My view is that S.J is taking dumb shots. They're getting too close to the net and the model is having a hard time evaluating the shots. [There's a steep drop off for shot quality for shot of distance 5-7'] .

When a good team plays against a good goalie they will dominate the game and get significant number of EG, however the goalie will lower the actual goals. So teams who are very good on average should under perform expected goals (but not the the magnitude S.J is).

Earl Sleek said...

I don't think it would change too much, but just for my head, are your DAL-ANA odds reflecting a series that starts in Dallas? Or in a neutral barn?

Also, are you planning on doing these updated percentages this spring? It became a favorite link of mine in last spring's gameday posts.

T.C. said...

Good stuff, Chris.