Round Two Review For The West

by Dirk Hoag on May 8, 2007 2:32 PM EDT

With Detroit and Anaheim pushing their way into the Western Conference Final, it's time to recap the predictions made here a couple weeks back, to see how well they performed:

Detroit/San Jose: I picked Detroit (albeit in 7 games), based on an expected advantage of 6-7 shots per game, and strong goaltending from Dominik Hasek. That's pretty close to how things actually turned out, as the Red Wings outshot the Sharks by 7.7 shots per contest and Hasek (despite an exciting trip or two behind the net) generally outplayed Nabokov.

All the highlights from Detroit's series clinching win Monday night.

Anaheim/Vancouver: This was my lone miss of the second round, as I had called for a Canucks victory, since the Ducks weren't exactly lighting up the scoreboard as of late and faced the outstanding Roberto Luongo in net. While Luongo held up his end of the bargain, the Vancouver defense (ravaged by injury) gave up too many shots, particularly close-range shots, to consistently keep the Ducks down. Although the actual per-game numbers look quite high compared to expectations, remember that due to overtimes, the teams played about 20% more minutes than you'd usually have across five games, so that accounts for a good portion of the difference. The real deciding factor in this series, however, is the shooting percentages for Vancouver - did their collective sticks go cold, or did J.S. Giguerre deserve as much credit as is heaped on Luongo? I guess we'll see based on how Giguerre handles the Red Wings in the conference finals...

Here's how the Ducks finished off Vancouver in Game Five.

So far, this method has resulted in a 9-3 predictive record (6-2 in the 1st round, 3-1 in the second). Check back tomorrow for the Conference Final predictions...

Tweet Comment 2 comments | 0 recs |

Do you like this story?

Comments

So far, this method has resulted in a 9-3 predictive record (6-2 in the 1st round, 3-1 in the second).

I prefer to think of the breakdown as 9-1 when picking series that Anaheim is not involved in; 0-2 when picking against the Ducks.

You'd probably have been better if you ignored the games Anaheim played without Giguere, Pronger, and Beauchemin. Those games really warp the team's post-Christmas statistics.

by Earl Sleek on May 8, 2007 5:23 PM EDT reply actions

Yeah, you could make the argument that there's some "Special Case" adjustments that need to be made for Anaheim given the coincidence of three major players being injured all at one time, but there are a couple factors just keeping within the method that I'm sure will warrant some fine tuning over the summer:

1) I picked a weighting coefficient of 0.97 out of thin air - backtesting this method against results from this season and the previous one will help me hone in on a more applicable figure, and
2) The moving averages are made up of raw performance data, unadjusted for the quality of opposition. Ideally whether you racked up 5 goals on L.A. or Minnesota would be reflected within the representation of a given team's offense.

For now, I'm pretty happy with how it's worked, in the sense that I think it's worth fleshing out to a further degree during the offseason. Picking the Detroit/Anaheim series will be tough, as Anaheim is outperforming relative to predictions, and with Detroit now losing Mathieu Schneider, they're likely to see some downside that's not reflected yet.

by The Forechecker on May 8, 2007 6:36 PM EDT reply actions