Eastern Conference Finals are set...

Now that Buffalo and Ottawa have made it to the Eastern Conference Finals, I wanted to check in how my predictive modeling performed compared to actual results. Obviously, it wasn't too far off the mark, as I did pick both the Sabres and Senators to make it through.

Just to recap, the basic idea behind this model is that I try to map out each team's offense against the opposing defense, projecting shots/game from various distances, shooting percentages, and thereby goals per game. The two key concepts are, first, an interaction between one team's offense and the other's defense (both in terms of shots allowed and goaltending), and secondly, trying to reflect recent performance by the use of exponential moving averages to represent the per-game values*.

Bufallo/New York: I was hardly alone in picking the Sabres here, as they were the top team in the regular season, although the Rangers did put up a good fight, and if they had successfully handled the final minute of Game Five, who knows what might have happened as a result. The story here appears to be one of Buffalo getting too many good scoring opportunities for Henrik Lundqvist to stop - from medium and close range, shot totals for Buffalo were well above expectations, and while New York held the overall shooting edge, as expected that numerical advantage was mostly due to outside shots, which are less dangerous.


Here's a look at Jochen Hecht's series-clinching goal for Buffalo.


New Jersey/Ottawa: I called this one in six games for the Senators, based on an impression from the numbers that goaltending was not going to be a significant edge for New Jersey (as most assumed), and a likely advantage for Ottawa in terms of both overall shot totals, as well as shots from within 29 feet. Pretty much all of that played to form, and while the level of goal-scoring was higher than predicted, the expected gap between the teams (0.47 goals per game in favor of Ottawa) was right on the money. One interesting point to be seen here is that while the Senators generated lots of close-in shots, they scored only 0.4 goals per game from the 10-19 foot range, which is far-and-away the most common scoring area. Instead, they consistently scored from all over the place.



Here's how the Sens got it done in Game Five.

So far, I'd say this method of modeling playoff matchups is working fairly well - while the exact numbers are often off the mark, the differences between teams are showing up quite clearly, and seem to be useful in predicting which team is likely to win. I'll take a look at the Anaheim/Minnesota matchup tomorrow (that's one that didn't work out as expected), and Detroit/San Jose whenever that finishes up. Since the Eastern Conference Finals won't start until Thursday (updated), I'll wait to do a preview for a couple more days.

*As always, I exclude empty-net goals from this analysis, so take that into account. Basically, empty-net goals have a negligible impact on the outcome of a game, since the team that scores one is already winning with little time left.



X
Log In Sign Up

forgot?
Log In Sign Up

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior users will need to choose a permanent username, along with a new password.

Your username will be used to login to SB Nation going forward.

I already have a Vox Media account!

Verify Vox Media account

Please login to your Vox Media account. This account will be linked to your previously existing Eater account.

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior MT authors will need to choose a new username and password.

Your username will be used to login to SB Nation going forward.

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join On the Forecheck

You must be a member of On the Forecheck to participate.

We have our own Community Guidelines at On the Forecheck. You should read them.

Join On the Forecheck

You must be a member of On the Forecheck to participate.

We have our own Community Guidelines at On the Forecheck. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9355_tracker