NHLNumbers Predictive Power Rankings, 3/3/13

Derek Zona
March 03 2013 09:39AM

Ray Emery
Photo by Leech44, via Wikimedia Commons

We're 40% of the way through the lockout-shortened 2013 season and three things are certain - Chicago, Anaheim, Boston and Montreal are in the playoffs; Columbus and Florida aren't; and the New York Islanders would be in the playoffs even if they had just average goaltending, like Minnesota.  

These rankings aren't presented as a look at the current standings or last couple of weeks' worth of performances.  Our goal is to build a predictive model that gives us a glimpse into the season-ending standings and first-round playoff matchups.  Our rankings aren't based solely on the current NHL standings, though points earned are a significant portion of the model, they are based on a number of underlying metrics so the rankings are constantly in flux. This week we've added remaining opponents in the hopes of building a better model.  Many have asked about how injuries are accounted for in the predictive model.  They aren't included - yet. 

We're still tweaking that model, so we're not quite ready to unveil it, but consider this NHLNumbers' current best effort at predicting the standings and playoff pairings at the end of the season.  One item of note - 38% of the game is luck, and we don't attempt to predict or model that 38%, and we don't plan to.

The second edition of the NHLNumbers Predictive Power Rankings:

Rank Team NHLNShare
1 Chicago Blackhawks 2.02
2 Boston Bruins 2.05
3 St. Louis Blues 4.92
4 Montreal Canadiens 5.13
5 Pittsburgh Penguins 5.81
6 Vancouver Canucks 5.90
7 San Jose Sharks 6.13
8 Ottawa Senators 7.22
9 Anaheim Ducks 7.48
10 Los Angeles Kings 8.00
11 New Jersey Devils 9.75
12 New York Rangers 11.15
13 Detroit Red Wings 11.65
14 Carolina Hurricanes 11.66
15 Toronto Maple Leafs 12.06
16 Phoenix Coyotes 12.58
17 Dallas Stars 13.01
18 Philadelphia Flyers 13.38
19 Minnesota Wild 13.87
20 Edmonton Oilers
13.91
21 Winnipeg Jets 15.89
22 Calgary Flames 16.17
23 Colorado Avalanche 16.63
24 New York Islanders 16.64
25 Nashville Predators
16.73
26 Washington Capitals 18.15
27 Tampa Bay Lightning 18.64
28 Florida Panthers 20.38
29 Buffalo Sabres 21.15
30 Columbus Blue Jackets 22.36

Based on these rankings, the predicted playoff matchups look like this:

1 Chicago Blackhawks
1 Boston Bruins
8 Phoenix Coyotes
8 Toronto Maple Leafs





4 St. Louis Blues
4 Montreal Canadiens
5 Anaheim Ducks
5 Ottawa Senators










2 Vancouver Canucks
2 Pittsburgh Penguins
7 Detroit Red Wings
7 New York Rangers





3 San Jose Sharks
3 Carolina Hurricanes
6 Los Angeles Kings
6 New Jersey Devils

These other 14 teams will be in the running for the #1 pick overall:

17 Dallas Stars
18 Philadelphia Flyers
19 Minnesota Wild
20 Edmonton Oilers
21 Winnipeg Jets
22 Calgary Flames
23 Colorado Avalanche
24 New York Islanders
25 Nashville Predators
26 Washington Capitals
27 Tampa Bay Lightning
28 Florida Panthers
29 Buffalo Sabres
30 Columbus Blue Jackets

E976572df3ea71edf6bafb00adc1c546
Avatar
#1 Ty
March 03 2013, 01:49PM
Trash it!
0
trashes
+1
0
props

I have a question, does power play PDO regress to the mean like Even Strength PDO? Would a team with a penalty kill rate of 90% regress towards 81-82%?

Avatar
#2 chinook
March 03 2013, 04:46PM
Trash it!
0
trashes
+1
0
props

Hmmm ... interesting and plausible, though as you haven't factored in injuries we should still play out the rest of the season. Also, I agree Anaheim is playing over their heads - because Getzlaf is after a new contract? (can you factor that in?) - and may revert to form soon. But my guess is LA will reach top of the Pacific, not San Jose.

Avatar
#3 garik16
March 03 2013, 06:41PM
Trash it!
0
trashes
+1
0
props
Ty wrote:

I have a question, does power play PDO regress to the mean like Even Strength PDO? Would a team with a penalty kill rate of 90% regress towards 81-82%?

In theory Power Play PDO should regress to the mean eventually.

But remember how tiny the sample size is for special teams - it's so tiny that there's no reason to expect regression to necessarily occur over even an 82 game season.

Avatar
#4 Ty
March 03 2013, 06:48PM
Trash it!
0
trashes
+1
0
props

Garik16: Then what differentiates the best special teams from the worst? Is it the ability to draw and avoid penalties?

Avatar
#5 Ralph
March 03 2013, 11:35PM
Trash it!
0
trashes
+1
0
props

@Ty

It seems like shots + missed shots for rate is the best predictor of future PP%, while goals against rate is the best predictor of future PK%. Source: http://objectivenhl.blogspot.com/2011/05/loose-ends-part-iii-c-power-play.html

Drawing/avoiding penalties is no doubt important, too. Montreal has been on the power play 31 more times than Boston already, and that's probably an extra 5 goals or so--nearly one win in the standings thanks to drawing PPs alone.

Comments are closed for this article.