Oct 312006

Hockey is, maybe more than any other sport, a team sport. A hitter in baseball can be evaluated based on a series of one on one battles with pitchers which are largely independent of the ability of the hitters teammates. In hockey this is not the case. Jaromir Jagr may be a great goal scorer but if he didn’t have quality teammates around him the number of goals he scores would be significantly impacted. A great defensive play by Marek Malik causing a turnover in the defensive zone followed by a great breakout pass might be just as important in scoring a goal as Jaromir Jagr being in the right spot at the right time to chip in a rebound. Conversely, a good forcheck by Peter Forsberg might be just as important for keeping the puck out of his teams net as a poke check by Antero Niittymaki taking away a shot and a scoring opportunity. As some say, the best defense is a good offense. But there is no stat that “adequately” rewards either Malik’s great defensive play resulting in an offensive opportunity nor Forsberg’s great offensive play limiting the number of offensive opportunities against.

I say “adequately” because +/- attempts to do that and though it has some use as a statistic it is seriously flawed because it doesn’t isolate a players worth independent from his teammates. Conceptually it is a good statistic. Winning games is all about scoring more goals than you give up and +/- measures how many goals are scored by your team while you are on the ice in relation to how many goals are scored by the opposition while you are on the ice. Problem is, if you play on a really good team you are likely to have a decent +/- regardless of how good or bad you actually are. Conversely, if you play on a really bad team you are likely to have a poor +/- regardless of how good or bad you really are. Was Vaclav Varada at +21 really significantly better than Sidney Crosby at -1? No, of course not. There is a reason why Varada is no longer in the NHL.

So the question is, how best can we isolate a players individual ability in such a team oriented sport like hockey. Is it even possible? I believe that this can be done and I believe that I have developed an algorithm that does so adequately. Tomorrow I will describe in more detail what I have done and present some player rankings from last season and Thursday I’ll post this seasons player rankings. I am sure the results with be both surprising and controversial but most of all I hope that are thought provoking and make you think about certain players a little differently.

NHL Power Rankings 10/31/2006

 Uncategorized  Comments Off on NHL Power Rankings 10/31/2006
Oct 312006

Here are this weeks NHL Power Rankings. As one might expect, the western conference teams still dominate the top rankings. Clearly the western conference has been better so far and has more top teams than the east.

Rank Last Week Team AdjWinP SchedStr Power Rank
1 3 Anaheim 0.818 0.571 0.861
2 2 Minnesota 0.682 0.602 0.858
3 8 Detroit 0.591 0.583 0.765
4 4 Dallas 0.773 0.543 0.757
5 1 San Jose 0.667 0.548 0.733
6 12 Buffalo 0.818 0.454 0.717
7 6 Vancouver 0.583 0.562 0.678
8 5 Edmonton 0.636 0.551 0.671
9 10 Colorado 0.455 0.611 0.600
10 13 Montreal 0.600 0.455 0.540
11 9 Atlanta 0.667 0.428 0.539
12 11 Washington 0.450 0.533 0.520
13 20 Carolina 0.458 0.475 0.484
14 25 Pittsburgh 0.667 0.417 0.472
15 24 New Jersey 0.545 0.421 0.461
16 17 Chicago 0.400 0.536 0.436
17 21 Ottawa 0.450 0.463 0.436
18 26 Nashville 0.500 0.480 0.434
19 7 Los Angeles 0.269 0.650 0.409
20 18 NY Islanders 0.350 0.517 0.394
21 16 Tampa Bay 0.409 0.471 0.391
22 19 Toronto 0.458 0.443 0.388
23 14 Florida 0.462 0.414 0.361
24 22 Columbus 0.333 0.528 0.348
25 27 Boston 0.389 0.441 0.338
26 23 Calgary 0.333 0.491 0.326
27 29 NY Rangers 0.400 0.410 0.313
28 15 St. Louis 0.278 0.552 0.303
29 28 Phoenix 0.250 0.521 0.261
30 30 Philadelphia 0.200 0.449 0.204

AdjWinP is a teams winning percentage when shootouts are considered ties and there are no points awarded for overtime losses
SchedStr is an indication of a teams relative difficulty of schedule
Power Rank is the teams expected winning percentage if team played all .500 teams

Game Predictions 10/31/3006

 Game Predictions  Comments Off on Game Predictions 10/31/3006
Oct 312006

Four games on schedule tonight with a couple of pretty interesting ones. The Ottawa-Montreal game is on I think is worth watching. Ottawa got off to a pretty slow start to the season scoring 10 goals in their first 6 games, then they went on fire scoring 21 goals in 3 games only to score 1 goal against Boston on the weekend. Which team is going to show up tonight? The other interesting game is out west where Nashville visits Vancouver. I am a bit surprised that Vancouver is listed as a strong favourite because Nashville had been outstanding after a slow start. They lossed their 3 first games but have been 6-0-1 since though their one overtime loss during that span did come at the hands of the Canucks. We’ll see what happens but I think this game could go either way.

Home Team Fair
Road Team Fair
Predicted Winner Confidence
NY Islanders -188 Chicago 146 NY Islanders Strong
Montreal -145 Ottawa 131 Montreal Good
Florida 140 San Jose -167 San Jose Strong
Vancouver -202 Nashville 150 Vancouver Strong
Oct 302006

Five games on the schedule tonight with the algorithm picking the road team in 4 games. Should be interesting to see how those predictions pan out as home teams generally win more games.

Home Team Fair
Road Team Fair
Predicted Winner Confidence
Philadelphia 143 Chicago -175 Chicago Good
Toronto 112 Atlanta -114 Atlanta Some
St. Louis 157 Anaheim -233 Anaheim Strong
Calgary 123 Washington -130 Washington Good
Los Angeles -158 NY Rangers 137 Los Angeles Good

Game Predictions – 10/29/2006

 Game Predictions  Comments Off on Game Predictions – 10/29/2006
Oct 292006

Just two games on the schedule today. The first is an interconference game featuring the San Jose Sharks and the Tampa Bay Lightning and the seond is a divisional game featuring the Minnesota Wild and Colorado Avalanche. Like the algorithm predicts, I like San Jose and Minnesota.

Home Team Fair
Road Team Fair
Predicted Winner Confidence
Tampa Bay 155 San Jose -225 San Jose Strong
Colorado 137 Minnesota -159 Minnesota Strong

NHL Game Predictions – 10/28/2006

 Game Predictions  Comments Off on NHL Game Predictions – 10/28/2006
Oct 282006

In my opinion the biggest game of the night is Atlanta at Buffalo. In my mind these have been the two best teams in the east so far this year as Buffalo is 10-0-0 (3 shootout wins) and Atlanta is 7-1-3 (2 shootout losses, 1 win). Both teams feature top tier young NHL goalies and underrated somewhat nameless (unknown) defenses. Both teams can score goals too with Atlanta having more top end talent with Kovalchuk and Hossa and Buffalo having more depth. Should be a fun game to watch and a good test game for both teams, particularly Atlanta who still needs to prove themselves, the battle of Pennsylvania this time where Pittsburgh is probably the better team over Philadelphia, .

In other interesting games are have a Toronto-Montreal rivalry game, Zdeno Chara facing his old Ottawa team for the first time, and the battle of Pennsylvania with Pittshburgh the better team than Philadelphia for probably the first time in a long long time (maybe a decade). The only real interesting western conference matchup is Nashville at Calgary. Both these cup contenders got off to slow starts to the season and both are just start and while Nashville has broken their early season slump Calgary is still looking to break out and put a few wins together. A big win over Nashville could go a long way to making that happen.

Home Team Fair
Road Team Fair
Predicted Winner Confidence
Boston 142 Ottawa -172 Ottawa Good
Buffalo -148 Atlanta 132 Buffalo Good
Montreal -138 Toronto 127 Montreal Some
NY Islanders -120 Florida 116 NY Islanders Some
Philadelphia 149 Pittsburgh -196 Pittsburgh Good
Carolina -131 Tampa Bay 123 Carolina Some
New Jersey -175 Columbus 143 New Jersey Strong
St. Louis 131 Detroit -145 Detroit Good
Edmonton -198 Washington 149 Edmonton Strong
Dallas -201 Los Angeles 150 Dallas Strong
Chicago 128 Anaheim -140 Anaheim Strong
Calgary -103 Nashville 102 Nashville Some
Phoenix -126 NY Rangers 121 Phoenix Some

NHL Predictions 10/27/2006

 Game Predictions  Comments Off on NHL Predictions 10/27/2006
Oct 272006

It’s a western conference night tonight as the only eastern conference team in action is the Washington Capitals who are in Vancouver to face the Canucks. This should be a fun game to watch as Washington is off to a pretty good start (3-2-3) and seemingly aren’t the horrible team they were last year. Two of the other games feature four of the top teams in the west as Detroit visits Dallas and Anaheim visits Minnesota. The other game features a weak Los Angeles team facing a Columbus team who many expected to be better this season but is off to a slow start.

Home Team Fair
Road Team Fair
Predicted Winner Confidence
Columbus 127 Los Angeles -137 Los Angeles Good
Minnesota -119 Anaheim 116 Minnesota Some
Dallas -144 Detroit 130 Dallas Good
Vancouver -148 Washington 132 Vancouver Good
Oct 262006

We have 8 games on schedule tonight. The games I think will be the most interesting are:

Toronto at Ottawa – This game should be a fun game to watch. Toronto will be seeking revenge for a 6-2 loss on Tuesday while Ottawa will be seeking revenge for a Darcy Tucker attack on Patrick Eaves and a Chad Kilger stick attack on Christolf Shubert. Speaking of Tuesdays games, whats with the refs? I enjoyed the added physical play but there were tons of partial hooks and trips that would have been called for penalties in most other games. Not that I am complaining because I think some of this crackdown on obstruction has gone too far.

Atlanta at Philadelphia – Atlanta clearly has been the better team so far this season while Philadelphia has been a disaster. But most teams respond very positively for the first few games after a coaching change plus Atlanta played last night so this game coule end up going either way. The keys to Philadelphia will always ben goaltending and Peter Forsberg. If the Flyers get good goaltening and Forsberg is healthy and play at the top of his game the Flyers will do well. If either of those guys fail there just isn’t enough else there to allow them to win consistantly.

San Jose at Nashville – Yesterday I wrote that San Jose was a big test for Detroit to determin whether they were an elite western conference team or a step below them. This is kind of the same for Nashville. Nashville has gotten off to a bit of a poor start and this is a chance for them to win a big game against a real tough team and prove that they should be considered contenders as well.

Home Team Fair
Road Team Fair
Predicted Winner Confidence
Boston 120 Montreal -125 Montreal Some
Philadelphia 162 Atlanta -268 Atlanta Good
Ottawa -146 Toronto 131 Ottawa Good
New Jersey -136 Florida 126 New Jersey Good
NY Islanders 127 Buffalo -137 Buffalo Good
Tampa Bay -143 Carolina 130 Tampa Bay Good
Nashville 145 San Jose -184 San Jose Strong
Phoenix 154 Edmonton -222 Edmonton Strong
Oct 252006

I debated whether I should post these power rankings or not because on the surface they don’t seem to make any sense. How could 9-0 Buffalo be ranked 12th overall? But finally I decided that I would post them with some explanation of what we are seeing.

First off, you will notice that the western conference teams dominate the top 10 or so positions. The reason for this is because the western confernce has a dominant record (14-5-3) against the eastern conference and since strength of schedule is a significant factor in the algorithm to calculate these power rankings most of the western conference teams get a big boost. Although I do see the western conference as being the better conference it is unlikely that they will continue to win at the same rate versus the eastern conference so over time things should settle down a bit.

So with that said, it is best to consider the power rankings as how eastern conference teams rank against eastern conference teams and how western conference teams rank against other western conference teams.

Now, some of you migh tthen point out that Atlanta and Washington are ahead of undefeated Buffalo in the standings and that doesn’t make sense either. Well, Buffalo has 3 shoot out wins which the algorithm considers as ties so Buffalo’s record is, for the purposes of calculating these rankings, 6-0-3. Atlanta’s record is 7-1-1 so in 9 games they have an equal number of points. So the difference comes in the fact that Atlanta has played a tougher schedule. As for Washington, they are getting a huge boost from playing one of the toughest schedules of any eastern conference team largely because they have played Minnesota once and Atlanta twice and going to overtime or a shoot out in all 3 of those games.

So I guess my final word of advice is to take these power rankings with a grain of salt. While there is some useful information held within them it is still pretty early in the season for them to be real accurate. That said, the prediction algorithm, which is loosely based on these rankings, has performed fairly well so they might not be completely out of whack.

Rank Last Week Team AdjWinP SchedStr Power Rank
1 5 San Jose 0.778 0.608 0.972
2 4 Minnesota 0.750 0.643 0.969
3 2 Dallas 0.833 0.580 0.891
4 8 Anaheim 0.812 0.603 0.862
5 7 Edmonton 0.750 0.598 0.768
6 1 Vancouver 0.550 0.624 0.760
7 19 Atlanta 0.778 0.437 0.634
8 21 Colorado 0.444 0.638 0.628
9 3 Detroit 0.438 0.616 0.602
10 10 Los Angeles 0.350 0.693 0.586
11 16 Washington 0.429 0.535 0.541
12 6 Buffalo 0.833 0.373 0.522
13 13 Ottawa 0.429 0.470 0.454
14 18 Montreal 0.500 0.426 0.452
15 20 Florida 0.450 0.444 0.427
16 9 Toronto 0.556 0.430 0.418
17 25 Tampa Bay 0.438 0.444 0.391
18 24 Chicago 0.500 0.458 0.382
19 17 Carolina 0.389 0.430 0.379
20 14 New Jersey 0.500 0.387 0.370
21 23 St. Louis 0.312 0.561 0.357
22 12 Columbus 0.286 0.591 0.357
23 27 NY Islanders 0.375 0.497 0.356
24 11 Calgary 0.286 0.564 0.349
25 15 Pittsburgh 0.571 0.366 0.327
26 28 Nashville 0.375 0.455 0.312
27 30 Boston 0.357 0.415 0.288
28 22 NY Rangers 0.375 0.353 0.259
29 29 Phoenix 0.222 0.560 0.253
30 26 Philadelphia 0.188 0.357 0.134

AdjWinP is a teams winning percentage when shootouts are considered ties and there are no points awarded for overtime losses
SchedStr is an indication of a teams relative difficulty of schedule

Game Predictions – 10/25/2006

 Game Predictions  Comments Off on Game Predictions – 10/25/2006
Oct 252006

Some of the games you should keep an eye on tonight are:

Atlanta vs Carolina: These two teams look like they might battle for top spot in the division. Should be a fun game too as both teams are entertaining to watch.

San Jose at Detroit: Detroit needs to figure out if they are a contending team or a step below. Games such as this one against other top western conference teams will help determine which they are.

Edmonton at Anaheim: Rematch of last years western conference finals and the return of Joffrey Lupul to his former home in Anaheim. We all know that Anaheim has the big two defensemen but after that they are somewhat weak. If you watch this game observe whether Edmonton’s depth of talent up front can expose that weakness. In the salary cap era you can’t really have a lot of top end talent as well as a lot of depth so I wonder whether it is better to have more top end talent like the Ducks or more depth like the Oilers. Personally I am leaning to depth (see Carolina last year, Buffalo, etc.).

Home Team Fair
Road Team Fair
Predicted Winner Confidence
NY Rangers 125 Florida -134 Florida Some
Carolina 125 Atlanta -135 Atlanta Good
Detroit 125 San Jose -134 San Jose Strong
Minnesota -205 Los Angeles 151 Minnesota Strong
Chicago 139 Vancouver -166 Vancouver Strong
Colorado -146 Washington 131 Colorado Good
Anaheim -123 Edmonton 118 Anaheim Some