Feb 032006
 

To win the Stanley Cup, you have got to win in the playoffs and that means beating the best of the rest. That led me to ask the question, who is the best at beating the best?

To answer that question I started with my power ranking calculation based on games through last night. Just for reference, here are the updated power rankings.

Rank Last Week Team AdjWinP SchedStr Power Rank
1 1 Ottawa 0.696 0.511 0.723
2 4 Dallas 0.604 0.522 0.639
3 2 Colorado 0.565 0.538 0.634
4 3 Carolina 0.683 0.476 0.632
5 5 Calgary 0.594 0.518 0.615
6 6 Vancouver 0.575 0.523 0.607
7 8 Buffalo 0.647 0.478 0.595
8 7 Detroit 0.657 0.479 0.594
9 9 Edmonton 0.519 0.528 0.561
10 12 Nashville 0.593 0.484 0.547
11 13 San Jose 0.520 0.517 0.543
12 10 Los Angeles 0.509 0.524 0.542
13 11 Toronto 0.500 0.518 0.538
14 14 Minnesota 0.491 0.525 0.531
15 15 Philadelphia 0.604 0.461 0.525
16 16 Phoenix 0.473 0.534 0.521
17 17 Anaheim 0.490 0.518 0.507
18 20 Boston 0.472 0.508 0.491
19 19 Tampa Bay 0.528 0.473 0.489
20 18 Montreal 0.441 0.519 0.484
21 22 NY Rangers 0.537 0.453 0.453
22 21 NY Islanders 0.433 0.496 0.448
23 23 Atlanta 0.462 0.474 0.433
24 24 New Jersey 0.472 0.469 0.433
25 25 Florida 0.396 0.482 0.383
26 26 Columbus 0.370 0.500 0.372
27 27 Chicago 0.337 0.500 0.334
28 28 Washington 0.324 0.486 0.319
29 29 St. Louis 0.269 0.495 0.265
30 30 Pittsburgh 0.241 0.490 0.244

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

The next thing I did was determined each teams record against the top 10 power ranked teams listed above. Here is what resulted.

Team Games Wins Loss OTW SOW OTL SOL Points AdjPts AdjWinP
Carolina 8 5 3 1 0 0 1 11 11 0.688
Ottawa 12 7 5 0 0 1 1 16 15 0.625
Dallas 19 13 6 1 4 0 0 26 22 0.579
Calgary 24 14 10 1 2 2 1 31 27 0.563
Nashville 17 9 8 2 0 1 0 19 18 0.529
Colorado 27 11 16 1 0 1 4 27 26 0.481
Edmonton 25 13 12 3 2 0 0 26 24 0.480
Atlanta 12 6 6 0 1 1 0 13 11 0.458
Florida 12 5 7 1 0 0 1 11 11 0.458
Los Angeles 24 12 12 1 2 1 0 25 22 0.458
Tampa Bay 12 5 7 1 0 0 1 11 11 0.458
New Jersey 13 6 7 0 2 0 1 13 11 0.423
Vancouver 24 10 14 1 2 2 2 24 20 0.417
Phoenix 24 10 14 1 2 0 2 22 20 0.417
San Jose 20 7 13 2 1 2 3 19 16 0.400
Philadelphia 13 4 9 1 1 1 3 12 10 0.385
Detroit 17 6 11 0 0 3 1 16 13 0.382
NY Rangers 12 5 7 1 1 0 0 10 9 0.375
Buffalo 9 3 6 0 0 0 0 6 6 0.333
Anaheim 18 5 13 0 1 1 2 13 11 0.306
Boston 15 4 11 1 0 1 1 10 9 0.300
NY Islanders 18 5 13 0 1 0 1 11 10 0.278
Montreal 18 5 13 0 1 2 0 12 9 0.250
Chicago 23 6 17 0 1 1 0 13 11 0.239
St. Louis 19 4 15 0 1 0 2 10 9 0.237
Toronto 17 3 14 1 1 0 3 9 8 0.235
Columbus 24 5 19 1 2 0 2 12 10 0.208
Washington 10 2 8 0 0 0 0 4 4 0.200
Pittsburgh 13 1 12 0 0 2 1 5 3 0.115

Games-# of games played
Wins-All wins including overtime and shootout wins
Loss-All losses including overtime and shootout losses
OTW-Overtime win
OTL-Overtime loss
SOW-Shootout win
SOL-Shootout loss
Points-Points as awarded by NHL
AdjPoints-2 points for a win in regulation or overtime and 1 point for a shootout win or loss
AdjWinP- Adjusted winning percentage calculated using AdjPoints/(2*Games)

Only 5 teams have an adjusted winning percentage (which doesn’t count points for an overtime or shootout loss and treats a shootout game as a tie for both teams and 1 point each) over .500 and one would have to consider that these teams are the top 5 favourites to win the Stanley Cup. The usual suspect sit among the bottom of the standings including St. Louis, Columbus, Pittsburgh, Washington, and Chicago but with the addition of Toronto. Toronto has struggled mightily against the top teams in the league.

To take the analysis a bit further I also looked at each teams records against the bottom 10 and middle 10 teams in the league.

Bottom 10:

Team Games Wins Loss OTW SOW OTL =SOL Points AdjPts AdjWinP
Toronto 14 13 1 2 0 0 0 26 26 0.928
Dallas 10 9 1 0 1 0 0 18 17 0.850
Ottawa 12 10 2 0 0 0 0 20 20 0.833
Los Angeles 10 8 2 0 0 0 0 16 16 0.800
Detroit 22 18 4 1 1 1 0 37 35 0.795
Philadelphia 28 22 6 5 1 2 1 47 44 0.785
Carolina 26 21 5 1 2 0 0 42 40 0.769
San Jose 10 7 3 0 0 0 1 15 15 0.750
Nashville 21 15 6 0 1 1 2 33 31 0.738
Anaheim 11 8 3 0 1 1 0 17 15 0.681
Buffalo 22 15 7 2 1 0 1 31 30 0.681
Columbus 15 11 4 3 2 0 0 22 20 0.666
Atlanta 22 13 9 1 0 0 3 29 29 0.659
Vancouver 11 7 4 0 0 0 0 14 14 0.636
NY Rangers 29 19 10 1 3 3 1 42 36 0.620
Tampa Bay 26 16 10 2 2 1 2 35 32 0.615
NY Islanders 22 14 8 2 3 0 2 30 27 0.613
Edmonton 11 5 6 0 0 1 3 14 13 0.590
Boston 17 9 8 2 0 1 2 21 20 0.588
Calgary 12 6 6 0 0 0 2 14 14 0.583
Phoenix 9 5 4 1 0 1 0 11 10 0.555
Colorado 9 5 4 0 2 0 1 11 9 0.500
Florida 22 11 11 2 1 2 1 25 22 0.500
New Jersey 27 14 13 1 2 3 1 32 27 0.500
Chicago 16 7 9 3 1 3 2 19 15 0.468
Montreal 16 7 9 3 1 0 1 15 14 0.437
St. Louis 18 6 12 0 1 2 2 16 13 0.361
Pittsburgh 25 8 17 2 0 1 2 19 18 0.360
Washington 22 6 16 0 3 1 2 15 11 0.250

What you quickly notice is that Toronto is now on the top of the list. It appears that Toronto is an excellent example of a team who almost always beats the teams they should but rarely steps up and beats teams better than they are.

Middle 10:

Team Games Wins Loss OTW SOW OTL SOL Points AdjPts AdjWinP
Detroit 14 11 3 2 0 0 0 22 22 0.786
Buffalo 20 15 5 2 2 0 2 32 30 0.750
Vancouver 18 13 5 0 0 0 1 27 27 0.750
Colorado 18 13 5 1 0 0 0 26 26 0.722
Ottawa 27 18 9 1 2 0 2 38 36 0.667
Calgary 17 10 7 0 0 0 2 22 22 0.647
Montreal 17 11 6 2 0 3 0 25 22 0.647
Carolina 18 11 7 1 2 3 0 25 20 0.556
Anaheim 23 11 12 1 0 3 3 28 25 0.543
Edmonton 17 10 7 0 3 2 1 23 18 0.529
San Jose 21 10 11 2 0 0 2 22 22 0.524
Dallas 24 14 10 1 3 2 0 30 25 0.521
Boston 21 10 11 1 0 2 1 23 21 0.500
NY Rangers 13 7 6 0 1 4 0 18 13 0.500
Phoenix 22 12 10 3 2 0 0 24 22 0.500
Washington 19 10 9 1 3 1 1 22 18 0.474
Nashville 15 8 7 0 3 1 1 18 14 0.467
New Jersey 13 7 6 0 2 1 0 15 12 0.462
Tampa Bay 15 7 8 0 1 0 0 14 13 0.433
Los Angeles 22 10 12 1 2 3 1 24 19 0.432
Toronto 21 9 12 2 1 1 1 20 18 0.429
Philadelphia 12 5 7 1 0 1 0 11 10 0.417
Chicago 13 4 9 1 0 0 1 9 9 0.346
NY Islanders 12 4 8 1 0 0 0 8 8 0.333
Columbus 15 6 9 0 2 0 0 12 10 0.333
Atlanta 19 4 15 1 0 1 1 10 9 0.237
Florida 19 4 15 1 0 3 1 12 9 0.237
St. Louis 15 3 12 0 1 1 1 8 6 0.200
Pittsburgh 16 3 13 1 1 4 0 10 5 0.156

Another interesting team to look at is the Philadelphia Flyers. Many people have criticized my power rankings and used the Flyers as an example saying that they should be ranked much higher. But these results don’t back that up. Philadelphia plays fairly poorly against the top 10 teams (.385), not much better against the middle 10 (.417) but does real well against the bottom 10 (.785). Add it all up and it doesn’t make sense to have them much higher than a middle of the pack team. All totaled, 47 of Philadelphia’s 70 points (67%) have come against the worst 10 teams in the NHL. The New York Rangers show a similar pattern with 60% of their points coming against the leagues 10 worst teams.

  4 Responses to “Beating the best”

  1.  

    Your premise regarding overtime points seems misdirected.

    If I’m evaluating teams’ chances of playoff success, I want to weed out the misleading information. That misleading information is not, repeat, NOT points awarded to overtime losers. The point going to an overtime loser is earned over 60 minutes of regulation five-on-five play.

    The misleading information to be eliminated from the standings in order to make a valid analysis is the points awarded to overtime/shootout WINNERS.

    Do I value a team that racks up its points playing four-on-four or in a skills competition? Hell, no, give me the team that earns its points playing HOCKEY, regulation, five-on-five, pull-the-goalie-any-time-you-want hockey.

    The NHL should publish their standings with regulation wins as the first column, because that’s the stat that tells you how good the team is compared to the others.

    Note that, when three-point games became NHL law, the NHL started to refer to the statistic “points-percentage” and no longer referred to “winning percentage.”

    Now that “wins” can be real wins or so-called “wins,” decisions forced by a coin toss (oops, I meant a shootout), “winning percentage,” as interpreted by the NHL, is just as meaningless as “points-percentage.”

    A meaningful analysis would compare the teams’ records in regulation time only which, since the NHL offers up only standings formatted to hide the glaring frickups of their new scoring system, takes some effort.

    That’s right, just take away all the SO/OT points awarded to the SO/OT winners, and you’ll have the most valid comparative analysis.

    Okay, okay, I suppose some kind of team hockey ability is represented in a team’s overtime play, so, allow a team’s overtime win points to comprise, say, one-tenth of one percent of the total weight. Shootouts? Sure, I suppose there might be the odd penalty shot in the playoffs, so in this analysis, shootout win points warrant a weight of, oh, say one-billionth of the total.

    I miss Clarence Campbell.

  2.  

    You make some good points but let me mention a few things.

    1. I don’t count shootout wins. Games that go to a shootout are considered a tie. If a game is tied after 65 minutes it is probably fairly safe to assume that each team has an equal chance of winning had it played on in sudden death. Thus giving each team 1 point is appropriate.

    2. Yes, the 4-on-4 overtime isn’t exactly the same as 5-on-5 play but I think it still has a lot of validity for a number of reasons but most significantly because 4-on-4 is a relatively common occurance in regulation play (offsetting minors) and certainly could happen more than once in a playoff overtime anyway. So I treat an overtime win the same as a regulation win and give the winner 2 points and the loser 0 points. Now, I could give the winner 1.5 points and the loser 0.5 points or I could even treat them as a tie and give each team 1 point but I honestly don’t think it would make a huge difference.

  3.  

    The 4-on-4 situation used in the 5 minute overtime period is a good thing and a bad thing. It is good because there is much more open ice to convert passes, more shots get through to the goalies, and players have shorter shifts, which also contributes to the fast-paced period. It is not in my opinion a good indicator as to the overall strength of any one team. This is because, even with there being an increase in penalties called this year, there still aren’t as many 4-on-4 situations in the regulation 60 minutes. In the overtime, coaches have to use different strategies then the ones used in regulation because it is a different scenario. Although I doubt this would happen any time soon, I would like to see a 10-minute overtime, 5-on-5, and play the whole 10 minutes, even if a goal is scored. Then, get rid of the shootout because it shows how good any 3 players are in breakaway situations and not how good the team is. Bring back the ties to show that any two teams were even against each other on that one particular occasion.

  4.  

    [...] Prior to the Olympics I posted an article on each teams records against teams ranked in top 10, middle 10 and bottom 10 in my rankings. Once again here is how they rank. If you are wondering why a team like Philadelphia gets ranked 18th these tables will explain why. Philadelphia has just 4 wins against teams ranked in the top 10 and just 5 more against teams in the middle 10. That leaves the bottom 10 teams which Philadelphia has beaten 29 times. 54 of Philadelphia’s 78 points have come against teams in the bottom 10 teams in my power ranking system. That is why Philadelphia is 18th in the power rankings. [...]

Sorry, the comment form is closed at this time.