Author: Jim Dahl

North Dakota seemingly a lock for NCAAs

At a first pass, here’s what I see as the remaining NCAA hockey tournament possibilities:

Secured a bid: Quinnipiac, Minnesota, Miami, Boston College, Mass.-Lowell, North Dakota, New Hampshire

Can get an at-large bid: Yale, Notre Dame, Mankato, Niagara, St. Cloud, Denver, Western Michigan, Union, Wisconsin, Providence, Boston University, Rensselaer, Alaska (.01% chance), Robert Morris

Needs to win the conference tournament: Brown, Colorado College, Ohio St, Michigan, Connecticut, Canisius, Mercyhurst

Fire up the You are the committee and Pairwise predictor calculators and let me know if I missed anything, or let me know if there are particular scenarios you’d like explained. I’ll post another article later that gives some team-by-team analysis.

Updated — more detail on a per-win basis

Team PWR Possibilities
Overall By number of wins
Quinnipiac #1 100.0%
Tournament invites: 100.0%
PWR Win 0 Win 1 Win 2
#1 100.0% 100.0% 100.0%
Tournament invites: 100.0% 100.0% 100.0%
UMN #2 100.0%
Tournament invites: 100.0%
PWR Win 0 Win 1 Win 2
#2 100.0% 100.0% 100.0%
Tournament invites: 100.0% 100.0% 100.0%
Miami #3 49.8%
#4 33.8%
#5 13.1%
#6 3.1%
#7 0.2%
#8 0.0%
Tournament invites: 100.0%
PWR Win 0 Win 1 Win 2
#3 31.0% 42.4% 94.8%
#4 42.9% 44.1% 5.2%
#5 20.3% 11.6%  
#6 5.4% 1.8%  
#7 0.4% 0.1%  
#8 0.0%    
Tournament invites: 100.0% 100.0% 100.0%
Boston College #3 17.6%
#4 17.5%
#5 22.5%
#6 20.8%
#7 13.0%
#8 5.5%
#9 2.0%
#10 0.8%
#11 0.3%
#12 0.0%
Tournament invites: 100.0%
PWR Win 0 Win 1 Win 2
#3     70.6%
#4 10.7% 20.5% 27.9%
#5 23.6% 41.3% 1.5%
#6 27.5% 28.3%  
#7 21.6% 8.6%  
#8 10.4% 1.3%  
#9 4.0% 0.0%  
#10 1.6%    
#11 0.6%    
#12 0.1%    
Tournament invites: 100.0% 100.0% 100.0%
Yale #3 2.1%
#4 13.7%
#5 15.1%
#6 13.6%
#7 13.0%
#8 10.2%
#9 6.2%
#10 4.2%
#11 3.8%
#12 5.1%
#13 6.8%
#14 5.1%
#15 1.0%
#16 0.1%
#17 0.0%
#18 0.0%
Tournament invites: 96.9%
PWR Win 0 Win 1 Win 2
#3     8.2%
#4   2.3% 51.1%
#5 1.2% 16.6% 31.1%
#6 3.1% 25.2% 8.3%
#7 6.3% 25.6% 1.2%
#8 8.3% 17.8%  
#9 8.6% 7.9%  
#10 8.2% 3.5%  
#11 10.1% 1.1%  
#12 15.1% 0.1%  
#13 20.5%    
#14 15.2%    
#15 2.9%    
#16 0.4%    
#17 0.1%    
#18 0.0%    
Tournament invites: 90.7% 100.0% 100.0%
Mass.-Lowell #3 18.2%
#4 8.2%
#5 16.1%
#6 24.4%
#7 19.4%
#8 9.7%
#9 3.3%
#10 0.7%
#11 0.0%
Tournament invites: 100.0%
PWR Win 0 Win 1 Win 2
#3   1.7% 71.3%
#4 0.4% 12.9% 19.0%
#5 10.3% 34.6% 9.4%
#6 31.5% 34.2% 0.4%
#7 32.2% 13.2%  
#8 18.1% 2.6%  
#9 6.1% 0.7%  
#10 1.4% 0.1%  
#11 0.1%    
Tournament invites: 100.0% 100.0% 100.0%
UND #3 2.4%
#4 6.5%
#5 9.5%
#6 13.4%
#7 22.4%
#8 18.9%
#9 15.7%
#10 9.0%
#11 2.2%
#12 0.1%
Tournament invites: 100.0%
PWR Win 0 Win 1 Win 2 Win 3
#3       19.0%
#4     5.7% 46.3%
#5 2.4% 6.9% 19.3% 32.8%
#6 7.5% 23.1% 28.9% 1.9%
#7 22.3% 31.8% 26.4%  
#8 22.8% 23.3% 13.3%  
#9 25.0% 10.4% 4.9%  
#10 15.9% 3.6% 1.2%  
#11 3.9% 0.9% 0.2%  
#12 0.1% 0.0%    
Tournament invites: 100.0% 100.0% 100.0% 100.0%
New Hampshire #5 0.2%
#6 3.9%
#7 12.4%
#8 34.1%
#9 35.7%
#10 11.8%
#11 1.9%
#12 0.1%
#13 0.0%
Tournament invites: 100.0%		 n/a
Notre Dame #3 4.6%
#4 6.3%
#5 2.7%
#6 1.3%
#7 1.9%
#8 2.8%
#9 7.0%
#10 17.6%
#11 13.6%
#12 6.3%
#13 9.3%
#14 6.5%
#15 9.7%
#16 7.9%
#17 2.2%
#18 0.3%
#19 0.0%
Tournament invites: 81.8%
PWR Win 0 Win 1 Win 2
#3     18.3%
#4   0.1% 25.2%
#5   1.3% 9.4%
#6   2.1% 3.2%
#7 0.0% 6.5% 1.2%
#8 0.5% 9.3% 1.1%
#9 4.5% 13.9% 5.0%
#10 16.4% 13.2% 24.4%
#11 19.3% 4.3% 11.5%
#12 7.9% 8.7% 0.8%
#13 2.5% 32.2% 0.0%
#14 8.9% 8.1%  
#15 19.2% 0.5%  
#16 15.7% 0.0%  
#17 4.4%    
#18 0.7%    
#19 0.0%    
Tournament invites: 64.0% 99.0% 100.0%
Mankato #3 5.2%
#4 5.3%
#5 3.9%
#6 4.5%
#7 6.0%
#8 9.1%
#9 16.6%
#10 23.5%
#11 17.5%
#12 7.2%
#13 1.1%
#14 0.1%
Tournament invites: 99.7%
PWR Win 0 Win 1 Win 2 Win 3
#3       41.7%
#4   0.2% 4.8% 37.1%
#5 0.0% 1.1% 12.5% 16.4%
#6 0.2% 3.9% 22.8% 4.7%
#7 1.3% 9.3% 24.4%  
#8 6.4% 14.9% 17.0%  
#9 19.1% 22.2% 12.4%  
#10 32.2% 27.1% 5.2%  
#11 26.1% 17.3% 0.8%  
#12 12.3% 4.1%    
#13 2.2%      
#14 0.2%      
Tournament invites: 99.3% 100.0% 100.0% 100.0%
Niagara #4 3.9%
#5 8.5%
#6 8.0%
#7 6.1%
#8 4.7%
#9 4.7%
#10 7.4%
#11 14.3%
#12 21.7%
#13 18.4%
#14 2.2%
#15 0.1%
Tournament invites: 95.9%
PWR Win 0 Win 1 Win 2
#4     15.6%
#5   0.4% 33.6%
#6   3.1% 28.7%
#7   8.8% 15.7%
#8 0.0% 13.5% 5.2%
#9 0.1% 17.7% 0.9%
#10 2.3% 25.0% 0.1%
#11 16.7% 23.7%  
#12 39.8% 7.3%  
#13 36.5% 0.5%  
#14 4.4%    
#15 0.2%    
Tournament invites: 91.9% 99.9% 100.0%
SCSU #3 0.1%
#4 4.8%
#5 8.5%
#6 7.0%
#7 5.5%
#8 4.7%
#9 4.6%
#10 7.4%
#11 10.3%
#12 19.6%
#13 21.0%
#14 6.0%
#15 0.6%
Tournament invites: 94.1%
PWR Win 0 Win 1 Win 2
#3     0.4%
#4     19.3%
#5   0.3% 33.6%
#6   1.8% 26.3%
#7   7.6% 14.4%
#8   14.2% 4.8%
#9 0.1% 17.0% 1.0%
#10 1.8% 25.9% 0.1%
#11 9.4% 22.3%  
#12 33.9% 10.4%  
#13 41.7% 0.5%  
#14 11.9%    
#15 1.1%    
Tournament invites: 88.5% 99.6% 100.0%
Denver #7 0.0%
#8 0.3%
#9 4.2%
#10 16.9%
#11 32.1%
#12 31.2%
#13 14.9%
#14 0.3%
#15 0.0%
Tournament invites: 98.0%		 n/a
Western Michigan #13 13.5%
#14 55.7%
#15 26.5%
#16 4.2%
#17 0.1%
#18 0.0%
Tournament invites: 56.5%		 n/a
Union #8 0.0%
#9 0.1%
#10 0.5%
#11 2.7%
#12 7.2%
#13 9.5%
#14 12.2%
#15 28.6%
#16 24.4%
#17 6.5%
#18 4.3%
#19 2.6%
#20 1.2%
#21 0.2%
#22 0.0%
Tournament invites: 41.5%
PWR Win 0 Win 1 Win 2
#8     0.0%
#9     0.2%
#10     2.0%
#11     10.9%
#12     28.8%
#13   0.6% 37.1%
#14 5.0% 12.7% 21.0%
#15 24.1% 49.3%  
#16 29.0% 35.2%  
#17 16.9% 2.1%  
#18 12.8%    
#19 7.7%    
#20 3.6%    
#21 0.6%    
#22 0.1%    
Tournament invites: 15.5% 27.2% 100.0%
UW #10 0.2%
#11 1.2%
#12 1.6%
#13 2.4%
#14 4.7%
#15 3.7%
#16 5.1%
#17 11.8%
#18 21.1%
#19 20.0%
#20 14.6%
#21 9.7%
#22 3.7%
#23 0.2%
Tournament invites: 13.3%
PWR Win 0 Win 1 Win 2 Win 3
#10       1.5%
#11       9.6%
#12       12.7%
#13       19.4%
#14     2.0% 35.6%
#15     12.2% 17.2%
#16   0.1% 36.9% 4.0%
#17 7.5% 18.4% 27.8%  
#18 19.9% 36.5% 16.0%  
#19 24.8% 28.2% 4.7%  
#20 22.4% 13.3% 0.3%  
#21 17.7% 3.4%    
#22 7.2% 0.1%    
#23 0.4%      
Tournament invites: 0.0% 0.0% 6.6% 100.0%
Providence #13 0.0%
#14 0.9%
#15 7.3%
#16 9.2%
#17 8.4%
#18 10.8%
#19 14.2%
#20 15.2%
#21 12.0%
#22 8.8%
#23 7.7%
#24 4.0%
#25 1.3%
#26 0.1%
Tournament invites: 25.1%
PWR Win 0 Win 1 Win 2
#13     0.2%
#14     3.7%
#15     29.3%
#16   1.9% 35.0%
#17 0.2% 11.8% 21.3%
#18 4.4% 27.0% 7.6%
#19 11.5% 31.7% 2.1%
#20 20.2% 19.3% 0.8%
#21 20.4% 7.4% 0.0%
#22 17.2% 0.8%  
#23 15.4% 0.0%  
#24 7.9%    
#25 2.7%    
#26 0.1%    
Tournament invites: 0.0% 0.2% 100.0%
Boston University #12 0.0%
#13 2.7%
#14 4.5%
#15 12.2%
#16 8.8%
#17 9.3%
#18 15.1%
#19 22.8%
#20 15.6%
#21 7.0%
#22 1.8%
#23 0.1%
Tournament invites: 25.8%
PWR Win 0 Win 1 Win 2
#12     0.0%
#13     11.0%
#14   0.3% 17.9%
#15   2.4% 46.3%
#16   15.9% 19.3%
#17 1.0% 30.4% 4.7%
#18 15.1% 29.3% 0.9%
#19 36.7% 17.7%  
#20 29.4% 3.5%  
#21 13.9% 0.3%  
#22 3.7% 0.1%  
#23 0.3%    
Tournament invites: 0.0% 3.1% 100.0%
Rensselaer #13 0.3%
#14 1.8%
#15 9.8%
#16 31.2%
#17 31.9%
#18 17.3%
#19 6.0%
#20 1.5%
#21 0.2%
Tournament invites: 8.2%		 n/a
Alaska #16 0.3%
#17 2.6%
#18 7.2%
#19 12.1%
#20 14.3%
#21 13.7%
#22 13.5%
#23 15.1%
#24 13.3%
#25 6.5%
#26 1.3%
#27 0.1%
Tournament invites: < 1%		 n/a
Cornell #18 0.2%
#19 2.1%
#20 3.8%
#21 5.4%
#22 11.3%
#23 23.5%
#24 29.3%
#25 19.6%
#26 4.5%
#27 0.3%
Tournament invites: 0.0%		 n/a
Brown #16 0.2%
#17 2.4%
#18 6.4%
#19 10.5%
#20 17.4%
#21 20.4%
#22 15.4%
#23 6.4%
#24 6.3%
#25 6.9%
#26 5.8%
#27 1.7%
#28 0.2%
#29 0.0%
Tournament invites: 25.0%
PWR Win 0 Win 1 Win 2
#16     0.6%
#17     9.6%
#18   0.5% 24.7%
#19 0.3% 7.0% 30.0%
#20 5.0% 24.7% 21.9%
#21 10.9% 33.4% 11.3%
#22 12.1% 26.1% 1.9%
#23 12.5% 5.2%  
#24 15.4% 2.9%  
#25 20.6% 0.1%  
#26 17.5%    
#27 5.0%    
#28 0.5%    
#29 0.1%    
Tournament invites: 0.0% 0.0% 100.0%
Dartmouth #18 0.2%
#19 2.2%
#20 8.8%
#21 20.6%
#22 31.3%
#23 24.5%
#24 10.8%
#25 1.7%
Tournament invites: 0.0%		 n/a
Robert Morris #14 0.0%
#15 0.6%
#16 8.6%
#17 24.4%
#18 14.4%
#19 2.9%
#20 1.7%
#21 3.1%
#22 5.1%
#23 7.6%
#24 14.3%
#25 13.1%
#26 3.8%
#27 0.5%
Tournament invites: 0.7%		 n/a
CC #18 0.2%
#19 0.8%
#20 1.6%
#21 2.6%
#22 3.4%
#23 5.1%
#24 7.1%
#25 13.8%
#26 16.4%
#27 16.5%
#28 22.0%
#29 8.0%
#30 2.2%
#31 0.5%
#32 0.0%
Tournament invites: 12.5%
PWR Win 0 Win 1 Win 2 Win 3
#18       1.9%
#19       6.4%
#20       12.6%
#21     1.5% 19.4%
#22     6.6% 20.2%
#23   1.1% 13.9% 24.5%
#24   6.4% 30.3% 13.4%
#25 0.8% 34.7% 36.0% 1.4%
#26 5.9% 47.7% 11.7% 0.1%
#27 28.0% 10.1%    
#28 44.0%      
#29 16.0%      
#30 4.4%      
#31 0.9%      
#32 0.0%      
Tournament invites: 0.0% 0.0% 0.0% 100.0%
Ohio State #17 0.5%
#18 2.4%
#19 3.7%
#20 4.4%
#21 4.8%
#22 5.1%
#23 6.4%
#24 4.4%
#25 4.3%
#26 6.4%
#27 6.5%
#28 19.7%
#29 21.2%
#30 3.3%
#31 6.2%
#32 0.7%
Tournament invites: 25.0%
PWR Win 0 Win 1 Win 2
#17     2.2%
#18     9.7%
#19   0.0% 14.9%
#20   0.2% 17.3%
#21   1.8% 17.6%
#22   6.0% 14.6%
#23   12.3% 13.2%
#24   12.5% 4.9%
#25   12.3% 4.8%
#26 0.2% 24.5% 0.9%
#27 2.1% 21.8%  
#28 35.4% 8.0%  
#29 42.1% 0.5%  
#30 6.5%    
#31 12.3%    
#32 1.4%    
Tournament invites: 0.0% 0.0% 100.0%
Ferris State #25 8.5%
#26 26.5%
#27 36.1%
#28 22.8%
#29 5.7%
#30 0.4%
Tournament invites: 0.0%		 n/a
St. Lawrence #24 0.0%
#25 5.3%
#26 21.8%
#27 32.1%
#28 25.2%
#29 12.0%
#30 2.8%
#31 0.7%
Tournament invites: 0.0%		 n/a
Nebraska-Omaha #27 0.0%
#28 0.0%
#29 20.9%
#30 38.3%
#31 26.6%
#32 14.2%
Tournament invites: 0.0%		 n/a
Michigan #21 0.0%
#22 0.3%
#23 2.0%
#24 6.7%
#25 11.5%
#26 4.7%
#27 3.2%
#28 8.1%
#29 10.9%
#30 2.1%
#31 1.6%
#32 0.7%
Non-TUC 48.2%
Tournament invites: 25.0%
PWR Win 0 Win 1 Win 2
#21     0.2%
#22     1.1%
#23     8.0%
#24     26.6%
#25     46.0%
#26   1.1% 17.6%
#27   12.3% 0.5%
#28   32.5%  
#29   43.5%  
#30   8.4%  
#31 2.3% 1.9%  
#32 1.2% 0.3%  
Non-TUC 96.5% 0.0% 0.0%
Tournament invites: 0.0% 0.0% 100.0%
Connecticut #21 0.1%
#22 0.3%
#23 1.4%
#24 4.0%
#25 7.5%
#26 8.8%
#27 2.9%
#28 0.0%
#29 7.0%
#30 8.3%
#31 5.7%
#32 4.0%
Non-TUC 49.9%
Tournament invites: 25.0%
PWR Win 0 Win 1 Win 2
#21     0.4%
#22     1.3%
#23     5.6%
#24     16.0%
#25     29.9%
#26     35.2%
#27     11.5%
#28   0.1%  
#29   28.0%  
#30   33.2%  
#31 0.0% 22.9%  
#32 0.1% 15.9%  
Non-TUC 99.9% 0.0% 0.0%
Tournament invites: 0.0% 0.0% 100.0%
Holy Cross #26 0.0%
#27 0.2%
#28 1.9%
#29 14.3%
#30 42.7%
#31 34.9%
#32 6.0%
Tournament invites: 0.0%		 n/a
Canisius Non-TUC 100.0%
Tournament invites: 25.0%
PWR Win 0 Win 1 Win 2
Non-TUC 100.0% 100.0% 100.0%
Tournament invites: 0.0% 0.0% 100.0%
Mercyhurst #32 0.0%
Non-TUC 100.0%
Tournament invites: 25.0%
PWR Win 0 Win 1 Win 2
#32     0.0%
Non-TUC 100.0% 100.0% 100.0%
Tournament invites: 0.0% 0.0% 100.0%

Methodology

Unlike the regular season forecasts which are based on monte carlo simulations, the above tables were produced by exhaustively searching all remaining possible outcomes, so the table should be complete. The probabilities presented in the forecasts are the share of remaining possible outcomes in which that event occurs. The predictions about the likelihood of a team making the NCAA tournament assume that the PWR rankings will perfectly mimic the NCAA selection process, as they have in the past, and that any ties will be broken by RPI.

Resources

Best of three weekend Pairwise possibilities

This is the last week we look at single week ranking probabilities, after this the “You are the Committee” calculators will go live and we’ll give you the rundown on all remaining possible outcomes.

It’s an interesting week, with a lot of teams’ fortunes still at play.

While the first three (#1 Quinnipiac, #2 Minnesota, and #3 Miami) in the Pairwise Rankings are reasonably secure, it starts to get interesting at #4 Mass.-Lowell. The Riverhawks are likely to stay #4-5 if they emerge from the best of 3, but plummet if they get eliminated.

The same holds true for #5 North Dakota, #6 Denver, and #7 Boston College.

#8 New Hampshire is the first team with serious upside potential. It has a decent RPI and TUC record and is playing a fellow team under consideration, Providence. Most surprisingly, the Wildcats don’t drop much this weekend if they get eliminated (particularly if they win one).

#9 Mankato, #10 Western Michigan, #11 Yale, #12 St Cloud St, and #13 Niagara all have the opportunity to climb with success, or fall to the bubble if eliminated.

#14 Rensselaer is the highest ranked team that could fall definitively below the bubble if swept.

#15 Notre Dame, #16 Union, and #17 Boston University can all push themselves onto the bubble with success this weekend.

#18 Wisconsin can put itself into a good position but is unlikely to quite climb onto the bubble even with a sweep this weekend.

#22 Alaska is the highest ranked team not playing this weekend. Though incredibly unlikely (<1% chance), they could mathematically still climb onto the bubble.

#25 Air Force is the cutoff beyond which even active teams don’t seem to be able to climb onto the bubble through this weekend’s performance alone.

If there’s anything else you’d like to know (e.g. what games are most important for a particular team, what are the chances for a team note listed here), just let me know in the comments!

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

Final regular season PWR possibilities

#7 UND (Current PWR rankings) remains in the driver’s seat for its tournament hopes.

Its ranking remains anchored around 6-7 with a split, could rise a little with a sweep, or could fall a few spots if swept. The conference tournaments are annoyingly difficult to predict through, but a solid performance this weekend should put UND in position for an at large bid even, if they were to suffer an early exit from the conference tournament. A sweep could even (if all else goes well) vault UND into the top 4.

Games with the biggest impact on UND’s PWR
Result Number of
times		 Average increase
in UND’s
PWR
North Dakota over MSU-Mankato (2 of 2) 4.24
North Dakota over MSU-Mankato (1 of 2) 3.30
Providence over Mass.-Lowell (2 of 2) 0.81
Vermont over Boston College (2 of 2) 0.80
Wisconsin over St Cloud St (2 of 2) 0.75
Wisconsin over St Cloud St (1 of 2) 0.66
Vermont over Boston College (1 of 2) 0.60
Maine over New Hampshire (2 of 2) 0.57

Providence over Mass.-Lowell is UND’s most obvious chance to flip a comparison in its favor (UND’s current PWR comparisons). UND could easily take Mass.-Lowell on RPI if the River Hawks falter.

Boston College is a little harder to flip because UND already has RPI. Even a sweep of Mankato doesn’t quite give UND the TUC criterion, but gets them close enough (.5926 vs .6053) that a little movement around the TUC cliff could flip it.

While UND can overtake New Hampshire on RPI, that would just put the comparison in the same situation as that with BC. UND would fall just short on TUC (.5926 vs .5952), but a little movement on the TUC cliff could flip it.

Wisconsin over St. Cloud seems to matter defensively, if UND underperforms this weekend this would prevent St. Cloud from flipping the comparison.

St. Cloud St vs Wisconsin

This weekend’s matchup between #10 St. Cloud St and #20 Wisconsin has big PWR implications that could have big tournament implications.

St. Cloud can all but lock in a tournament berth with a sweep, or can put itself in danger of missing out if swept. Wisconsin has a lot to play for, with the possibility of rising to the bubble with a sweep.

Big movers

Overall movement looks to be relatively small this weekend. The most upside potential (that immediately seems to matter) is #22 Providence, which could find itself on the bubble with a sweep. The “look out below” award again goes to #19 Dartmouth, whose probable outcomes span a full 10 rankings if they get swept. An honorable mention for upside potential goes to #31 Colorado College, which has an outside chance of becoming relevant with a sweep over Michigan Tech (the Tigers’ probable outcomes again spanning an almost 10 ranking range if they sweep).

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

UND needs wins to maintain its lofty ranking

#5 UND has climbed its way back into needing a sweep this weekend just to maintain its current ranking (Current PWR rankings).

Looking at UND’s PWR comparison details, NoDak has very limited upside potential this week. Only the Boston College comparison is obviously within reach, but UND is already ahead of BC in the ranking because of the RPI tie-breaker, so flipping that comparison may not even result in a ranking increase.

UND’s PWR comparisons and RPI details reveal a significant downside, however. The RPI value of winning these games is only .5784, not significantly different enough from UND’s RPI of .5455 to budge it much. The RPI value of losing is .3284, enough to drag down UND’s RPI a fair amount. The PWR comparison details reveal that UND is winning a number of comparisons only by a slim RPI lead (Naigara .5427, Mass.-Lowell .5448, Western Michigan .5384, St Cloud St .5412, and Dartmouth .5216).

Weekend games with the largest effect on UND’s PWR:

Outcome Number of
games		 Average effect
on UND’s PWR
North Dakota over Bemidji State (2 of 2) 4.60
North Dakota over Bemidji State (1 of 2) 2.17
Merrimack over Mass.-Lowell (2 of 2) 1.05
Providence over Boston College (2 of 2) 1.05
Merrimack over Mass.-Lowell (1 of 2) 0.93
Air Force over Niagara (2 of 2) 0.92
Air Force over Niagara (1 of 2) 0.73
Alaska Anchorage over Alaska (2 of 2) 0.70

I did a minimum quick update on UND this week because I’m a little late and last week’s UND closes in on NCAA tournament berth was so comprehensive. However, if there’s anything else you’d like to see or any particular questions you have, just let me know!

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

UND closes in on NCAA tournament berth

#6 UND has moved much closer to clinching a tournament berth. Winning at least 2 of the remaining 6 would have UND more likely than not in the top 13 in PWR going into the conference tournaments. Even if UND won only 0 or 1 of its remaining games, an at large berth would still be within reach with a good showing in the conference tournament.

Given the above, UND faces a somewhat predictable situation this week of being able to make a small gain with a sweep over Denver, stay pretty much the same with a split, or fall a bit more if swept.

The final weeks of the regular season

Only three teams can’t fall from being a TUC by the end of the regular season: #1 Quinnipiac, #2 Minnesota, and #3 Miami.

Quinnipiac (PWR Details) is particularly safe, even a catastrophe seems to only drop them 3 spots. Winning only 2 of their remaining 4 would leave them with only about a 10% chance of falling from the #1 spot. Their .7778 vs. TUCs is unassailable, and .5828 RPI is miles ahead of #2 Minnesota .5658. To put that RPI into perspective, if Minnesota swept its final 6 games their RPI would only rise to about .5738 (Minnesota RPI details).

#2 Minnesota and #3 Miami each also have good RPI leads over the teams chasing them, coming in at .5658 and .5529 respectively, with #4 New Hampshire at .5477.

However, #2 Minnesota and #3 Miami aren’t alone in vying for the #2 PWR ranking at the end of the regular season. A staggering 9 teams could claim that position: #2 Minnesota, #3 Miami, #4 New Hampshire, #5 Boston College, #6 North Dakota, #7 MSU-Mankato, #9 St. Cloud St, #11 Denver, and #14 Mass.-Lowell.

Interestingly, the list of teams that can finish top 4 isn’t much longer, add only #8 Western Michigan and #10 Niagara to the list.

The team with the most upside potential for the remaining regular season is #29 Providence, which can climb to #8. That’s mostly just because big upward moves are possible from that low a rank (#28 Colgate and #30 Colorado College could each rise to #14).

The team with the most downside potential for the remaining regular season is #13 Boston University (BU PWR details), which could fall to not being a TUC. #13 BU is only #20 in RPI, and is already losing most of its common opponents comparisons.

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

Small rankings moves likely for UND in bye week

Idle #6 North Dakota could make some small moves in the Pairwise Rankings (PWR) this week, with anything between 5 and 8 being reasonably likely.

Key games for UND

Matchups that most affect UND’s PWR
Matchup Number
of wins		 Effect on
UND’s PWR
Ohio State over Western Michigan (1 of 2) 1.13
Ohio State over Western Michigan (2 of 2) 1.05

The reason those games (and others) are important can be deduced by studying the PWR comparison details for UND.

UND’s downside potential

Western Michigan is clearly the bigggest threat this week, with UND currently winning the comparison on the back of a very narrow .5470 to .5454 RPI lead.

Niagara is similarly knocking on the door, losing the comparison to UND only on the basis of Niagara’s RPI of .5381.

Finally, Yale could take the comparison with UND by defeating both Union and Rensselaer, thus raising their TUC record to .5667 (vs. UND’s .5470).

UND’s upside potential

Though New Hampshire is winning the comparison to UND 3-0, two of those criteria would flip to UND if New Hampshire got swept. UND could take both RPI and TUC.

Other interesting teams this week

Smallest range of outcomes — #1 Quinnipiac (#1-#1). Sorry Gopher fans, not this week.

Of the teams that have a two comparison or less deficit with Quinnipiac [Quinnipiac PWR comparisons] (only North Dakota, MSU-Mankato, Niagara, Wisconsin, Providence, Holy Cross, and Robert Morris), none can hope to catch their RPI of .5885 any time soon.

Largest range of outcomes — #23 Rensselaer (#13-#32), #25 Colgate (#13-#32), and #20 Merrimack (#12-#31)

Looking at Rensselaer PWR comparisons, Colgate PWR comparisons, and Merrimack PWR comparisons, all have fairly middling RPIs in the .5100s and quite a few comparisons being decided by RPI. That creates a lot of opportunity for both upward and downward movement from that part of the comparison table.

Most upside potential — #31 Robert Morris (#16–non-TUC)

Robert Morris’s story is simple (Robert Morris PWR comparisons): The TUC criterion hasn’t come into play for them yet because they don’t have 10 games and a sweep this weekend (at least a win seems necessary to stay a TUC) would give them an impressive .700 record vs. TUCs. That would immediately flip a lot of the 1-1 comparisons, and some of the 0-2’s vs teams that Robert Morris can overtake on RPI.

Most downside potential — #14 Dartmouth (#9-#27), #18 Nebraska-Omaha (#17-#31)

Dartmouth is tricky; just looking at Dartmouth’s PWR comparisons, it’s not immediately obvious why #14 Dartmouth has so much more downside potential than #15 Alaska [Alaska PWR comparisons], as RPIs and TUCs are similar. Fortunately, the simulations keep track of which games have the biggest effects on each teams, and there’s a valuable clue there:

Matchups that most affect Dartmouth’s PWR
Matchup Number
of wins		 Effect on
Dartmouth’s PWR
Dartmouth over Colgate   5.55
Dartmouth over Cornell   3.94
Brown over Rensselaer   1.56
Brown over Union   1.52
Miami over Notre Dame (2 of 2) 0.98
Lake Superior over Alaska (2 of 2) 0.84
Minnesota over Wisconsin (2 of 2) 0.80
Minnesota over Wisconsin (1 of 2) 0.61
Massachusetts over Mass.-Lowell (1 of 2) 0.58
Lake Superior over Alaska (1 of 2) 0.52
Robert Morris over Niagara (2 of 2) 0.51

The first thing that jumps out is how much Dartmouth wants Brown to win. It turns out that Brown is in danger of not being a TUC, and Dartmouth has 3 wins vs. Brown. Losing those wins would drop Dartmouth’s TUC record from .5333 to .4167. That gives Dartmouth significantly more downside potential with a couple losses than similarly ranked teams with similar RPIs.

Nebraska-Omaha [PWR comparisons], on the other hand, just has a miserable TUC of .3824. Alaska-Anchorage is a weak enough opponent that getting swept would push UNOs RPI from .5196 to about .5086. That would be enough on today’s RPI chart to drop UNO from #17 to #27 in RPI, certainly flipping a lot of comparisons given the poor TUC record. UNO seems to need a sweep not to fall.

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

PWR forecasts for February 11

#10 North Dakota is facing another typical (for it) week of a little bit of upside potential if they sweep, but a fair amount of downside potential if they get swept. (Current PairWise Rankings)

Special Beanpot note — the simulations already include the results of this week’s Beanpot games, but forecast only through next Monday NOT including the Beanpot.

UND’s upside potential comes primarily from two games:

  • If Canisius sweeps Niagara, UND could take the RPI criterion and win the comparion with Niagara
  • If Minnesota sweeps St Cloud, UND could take the RPI criterion and win the comparison with St Cloud

(UND’s pairwise comparisons detailed)

Other teams of interest this week

Note that “likely” outcomes are those with a greater than 1% chance of occurring.

Team with the narrowest spread of likely outcomes: #1 Quinnipiac (#1-#2)

Team with the largest spread of likely outcomes: #19 Union (#9-#28)

Team with the most upside potential: #21 Nebraska-Omaha (#8-#27)

Team with the most downside potential: #12 MSU-Mankato (#7-#25)

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

UND has opportunity to rise in rankings

#10 UND has fallen to a much more natural position in the Pairwise Rankings (PWR) than its lofty perch of the last couple weeks. If UND wins, its ranking goes up; it UND loses, its ranking goes down.

Follow up on last week’s predictions

In last week’s Weekly rankings analysis column I observed (and the outcomes were) that:

  • UND had more downside than upside and was likely to slump with anything less than a sweep. (Another one point weekend later, UND indeed fell from #6 to #10, right between the predicted win 1 and win 0 curves).
  • Mass.-Lowell similarly had more downside than upside, with only the comparison with UND as a possible gain and six comparisons as possible losses. (Mass.-Lowell took 3 of 4 points for the weekend and stayed #7 by taking the comparison with UND but losing the comparison with Miami).
  • Ferris St had big movement potential from #21 with a sweep or being swept (Ferris St split the weekend and dropped just one spot to #22).
  • Alaska had big movement potential from #18 (Alaska swept the weekend and climbed to #12, a much higher PWR than RPI).
  • Cornell had big movement potential from #22 (Cornell lost two and dropped to #27).
  • Dartmouth had the most downside potential from #9 (Dartmouth split the weekend and dropped to #15).

Bottom-line: this stuff works and reveals useful information that would be difficult to uncover browsing the comparison tables. But enough back-patting.

Whither UND?

In my first PWR post two weeks ago, I noted that UND was artificially high at #7 and needed to win about 6 of 14 remaining regular season games to finish in the at-large bid range of 13-15. UND has since gone 0-2-2 (.250) but dropped only to #10 in PWR.

UND still has a ton of downside potential if they continue to lose, but things are evening up a little with some upside potential for a sweep.

In the longer run view, the rough equivalent of 1 win has left UND needing about 5 wins in the remaining 10 games to be most likely to finish the regular season in the at-large bid range. Remember, though, that big moves can occur with a winning streak in the conference tournament.

Outcomes that most affect UND’s PWR this week
Outcome Average effect
on UND’s PWR
UND over Wisconsin (2 of 2) 11.57
UND over Wisconsin (1 of 2) 6.84
Holy Cross over Canisius (2 of 2) 1.74
Holy Cross over Canisius (1 of 2) 1.59
Minnesota-Duluth over Denver (2 of 2) 1.30
Michigan Tech over Nebraska-Omaha (2 of 2) 1.10
Minnesota-Duluth over Denver (1 of 2) 1.04
Michigan Tech over Nebraska-Omaha (1 of 2) 0.89
Bentley over Niagara (2 of 2) 0.76
Ferris St over Western Michigan (2 of 2) 0.71
Bentley over Niagara (1 of 2) 0.56
Rensselaer over Dartmouth 0.56

A lot of those are obvious with a quick glance at UND’s PWR comparisons.

  • Holy Cross is in danger of not being a TUC, but UND needs its 2 wins over the Crusaders to contribute to its mediocre record vs. TUCs (currently .500).
  • Denver could pass UND in RPI with a better weekend than North Dakota. Also, Duluth becoming a TUC seems to give UND a little juice (probably due to UND’s 1-0-1 record vs the Bulldogs, better than UND’s record vs current TUCs).
  • Nebraska-Omaha’s RPI could climb to about .539 with a sweep this weekend, which could allow them to take the comparison if UND also drops.
  • Similarly, Niagara and Dartmouth could take their comparisons with UND with a climb in RPI and a drop by UND.
  • Finally, UND barely loses the comparison with Western Michigan on the basis of RPI. Outperforming the Broncos this weekend would probably flip that.

Teams to watch

Biggest potential outcome (likelihood greater than 1%) spread of the week goes to UND (likely to come out between #5 and #23) and Dartmouth (likely to come out between #6 and #24).

Biggest upside potential goes to Nebraska-Omaha, which could climb a staggering 13 slots from #21 to #8 if everything went their way. This is mostly due to their reasonably strong RPI for a #21 team.

Biggest downside potential goes to Mass.-Lowell, which could drop 14 slots from #7 to #21 if everything went wrong. The looming disaster for Mass.-Lowell is clearly their miserable .250 TUC record coming into play when they hit 10 games vs. teams under consideration. It won’t happen this week without a little movement on the cliff (Maine isn’t a TUC), but it will eventually if they don’t gain some wins vs. TUCs.

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

Weekly rankings analysis

#6 UND is still perched somewhat precariously high in the Pairwise Rankings (PWR), though can continue to stay there if they keep winning.

Follow-up on last week’s predictions

Let’s also take a look back at last week’s A first look at NCAA hockey PairWise Rankings (PWR). I observed (and the outcomes were) that:

  • #4 Minnesota was secure in its PWR and that anything short of being swept would be unlikely to result in a drop. (Minnesota took 3 points and climbed to #2.)
  • #7 UND was on a bit more of a cliff and had the potential to fall substantially if swept. (UND took 1 point and climbed to #6. Though well within the predicted range for being between 0 and 1 wins, UND came out near the top of the range because of an unlikely sweep of Denver by St. Cloud State, as further explained in this forum post.)
  • #3 Quinnipiac was the real deal and would be well protected against sharp downward movement. (Despite not playing, Quinnipiac climbed to #1.)
  • #6 Boston University was in an artificially high ranking and doomed to fall a bit when some inevitable losses came in. (BU fell to #13 after a pair of tough losses to Northeastern and Mass. Lowell.)

All-in-all, the predictions were pretty good. A little more information could have helped readers anticipate UND’s additional strength if St. Cloud St. swept, so I’ll try to include more of that sort of information. Otherwise, readers knew in advance of Minnesota’s and Quinnipiac’s strength and Boston University’s hidden weakness.

North Dakota this week

Currently ranked #6, UND continues to have significantly more downside than upside. A sweep is most likely to result in a rise of only one position (though staying #6 is also quite likely), whereas getting swept could result in a fall of up to around 10 positions.

Games that most affect UND’s PWR

Outcome Average effect on UND’s PWR
UND over St. Cloud (2 of 2) 8.33
UND over St. Cloud (1 of 2) 5.73
Holy Cross over Bentley (2 of 2) 1.03
Holy Cross over Bentley (1 of 2) 0.80
Clarkson over Dartmouth 0.63
Alaska Anchorage over Wisconsin (2 of 2) 0.58
Northeastern over Mass.-Lowell (2 of 2) 0.54
Minnesota over MSU-Mankato (2 of 2) 0.52
Cornell over Yale 0.52
St. Lawrence over Dartmouth 0.50

UND comparisons most likely to flip

To figure out why those games matters takes some diving into North Dakota’s PWR comparisons.

Upside potential

  • Cornell and/or Colgate wins over Yale could help UND take the comparison with Yale as TUC comes into play when Yale reaches 10 games vs. TUCs.

Downside potential

  • A Bentley win over Holy Cross could knock Holy Cross out of consideration. This would considerably harm UND’s winning percentage vs. TUCs.
  • A Clarkson win over Dartmouth would be enough to give Dartmouth the edge over UND’s current RPI. A St. Lawrence win over Dartmouth would have a slightly smaller effect.
  • A pair of Wisconsin wins over Alaska Anchorage could be enough to take RPI and COP, and thus the comparison with UND.
  • A Mass.-Lowell win over Northeastern would give Mass.-Lowell the edge over UND’s current RPI and the comparison with UND.

On the cliff

The two teams with the most lopsided downside vs. upside potential this week are Mass.-Lowell and North Dakota.

Team Current PWR Highest PWR
with >1% chance		 Lowest PWR
with >1% chance
Mass.-Lowell 7 5 19
North Dakota 6 5 17

Having already beat to death why UND has little upside potential but high downside potential this week, let’s take a quick look at Mass.Lowell.

First, looking at Mass.-Lowell’s PWR comparisons, notice that they haven’t yet played enough teams under consideration for the TUC criterion to come into play. Northeastern, their only opponent this weekend, also isn’t a TUC so only some movement along the TUC line would bring this into play (and none is obvious).

Second, notice that of the first 7 comparisons Mass.-Lowell wins (Miami through St. Cloud St), 6 could easily be flipped to losses via RPI (the RPIs are within .01 and flipping them would flip the comparison).

Finally, the only obvious upside potential for Mass.-Lowell is taking the comparison with North Dakota if UND stumbles.

Other superlatives

But those aren’t the biggest potential movers of the weekend (again, only measuring outcomes with over a 1% likelihood).

Teams in the high teens, low twenties seem poised to make the biggest moves in either direction. #21 Ferris St. could fall between #12 and #30, #18 Alaska between #10 and #28, and #22 Cornell between #13 and #30.

But #9 Dartmouth has the largest one way (downside) potential, with the possibility of falling up to 13 spots to #22 if everything goes wrong.

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Resources

A first look at the NCAA hockey PairWise Rankings (PWR)

January is the typical time to start paying attention to the PairWise Rankings (PWR) that mimic that NCAA’s tournament selection criteria. UND is currently ranked #7, and is poised to move this weekend with a pair of contests against arch-rival Minnesota.

North Dakota – Gophers

Both are highly ranked, Minnesota #4 and UND #7. However, the weekend’s games have only minor short term implications for Minnesota, but more significant short term implications for North Dakota if either team sweeps.

The following charts are the forecast likelihoods of each PWR ranking as of next Monday (Jan 21) for each team based on its own performance this weekend.

North Dakota’s remaining season

North Dakota seems pretty well positioned to make the tournament. Going about .500 over the rest of the regular season would likely leave UND in position for an at-large bid going into the conference tournaments.

The following chart contains the forecast likelihoods of each PWR ranking as of the end of the regular season for UND based on its own performance over the remainder of the regular season.

Minnesota’s remaining season

“Numbers” on the USCHO forum asserted that Minnesota’s 8-0-0 non-conference record particularly shores up their PWR ranking because of the Common OPponents criterion.

It seems true that running the rest of the season even under .500 would still leave Minnesota in a good position heading into the conference tournaments. Minnesota’s strength in COP does help, but once TUC is in play you still need to win one more criterion to take the comparison. Opponents’ strength held constant, going .500 would drop Minnesota’s RPI to about .555, a far more pedestrian number. Minnesota’s respectable TUC helps them in that regard, further supporting their lofty rating down the road.

TUC is coming into play

A lot of teams are just starting to hit 10 games vs. other Teams Under Consideration (TUCs), bringing that criterion into play in as-of-now PWR calculations (which, frankly, additionally reveals how absurd as-of-now calculations are; because we know most of these teams will hit 10 games vs. TUCs by the end).

The top 10 teams’ records vs TUCs are somewhat predictive:

Team Record vs TUCs
1. New Hampshire .7500
2. Boston College .6667
3. Quinnipiac .8077
4. Minnesota .6875
5. Notre Dame .6364
6. Boston University .4000
7. North Dakota .5000
8. Denver .5938
9. Yale .5000
10. Miami .6667

Indeed, Minnesota’s good TUC record is part of what helps insulate it a bit from big downward movement. But that’s true of all the top few teams.

The two most exceptional seem to be Quinnipiac, whose TUC record does seem to provide a fair amount of insurance against big downward moves; and Boston University, whose low TUC means it will take a strong effort down the stretch to maintain their current ranking.


Boston University is a particularly interesting case. Despite their dismal TUC, they’re currently #6 in the PWR on the back of the #1 strength of schedule in RPI. Looking at BU’s RPI details, it would take a noticeable increase in winning percentage (to about .800 over the remainder of the season vs. about .632 to date) to offset their upcoming decline in SOS and keep RPI constant. Short of that, expect BU’s RPI to fall and their PWR with it.

Of course, those TUC records can change not only based on teams’ own upcoming performances, but as former opponents play additional games that may add or remove them from being “under consideration”.

Methodology

Each forecast is based on at least one million monte carlo simulations of the games in the described period. For each simulation, the PairWise Ranking (PWR) is calculated and the results tallied. The probabilities presented in the forecasts are the share of simulations in which a particular outcome occurred.

The outcome of each game in each simulation is determined by random draw, with the probability of victory for each team set by their relative KRACH ratings. So, if the simulation set included a contest between team A with KRACH 300 and team B with KRACH 100, team A will win the game in very close to 75% of the simulations. I don’t simulate ties or home ice advantage.

Closing remarks

This is my first stab at firing up the PWR simulations and writing this sort of post this year. It does always seem to take me a little while to get back into the groove, so please point out any errors, questions, or points that need clarifying.

Though I started with an analysis of UND’s position, let me know if there’s anything else in particular that you’d like to see.

Resources