2003‑04 KRACH Ratings

KRACH is endorsed by College Hockey News as the best system to objectively rank teams. It stands for "Ken's Ratings for American College Hockey", because it was a statistician named Ken Butler who first implemented the methodology for college hockey.

Note: The ratings are immediately updated as results come in. For more, see below the chart or view the FAQ.

Rk Team KRACH Record   Schedule Strength
Rating RRWP Rk W-L-T Pct Ratio Rk SOS
1North Dakota983.3.8803229-7-3.78213.5882274.0
2Maine583.0.8187130-7-3.78753.70615157.3
3Boston College559.3.8130327-8-4.74362.90012192.9
4Minnesota555.8.8122726-13-3.65481.8971293.0
5Minnesota-Duluth507.8.7995526-12-4.66672.0004253.9
6Denver397.9.7625823-12-5.63751.7598226.2
7Wisconsin346.5.739912t21-12-8.60981.5629221.7
8Michigan283.2.7050626-13-2.65851.92920146.9
9Colorado College280.6.70332520-16-3.55131.2297228.4
10Ohio State277.5.7013926-15-0.63411.73313160.1
11St. Cloud State261.2.69042818-16-4.52631.1116235.1
12Michigan State235.4.671012t24-15-2.60981.56218150.6
13Miami226.3.66361123-13-4.62501.66727135.8
14Massachusetts213.1.65221519-12-6.59461.46723145.3
15New Hampshire211.6.65082120-14-6.57501.35316156.4
16Notre Dame200.9.64082020-14-4.57891.37521146.1
17Providence171.6.60992716-14-7.52701.11417154.0
18Alaska-Anchorage157.3.59254514-23-3.3875.6335248.7
19Northern Michigan151.7.58522321-16-4.56101.27831118.8
20Colgate151.4.58481022-12-5.62821.6904089.6
21Boston University150.5.583536t12-17-9.4342.76711196.1
22Cornell142.8.57291616-10-6.59381.4623797.7
23Harvard131.4.55602418-14-3.55711.25834104.5
24Dartmouth130.6.55472614-11-9.54411.19432109.4
25Rensselaer125.4.54641822-15-2.58971.4384287.2
26Northeastern118.7.53523811-16-7.4265.74414159.7
27Western Michigan117.6.53333117-18-4.4872.95029123.8
28Alaska-Fairbanks115.7.529933t16-19-1.4583.84626136.7
29Brown110.4.52032215-11-5.56451.2964385.1
30Minnesota State104.4.50895210-24-5.3205.47210221.3
31Michigan Tech97.3.4944558-25-5.2763.3823254.9
32Bowling Green96.3.49224211-18-9.4079.68924139.7
33Clarkson94.7.48883018-18-5.50001.0003894.7
34Ferris State93.8.486836t15-20-3.4342.76730122.2
35Bemidji State91.7.48231420-13-3.59721.4834461.9
36Merrimack87.8.47344411-19-6.3889.63625138.0
37Union84.5.46553215-16-5.4861.9464189.3
38Niagara80.9.45671721-14-3.59211.4524555.7
39Mass.-Lowell74.2.43935010-23-7.3375.50922145.7
40St. Lawrence71.6.43194014-21-6.4146.70836101.0
41Lake Superior67.1.4190489-20-7.3472.53228126.1
42Yale59.8.39644612-19-0.3871.6323994.7
43Holy Cross56.2.3843422-9-4.68572.1825425.8
44Nebraska-Omaha55.2.3809568-26-5.2692.36819149.8
45Vermont47.6.3529539-22-4.3143.45835103.9
46Mercyhurst44.9.34191920-14-2.58331.4005032.1
47Alabama-Huntsville35.7.30093911-16-4.4194.7224749.4
48Quinnipiac30.8.27612915-14-6.51431.0595129.1
49Findlay29.3.26784711-22-5.3553.5514653.1
50Princeton25.8.2474585-24-2.1935.24033107.3
51Sacred Heart23.1.230633t14-17-5.4583.8465227.2
52Air Force22.7.22834314-21-2.4054.6824933.3
53Connecticut20.8.21573512-16-7.4429.7955326.2
54Wayne State20.0.2097549-24-3.2917.4124848.5
55Canisius17.9.19484110-16-8.4118.7005525.5
56Bentley10.8.1346499-19-4.3438.5245620.5
57Army9.0.1165518-18-3.3276.4875818.4
58American Int'l5.2.0727575-25-4.2059.2595719.9
RRWPRound Robin Winning Percentage. A team's theoretical winning percentage would be if you played every other team, one time each.
RatioThe Ratio of wins to losses (as opposed to Win %, which is wins divided by games).
SOSStrength of Schedule

Note: For the purposes of NCAA eligibility (and therefore KRACH), a team's record is based only on games against other Division I hockey schools which are eligible for the NCAA Tournament.

KRACH Explained

As explained above, KRACH is the implementation of a sophisticated mathematical model known as the Bradley-Terry rating system, first applied to college hockey by a statistician named Ken Butler. While the model is sophisticated, and needs a computer to calculate, its essential meaning is actually quite simple.

The key to understanding KRACH is understanding that it's calculated recursively, so that the end result is self-evident by the results. In other words, if you took one team's schedule to date, and played a theoretical "game" for each game already actually played, using the KRACH ratings themselves in order to predict the winner, then the end result would be a theoretical won-loss percentage that matches the team's actual won-loss percentage. Pretty cool.

It is not possible to do any better than that with a completely objective method. Any other method introduces arbitrary-ness and/or subjectivity.

(See the FAQ for a more complete explanation.)

More detailed explanations of the KRACH ratings can be found under the following links

Send Feedback | Privacy Policy | Terms and Conditions

©2018 College Hockey News.