When Corsi Bites Back

uno-hockey-stats
For the records, the above does not count as puck possession.

For an Omaha hockey fan, last season was rough. In late February and March, I watched in despair as my UNO Mavericks lost the last eight games of the season and slid out of the NCAA tournament conversation. Six of those losses came against eventual Frozen Four teams, but it still stung. How could a team that reached the Frozen Four just a year prior – and was a #1 seed as late as January – fall so far, so fast?

Honestly, I checked out of college hockey for a little while. I whistled past the further carnage of losing UNO’s top player and two assistant coaches. I enjoyed my spring and summer, cautiously avoiding dealing with the emotional rollercoaster that was 2015-16.

Now we’re less than a month away from a new season. New year, clean slate. But… what to make of this year’s Mavericks? I can’t get 100% pumped for this season without coming to terms with how last season ended. I need closure. As a UNO fan, I need to process this unthinkable turn of events.

So I’m going to do that in the only way I know how: hockey stats.

First, a disclosure: I am not a hockey expert. I am not a hockey coach, nor a hockey player. I offer no prescriptions, indictments or playbooks. I simply intend to lay out the data and share the story it tells. Speaking as a UNO fan, I will warn that this might be painful, but it needs to be done.

Second, another disclosure: I’m going to analyze primarily by looking at two “advanced” stats – Even Strength Corsi For, and Even Strength PDO. Thanks to College Hockey News, we now have two seasons of these advanced stats available for every NCAA team. I’ve compiled this data for the last two seasons. For comparison, I also included RPI and Pairwise rankings from regular-season end (via siouxsports) and season-end ELO rankings (via Ebscer). I consider these three “rankings” more or less objective measures of team quality, and none of them directly factor in on-ice statistics.

This is the database I’ll be using. Feel free to play around and do your own digging:

Team
Year
W
L
T
SH%
SV%
PDO
CF%
PW
RPI
ELO
Tourney?
FF?
Conf
Air Force14-15162149.00%89.20%98.2153.6490.4521391NONOAHA
Air Force15-16181158.70%91.40%100.1853.8280.5091514NONOAHA
Alabama-Huntsville14-1582646.00%92.60%98.6339.1530.4411302NONOWCHA
Alabama-Huntsville15-1672168.40%91.20%99.6444.2570.4331352NONOWCHA
Alaska15-16102047.30%91.30%98.5947.6550.4461393NONOWCHA
Alaska 14-15191327.30%91.10%98.4154.6240.5201548NONOWCHA
Alaska-Anchorage14-1582247.60%90.70%98.2444.9520.4431357NONOWCHA
Alaska-Anchorage15-16112039.70%90.90%100.5944.6520.4481345NONOWCHA
American Int'l14-1542578.60%89.40%98.1038.8590.3951230NONOAHA
American Int'l15-1662737.70%89.80%97.5240.3600.4011235NONOAHA
Arizona State15-1632204.30%91.40%95.6836.7590.4011297NONOIND
Army14-1582246.10%91.30%97.3449.4570.4141284NONOAHA
Army15-16101396.00%94.30%100.2748.1450.4691426NONOAHA
Bemidji State14-15161758.60%92.60%101.2249.3270.5141504NONOWCHA
Bemidji State15-16161466.40%91.20%97.6951.6320.4981512NONOWCHA
Bentley14-15171557.30%92.80%100.1148.9410.4791443NONOAHA
Bentley15-16111768.00%92.00%99.9849.4500.4611387NONOAHA
Boston College14-15211438.20%92.20%100.3352.9110.5461591YESNOHE
Boston College15-16245510.00%93.70%103.6952.440.5801676YESYESHE
Boston University14-1528858.60%93.80%102.3455.030.5721683YESYESHE
Boston University15-16191058.40%91.60%99.9556.2110.5531604YESNOHE
Bowling Green14-15231159.40%91.40%100.8152.6160.5411540NONOWCHA
Bowling Green15-16201267.60%93.40%100.9852.5260.5131538NONOWCHA
Brown14-1582037.70%90.30%98.0242.4470.4561413NONOECAC
Brown15-1651777.30%89.50%96.7849.0470.4521375NONOECAC
Canisius14-15181276.40%92.70%99.1249.4400.4841498NONOAHA
Canisius15-16101958.30%92.30%100.5443.4540.4441371NONOAHA
Clarkson14-15122056.70%90.70%97.3456.1440.4721406NONOECAC
Clarkson15-16181337.30%91.30%98.6153.5230.5191534NONOECAC
Colgate14-15221247.70%92.40%100.1254.1170.5391604NONOECAC
Colgate15-16102227.90%89.40%97.3047.9440.4671434NONOECAC
Colorado College14-1562636.20%89.30%95.5843.2510.4491373NONONCHC
Colorado College15-1662716.10%89.30%95.4642.8530.4481339NONONCHC
Connecticut14-15101976.80%92.00%98.8842.7430.4771430NONOHE
Connecticut15-16111947.00%91.30%98.3044.6420.4631413NONOHE
Cornell14-15111465.30%92.40%97.7448.8360.4911485NONOECAC
Cornell15-1613978.10%92.60%100.6947.8170.5351552NONOECAC
Dartmouth14-15171248.30%92.50%100.8053.7220.5211548NONOECAC
Dartmouth15-16141418.30%90.60%98.9450.8270.5221555NONOECAC
Denver14-15241428.90%91.90%100.8553.250.5681628YESNONCHC
Denver15-1621859.00%92.50%101.4953.760.5791704YESYESNCHC
Ferris State14-15182026.50%92.40%98.9450.9340.4991510NONOWCHA
Ferris State15-16151467.60%92.10%99.7150.4330.5111563YESNOWCHA
Harvard14-15211339.60%92.30%101.8348.290.5501576YESNOECAC
Harvard15-1614948.60%92.80%101.4351.5130.5841591YESNOECAC
Holy Cross14-15141856.20%93.20%99.4249.7480.4541379NONOAHA
Holy Cross15-16181159.00%92.10%101.1051.9300.4941458NONOAHA
Lake Superior14-1582826.10%91.40%97.4743.9540.4371345NONOWCHA
Lake Superior15-16132055.00%92.70%97.6744.7460.4651439NONOWCHA
Maine14-15142238.60%89.90%98.4350.1420.4781478NONOHE
Maine15-1682265.30%90.50%95.7949.6510.4441387NONOHE
Massachusetts14-15112328.40%89.00%97.3743.5450.4641418NONOHE
Massachusetts15-1682247.10%89.90%97.0243.8480.4511341NONOHE
Mass-Lowell14-15211269.50%92.00%101.5353.9180.5371592NONOHE
Mass-Lowell15-1621858.50%94.20%102.6751.990.5561654YESNOHE
Mercyhurst14-15191649.20%92.40%101.6245.5390.4841459NONOAHA
Mercyhurst15-16171348.60%91.60%100.2742.9350.4841447NONOAHA
Merrimack14-15161846.90%92.60%99.4848.2300.5071459NONOHE
Merrimack15-16111677.50%91.00%98.5052.8360.4781460NONOHE
Miami14-15251418.20%91.90%100.0454.240.5691625YESNONCHC
Miami15-16151637.20%90.30%97.4948.5210.5161527NONONCHC
Michigan14-15221509.90%90.90%100.7953.6190.5291568NONOBIG10
Michigan15-16227510.80%91.30%102.0651.670.5681649YESNOBIG10
Michigan St14-15171626.90%93.40%100.3149.0310.5051542NONOBIG10
Michigan St15-16102247.70%89.50%97.2346.8430.4621439NONOBIG10
Michigan Tech14-15291029.20%94.20%103.4155.670.5651616YESNOWCHA
Michigan Tech15-1621858.90%92.70%101.5657.0140.5361630NONOWCHA
Minnesota14-15231338.50%91.90%100.4753.4100.5501622YESNOBIG10
Minnesota15-16191609.40%90.50%99.8752.5180.5281568NONOBIG10
Minnesota State14-1529839.20%91.50%100.7358.010.5921659YESNOWCHA
Minnesota State15-16181176.70%92.00%98.6360.2220.5191577NONOWCHA
Minnesota-Duluth14-15211637.40%91.80%99.2055.860.5661571YESNONCHC
Minnesota-Duluth15-16151456.80%92.50%99.3658.1120.5471624YESNONCHC
Nebraska-Omaha14-15201367.30%93.50%100.7447.580.5501565YESYESNCHC
Nebraska-Omaha15-16181518.20%90.70%98.8649.9150.5291517NONONCHC
New Hampshire14-15191928.20%91.10%99.2649.7290.5121567NONOHE
New Hampshire15-16101868.50%91.80%100.3044.0390.4721434NONOHE
Niagara14-1572846.80%89.30%96.1347.1580.4031319NONOAHA
Niagara15-1652366.60%89.60%96.1950.1580.4121270NONOAHA
North Dakota14-15291038.50%93.50%102.0051.520.5801681YESYESNCHC
North Dakota15-16285310.40%93.40%103.8756.510.6081769YESYESNCHC
Northeastern14-15161647.90%92.70%100.5750.3230.5201549NONOHE
Northeastern15-16161359.10%91.00%100.0451.2190.5431671YESNOHE
Northern Michigan14-15141867.30%92.20%99.5046.4350.4951417NONOWCHA
Northern Michigan15-16151478.10%92.80%100.9546.2340.4921452NONOWCHA
Notre Dame14-15181959.30%92.20%101.5048.7330.5041537NONOHE
Notre Dame15-1618778.80%93.30%102.0850.080.5461583YESNOHE
Ohio State14-15141938.80%90.50%99.3947.2370.4891506NONOBIG10
Ohio State15-16131749.60%90.50%100.0547.6310.4981531NONOBIG10
Penn State14-15181547.00%91.00%97.9955.1320.5041496NONOBIG10
Penn State15-16201247.20%91.50%98.7056.2200.5221516NONOBIG10
Princeton14-1542334.20%91.30%95.5044.7560.4171299NONOECAC
Princeton15-1652136.00%92.20%98.2145.1560.4301304NONOECAC
Providence14-15261327.50%93.80%101.3654.3150.5411655YESYESHE
Providence15-1625548.00%94.50%102.4455.150.5751683YESNOHE
Quinnipiac14-15231247.30%91.30%98.6556.5140.5441598YESNOECAC
Quinnipiac15-1625278.90%92.50%101.4255.620.5991738YESYESECAC
Rensselaer14-15122637.00%90.20%97.2348.0460.4631407NONOECAC
Rensselaer15-16151277.70%92.90%100.6045.9250.5191498NONOECAC
RIT14-15201558.20%91.50%99.6853.9380.4871528YESNOAHA
RIT15-16141467.80%89.40%97.2755.4410.4871505YESNOAHA
Robert Morris14-1524858.60%93.30%101.9248.5250.5201546NONOAHA
Robert Morris15-1621949.10%93.10%102.1648.9160.5261553NONOAHA
Sacred Heart14-15131967.90%91.20%99.1750.7500.4521401NONOAHA
Sacred Heart15-16121847.10%91.00%98.1051.0490.4541368NONOAHA
St. Cloud State14-15201917.10%92.30%99.4149.6120.5461615YESNONCHC
St. Cloud State15-16278111.30%93.10%104.3450.830.5961696YESNONCHC
St. Lawrence14-15201439.80%94.20%104.0548.0210.5271556NONOECAC
St. Lawrence15-16171349.00%93.50%102.5351.0240.5221563NONOECAC
Union14-15191828.10%92.00%100.0950.5280.5101553NONOECAC
Union15-16131296.70%92.50%99.1651.3290.5001499NONOECAC
Vermont14-15221547.40%91.90%99.3452.4200.5361560NONOHE
Vermont15-16122036.10%92.60%98.7653.2370.4921501NONOHE
Western Michigan14-15141857.20%91.40%98.6051.0260.5191515NONONCHC
Western Michigan15-1682136.90%90.50%97.4144.7380.4681413NONONCHC
Wisconsin14-1542656.00%90.60%96.5742.8550.4251380NONOBIG10
Wisconsin15-1681887.90%89.40%97.2749.7400.4711408NONOBIG10
Yale14-15181056.80%94.10%100.9152.6130.5411577YESNOECAC
Yale15-1619646.80%92.90%99.7954.2100.5451602YESNOECAC

Click here to open in Google Sheets.

Now, we can argue all day about what shot stats mean (I did some of that myself in this post last year). Certainly the usefulness of advanced stats has not yet been demonstrated to the level of sophistication in the NHL. There are certainly reasons to be skeptical. Alone, the data suggests Corsi and PDO aren’t particularly indicative of anything. But in combination, I believe these stats do illustrate larger trends. More on that in a minute.

Continue reading

PDO in College Hockey

providence_hockey

If you’re not convinced by now that puck possession matters, you might as well go ahead and close this tab. Corsi is real, and it’s here to stay – yes, even in college hockey. Ryan Lambert at College Hockey News pretty much dropped the mic on the issue as the 2015 conference tournaments started, so I won’t waste any more words on it.

Though if you’d like to be spared from an RL article, I’ll just say this: of the top 20 CF% teams in the NCAA, 60% made the tournament (12/20). Of the 39 other teams, 10% made the tournament (4/39). More possession, more shots, more goals, more wins. What’s not to get?

The next thing everyone always brings up is, “yeah but shot quality.” Sorry, but shot quality is not a game plan, nor something even the best players can sustain. If your team has a high shooting percentage, it’s likely to regress the more they play. Same with save percentage. Hot goalies are usually just that – hot. Until they’re not, because .925 simply isn’t sustainable for most goalies. Eventually they’re going to have a few .795 nights.

The stat that measures all of that, of course, is PDO. Puckology has a great post on this. PDO adds the team shooting percentage and save percentage into one stat. The baseline for this stat is 100.0, because that’s the league average – always was and always will be (the average of all shot percentages in the league plus the average of all save percentages will always be 100%). So every team in a league will ultimately regress toward a PDO of 100.0. Your incredible goaltending will falter, your super-lucky “shot quality” will come down. That’s the nature of the game – regression to the mean is a statistical fact.

So here’s where things get interesting, and where I always got a little hung up: what I just said above isn’t 100% true. The baseline for every team or player probably is not 100%. The Chicago Blackhawks have more talent than the Arizona Coyotes, yes. But this is the NHL we’re talking about – each of those 30 teams is among the most talented hockey teams in the world (yes, even Buffalo). The baseline for the Blackhawks might be higher than the Coyotes, but not much. We’re talking about a thin band: PDO in the NHL ranges between 97.0-102.0. So is PDO talent? Probably not very likely in the NHL – it’s just variance from the mean, indicating likelihood of regression.

But is that still true in college hockey? You don’t have the same talent parity as in the NHL – most people wouldn’t put Boston University and Niagara in the same sentence. No one will argue there’s not a significant talent gap there. Sure enough, the PDO range in the NCAA is much larger than in the NHL – 95.7-104.5.

Therefore, PDO in college hockey might not be as indicative of unusual good/bad luck as it is in the NHL. Some teams might just have a naturally higher or lower baseline because of a higher level of quote-unquote talent.

Continue reading

NCHC shot results

UNO-hockey-shots

Watching UNO hockey get manhandled for two consecutive games in St. Cloud this weekend, I couldn’t help but notice how few quality scoring chances the team was able to put together. On the flip side, the Huskies seemed to have a high-probability chance every minute or so. Both teams have played difficult schedules, yet the Mavericks have won many more games that St. Cloud. Meanwhile, on the stat sheet, UNO has much higher season shot and save percentages than St. Cloud, although SCSU’s possession is clearly better than UNO.

So… what’s going on? Is it just luck? Is UNO slumping and SCSU streaking? Is youth finally catching up to UNO and experience paying off for SCSU?

To answer these questions, I took a look at all shots taken this season by every NCHC team. Thanks to College Hockey News, I can now do that for multiple situations (5v5 and power play), and I can also see results of each shot.

There are some interesting observations to be made from looking at shot results. The chart below lists each team’s shot results as a percentage of the total shots they take:

 Shot results – All Shots

Team Blocked Wide Hit Post Saved Goal Total % on Goal
St. Cloud 19.6% (315) 23.2% (372) 1.0% (16) 50.9% (817) 5.2% (84) 1604 56.2%
Colorado College 24.1% (335) 19.4% (270) 0.9% (12) 47.8% (715) 4.1% (57) 1389 55.6%
Miami 22.9% (417) 21.1% (384) 1.3% (24) 50.1% (912) 4.6% (83) 1820 54.7%
North Dakota 24.5% (442) 20.0% (362) 1.2% (21) 48.7% (880) 5.7% (103) 1808 54.4%
Minnesota-Duluth 22.6% (435) 22.5% (434) 1.2% (23) 48.9% (943) 4.8% (93) 1928 53.7%
Western Michigan 23.4% (357) 22.4% (342) 0.6% (9) 48.5% (742) 5.2% (79) 1529 53.7%
Denver 23.7% (400) 21.8% (368) 1.4% (24) 47.8% (807) 5.3% (89) 1688 53.1%
Omaha 24.7% (372) 22.5% (339) 1.2% (18) 46.0% (693) 5.7% (86) 1508 51.7%

First, some minor details. More of Omaha’s shots get blocked than any other NCHC team. Meanwhile, St. Cloud has the fewest amount of shots blocked. However, UNO and North Dakota lead the league in percentage of shots that result in goals. The problem for Omaha is that they take fewer shots than anyone in the league save for Colorado College. And, if you add goals plus shots saved, to count all shots that make it on net, UNO is dead last with 51.7%. Who’s first? St. Cloud, with 56.2%. Even the hapless Colorado College Tigers get 55.6% of their shots on net.

So, let’s isolate even strength vs. power play situations.

Continue reading

NCHC Fridays and Saturdays

nchc_den_ndak

Denver has thrived on Friday nights this season. North Dakota likes Saturdays. 

A couple of weeks ago I looked at home vs. away performance for the eight NCHC teams. With four weekends left, each team has two home and two away series (with the exception of Denver and Colorado College, who have a home-and-home). So before the action starts, let’s take a look at the other piece of the puzzle – Friday vs. Saturday night.

I didn’t expect as big a Friday-Saturday difference in performance as we saw in home-away performance, but as it turns out, a few teams really do seem to have a preference.

Let’s dive in, shall we? I’ll update the home/away and add the Friday/Saturday, all stats as of February 12:

hlogo-CC  Colorado College

Situation Goals For Goals A. Shots For Shots A. Shot% Save% Poss% Points Remaining
Home 10 33 183 220 5.46% 85.00% 45.4% 0.0 5
Away 11 36 203 320 5.42% 88.75% 38.8% 0.3 3
Friday 12 35 229 295 5.24% 88.14% 43.7% 1.1 4
Saturday 9 34 157 245 5.73% 86.12% 39.1% 0.0 4

Whatever advantage the Tigers looked like they might have from a 6-game home stretch was quickly deflated by three straight losses with a combined score of 2-18. Not to pile on, but neither Friday nor Saturday nights look particularly good for CC either.

The predictive model suggest CC might see a sliver of luck Friday nights on the road. Do they have any of those left? Oh yeah, they do – against Omaha, the only NCHC team they’ve beat this season. So watch out, Omaha.

 

hlogo_DEN  Denver 

Situation Goals For Goals A. Shots For Shots A. Shot% Save% Poss% Points Remaining
Home 33 23 297 258 11.11% 91.09% 53.5% 24.8  3
Away 15 18 193 221 7.77% 91.86% 46.6% 14.3  5
Friday 36 16 265 257 13.58% 93.77% 50.8% 32.8 4
Saturday 12 25 225 222 5.33% 88.74% 50.3% 6.1 4

When we last looked, Denver was showing signs of stronger performance at home. That’s still true, and that’s bad news for Denver, because they only have three home games left. They’ve played well but haven’t exactly lit the world on fire, so the odds are still stacked against them in getting home ice. For some reason their possession drops a ton on the road – bad news for anyone.

Oddly, Denver is showing a very strong preference for Friday games – they’ve only lost one this entire season. If they can continue that success rate, they might have a chance. But they’ll have to do it on the road against North Dakota starting this weekend. And according to the data, their shooting percentage drops by more than half on Saturdays. Ouch.

Continue reading

NCHC home advantage

wmu_duluth_2015_01_16

Broncos gonna buck. 

Halfway through the NCHC season, nothing is settled, and most teams in the league are very much in the race for first-round conference tournament home ice. But does every team want to play on home ice? That’s the question that crossed my mind watching Western Michigan take five points from a superior Minnesota-Duluth team this weekend.

It seems like a dumb question. Obviously every team wants to play in front of a friendly crowd and avoid the hassle of travel. Sports psychologists have demonstrated that players get a boost from playing at home, and there is even evidence that refs are affected by the home crowd (PDF). But that doesn’t mean that every team is actually playing better at home. I took a look at home performance vs away performance in terms of shot-based stats (which I’m tracking here), and what they mean for the rest of the season.

Continue reading

NCAA Hockey Possession

uwm_cc

I’ve seen some debate over which is the best NCAA hockey conference this year. It’s a fun debate to have, however meaningless. The obvious way to make this determination is to look at inter-conference records. That chart is available here, and based on that metric, the six D-1 conferences are ranked as such:

  1. NCHC, .660 win percentage
  2. ECAC, .583
  3. WCHA, .558
  4. Hockey East, .500
  5. Big Ten, .438
  6. Atlantic Hockey, .256

Based on win percentage, the NCHC is the best conference. Its teams are winning 2 non-conference games for every one it loses. It’s not really close. Meanwhile the Big Ten seems to be doing a lot of choking (good job Wisconsin!).

Measuring who’s winning games is a fair enough way to compare leagues. But it doesn’t really measure the competitiveness of the league (that is, how hard of a conference it is in which to compete). Record comparison in non-conference games relies on goals scored, which has a huge component of luck.

I have a better idea. Let’s compare conferences by possession percentage – that is, by the percentage of the total shots they take vs. the other team. This is an established proxy for how much clock time said team is controlling the puck. Looking at this metric will tell us which teams/conference is most dominant and most controlling of the puck vs. other teams/conferences. And when you control the puck more, you shoot more, and you score more. The leagues with the highest possession numbers against other leagues are going to be the leagues with the most dominant teams in terms of gameplay, and thus the toughest leagues in which to play.

I grabbed a NCAA game data through Nov. 25, 2014. Looking at non-conference games only, I tallied up the number of shots each league has taken against each other league:

Inter-league shots by conference

Team shots vs. AH vs. B1G vs. ECAC vs. HE vs. NCHC vs. WCHA Total Shots
Atlantic Hockey 207 305 252 102 184 1050
Big Ten 360 34 601 325 435 1755
ECAC 448 22 555 231 122 1378
Hockey East 354 562 543 174 202 1835
NCHC 147 425 332 180 360 1444
WCHA 158 400 132 168 357 1215
Total shots vs. 1467 1616 1346 1756 1189 1303 8677

From there, it’s easy to calculate the possession numbers:

Inter-league puck possession

Team shots vs. AH vs. B1G vs. ECAC vs. HE vs. NCHC vs. WCHA Total Shots
Atlantic Hockey 36.51% 40.50% 41.58% 40.96% 53.80% 41.72%
Big Ten 63.49% 60.71% 51.68% 43.33% 52.10% 52.06%
ECAC 59.50% 39.29% 50.55% 41.03% 48.03% 50.59%
Hockey East 58.42% 48.32% 49.45% 49.15% 54.59% 51.10%
NCHC 59.04% 56.67% 58.97% 50.85% 50.21% 54.84%
WCHA 46.20% 47.90% 51.97% 45.41% 49.79% 48.25%
Total shots vs. 58.28% 47.94% 49.41% 48.90% 45.16% 51.75% 50.00%

So, in short, as of Nov. 25, the best possession conferences are:

  1. NCHC, 54.84%
  2. Big Ten, 52.06%
  3. Hockey East, 51.10%
  4. ECAC, 50.59%
  5. WCHA, 48.25%
  6. Atlantic Hockey, 41.72%

So yes, the NCHC is probably the toughest conference this year. They are dominating in possession, especially considering they are above 50% against every league, and they have taken almost a third of their shots against the Big Ten, the second best possession league. This also means the Big Ten isn’t as hapless as its win percentage would suggest. Meanwhile, the WCHA isn’t as good as its record would suggest. Though we have indeed confirmed that Atlantic Hockey is terrible.

Now, I don’t think that this means an NCHC team is favorite to win a national championship, or that the WCHA has no chance. But I do think your average Big Ten team is slightly better than your average ECAC team, for example. Also, I wouldn’t be surprised to see the NCHC to get five teams into the tournament while the WCHA gets two at most. And I would expect probably one more Big Ten team than ECAC teams.

I could do this same kind of comparison for shot% and save%, but I think NCAA hockey possession is the most helpful in determining which conference has the toughest teams, and the most equitable and data-rich way to compare conferences on shot-based metrics. Shot and save % will tell you more about individual talent, which really varies team to team, even within leagues.

Also, I would say I think poll voters should start taking these metrics into account, but that would require me to pretend that polls matter.

A mathematical 2015 NCHC prediction

Working toward a 2015 NCHC Prediction

It’s that time of year when NCAA hockey writers are prognosticating about the upcoming season. Naturally, NCHC fans are all abuzz about their 2015 NCHC prediction. I’m not a hockey writer, but I do enjoy the advanced analytics that are swiftly entering the hockey universe. I also love NCHC hockey.

It’s difficult to do any fancy stat work for NCAA hockey (mostly because of a lack of data), but on a team level, there’s enough to work with. Therefore, I’m going to attempt to make an NCHC prediction that’s based purely the numbers (aka the actual expected performances of the teams) instead of qualitative assessment.

(Also I’m going to rely on my wife Taylor, a PhD student who is much more skilled with statistics and regression than I ever will be. While I had an ok handle on the theoretical framework, I was coming up with some pretty wacky-but-reliable models, and she destroyed them and found models that were both statistically sound and even better fits than mine, so I owe this whole thing to her.)

The first step in this process will be to design the simplest-possible model that accurately predicts last year’s NCHC final standings. What a season, right? Miami, picked by writers to win the league, finished dead last. UNO, picked last, finished third. Denver, who finished sixth in the league, won the conference tournament! Surprise!

But were these results all that surprising? Let’s find out, and let’s use some actual, publicly available game data to do so.

Continue reading