2015 NCHC Model Check-in

massa

Now that each NCHC team has played at least one league series, let’s look again at the predictive model we developed. In September, before the season, I made the following predictions:

2015 NCHC Prediction

Team '15 Shot% '15 Save%  '15 Poss% '15 Points* '15 Finish
North Dakota 11.26%  92.03% 49.50% 48 1
Nebraska Omaha 10.82% 90.45% 49.50% 41 2
Minnesota Duluth 9.05% 89.35% 54.50% 36 3
St. Cloud State 10.02% 90.21% 49.00% 36 3
Denver 9.44% 91.45% 47.50% 36 3
Miami 7.81% 90.90% 52.75% 34 6
Western Michigan 8.68% 90.45% 48.50% 30 7
Colorado College 8.31% 89.72% 48.75% 27 8
ALL NCHC 9.43% 90.57% 50.00% 288
*All 14-15 expected points +-4.85

Six week into the season, how are these predictions holding up? To be honest, it’s way too early to assess, but it’s not too early to estimate when evaluation becomes appropriate.

But just for fun, let’s look briefly at how each team has performed since the start of the season, especially in terms of the model’s three input metrics. Note, though,  this looks at all games, not just the NCHC. This is just to get a better general idea about how the teams are playing, as of Nov. 9, 2014:

hlogo-CC  Colorado College

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
2-6-0 15 35 236 268 6.36% 86.94% 46.8%

Well, probably a bad place to start, for us and for CC. Low on shot percentage, save percentage, and possession share. They’re doing even worse than I predicted in all three areas so far. We knew this was going to be a rebuilding season, but it could be a long rebuilding season for Tigers fans, especially in NCHC play. I predicted CC would finish at the bottom of the conference, and I still feel pretty secure in that.

hlogo_DEN  Denver

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
5-3-0 22 21 267 226 8.24% 90.71% 54.2%
Denver is lagging a bit in shot and save percentage so far, but they’re more than making up for that in a possession turnaround. All that returning experience is paying off. I still think they’ll make a run at home ice, but they can’t keep giving up 7- and 10-goal weekends like they have to Duluth and Western Michigan, respectively.

hlogo_MIA Miami

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
7-3-0 32 20 396 244  8.08% 91.80% 61.9%
I picked Miami to finish sixth in the NCHC this year in the prediction I am mostly likely to regret. Admittedly, they have played better than expected. Their shooting percentage is still low – not unexpected considering how low it was last year. But their goaltending is improved so far, and HOLY SMOKES their possession numbers are dominant. They know how to control the play of the game. We’ll see how league competition pans out for Miami; last year, they started strong out of conference but tanked. The data suggests any team that can limit Miami’s shooting can beat them.. but good luck with that.

hlogo_UMD Minnesota-Duluth

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
6-4-0 35 24 311 294  11.25% 91.84% 51.4%
I’m sticking to my guns in saying this will be a good NCHC year for Duluth. Their shooting is great so far, their rookie goalie Kasimir Kaskisuo is a phenom-in-the-making, and their possession numbers are high, though not necessarily high for a UMD team. I feel pretty good about the Bulldogs hosting a first-round NCHC playoff series.

hlogo_UNO Nebraska-Omaha

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
6-1-1 26 15 196 235  13.27% 93.62% 45.5%
Nobody expected UNO to be good this season? Ahem. For the record, I predicted they’ll finish second in the NCHC. I still think so, because their early numbers suggest this team is a lot better than anyone thought. The Mavericks can play with anyone – UNO currently ranks 4th in RPI. Shooting is phenomenal. Goaltending (especially Ryan Massa) is solid so far. Possession is down a lot more than I thought it would be, but perhaps that’s to be expected with 11 freshman. I would be dumb not to note that this kind of performance is unsustainable and will regress, especially in the NCHC. The only question is how hard.

hlogo_UND North Dakota

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
7-1-1 35 18 301 237  11.63% 92.41% 55.9%
Anyone else watch the Air Force game? This is a dangerous North Dakota team. Sharpshooting, good goaltending, game control – but I expected all of that. Currently ranked second in the country and fifth in the RPI, North Dakota remains the team to beat in the NCHC, and I believe they’ll win the league, although maybe not as easily as I once thought. We’ll learn a lot about this team in their series against Miami this weekend.

hlogo_SCSU St. Cloud State

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
3-5-0 18 23 248 215  7.26% 89.30% 53.6%
Who is this St. Cloud team? I think they’re still figuring it out. So far, they’re not meeting any expected metrics, but they are inconsistently up on some and down on others. Will these level out? I have no idea. Also, let’s remember that they’ve played Colgate, Union, Minnesota and Minnesota-Duluth. Show me a more difficult opening schedule in NCAA men’s hockey. League play should be a different story for the Huskies, but I still expect them to just miss home ice.

 hlog_WMU Western Michigan

Record Goals For Goals Against Shots For Shots Against Shot% Save%  Poss%
3-5-0 22 24 234 228  9.40% 89.74% 50.6%
The Broncos will play you tough, and they’re not an easy out. They are performing better than expected so far this season, especially in possession share, but is it enough to make a difference? I think Western plays the spoiler this year. They’ll steal a few games, like they did from Denver this weekend, but I can’t see them getting home ice this year. They’re a decent team. But in this league, being decent is not enough.

Final note

If it seems like most NCHC teams are performing above average to start the 14-15 season, it’s because they are. The NCHC has so far been the best conference this year (six teams in the top 20), and it is clearly has the best teams from top to bottom. We will only really know how these teams compare, and how their performances match my predictions, once league play has been happening for a while. After all, the model only considers NCHC games, precisely because it wants to predict NCHC performance, and precisely because we know all of these teams will do well against teams from other conferences.

So let’s look at performance only in NCHC games, and let’s see what the model predicts if we plug in the current, limited 14-15 NCHC data:

2015 NCHC Prediction – Nov. 09 (at least 2 games per team)

Team Shot% Save%  Poss% Exp. Points Exp. Finish
Nebraska Omaha 16.00%  96.49% 46.7% 80 1
North Dakota 13.51% 94.55% 57.4% 78 2
Miami 9.15% 93.33% 59.3% 58 3
Minnesota Duluth 9.64% 92.97% 51.6% 47 4
Western Michigan 10.81% 86.99% 47.4% 27 5
Denver 9.60% 87.22% 48.4% 24 6
St. Cloud State 4.84% 88.00% 55.4% 15 7
Colorado College 4.35% 88.46% 37.1% -6 8
*All 14-15 expected points +-4.85

Obviously this makes no sense whatsoever. For starters, a team can’t earn negative points. Also, there are only 288 points up for grabs, not 323. The shooting, save and possession percentages ARE accurate, but they are so crazy high and low in some instances because there simply haven’t been enough games to provide a good prediction of final stats and final standings.

So…. when will there be?

I took a look at last season’s model prediction, week by week:

 model-chart-20141109

In addition, here are the correlations to the actual final points, week by week:

  model-correl-20141109

The model needs stable shot%, sv% and possession, which didn’t happen in 2013-14 until every team had played roughly 14 league games. This is also when we begin to see a consistent correlation to the final points (above .8), even though just more than half of points had been earned, and the point predictions finally begin to add up to something near 288, the actually number of points available. For 2014-15, it won’t be exactly 14 games when the model becomes reliable, but it should be around then. Depending on how erratic or stable team performance is this year, we might have a solid final prediction for 14-15 by the end of January. This is just eyeballing, by the way, and not “real” statistics.

How is that useful? Does it tell us anything the standings don’t tell us? Sure it does. Last season after 14 NCHC games, five teams had between 23 and 26 points, a virtual 5-way tie for first. But by that point in the season, the model is already pointing toward SCSU as regular season champs and 47 points (they finished with 48), and a tight race for the middle with UNO finishing slightly higher (42, finished with 42) and Denver finishing slightly lower (38, finished with 35).

Also, if the model predicts a team should be doing better than it is for 2015, we might guess the team is a victim of bad luck or is underperforming. And vice versa for a low prediction.

I’ll update the shot%, save% and possession after each week of play here, so we can all keep track of where these teams should end up. It probably won’t matter much until January, so don’t pay too much attention until then, but these early numbers might give us a good idea of whether teams need to make serious adjustments for the gauntlet of season play.

Leave a Reply

Your email address will not be published. Required fields are marked *