# NCHC Prediction 2016-17

Welcome back, NaCoHoCo fans. It’s been a long offseason – partying for some of us (North Dakota, Denver), pangs for others (Omaha, Miami), pleading for others still (Arizona State, Mankato). But the days are getting shorter, the lattes are getting pumpkin spicier and the rev of distant zamboni motors grows and grows. Let’s talk college hockey, eh?

It’s destined to be another exciting year in the National Collegiate Hockey Conference. Sure, it’s tough to beat a national championship season. How about another one? And sure, the conference lost a lot of talent to graduation and pro contracts, but there are incredibly talented goaltenders, forwards and defenders returning for another year of play. How will it all shake out?

The writers and experts have had their say, and once again, I’m taking a stab using shot data from last season to predict how the conference race might unfold this year.

We haven’t always agreed, these hockey writers and me (me being “data”, I mean). In the last couple of years I’ve been doing this, the data has uncovered some interesting trends that bucked conventional wisdom. Two years ago, this statistical model helped predict an insurgent UNO Maverick team (that ultimately made the Frozen Four). Last year, it pegged Denver as the conference favorite – wrong, but the Pios did make the Frozen Four.

Sometimes data finds the trends that we don’t normally see. Other times, it simply confirms what everyone already knows. This year, as we’re about to see, is one of those years.

I’ve collected individual-level data on all NCHC players from 2015-16, primarily goals, shots, shot%, save% and a derived possession-share (individual shots/all shots). That data is readily available thanks to better tracking by the NCHC and more in-depth shot statistics compiled by College Hockey News. (P.S. to NCHC’s marketing team: I love the new website and data page – huge improvements. Someone’s been reading?)

As is tradition, we’re going to adjust sh%, sv% and possession for each team based on what we know about roster changes, particularly about who is returning and who has left. To do this, I have to make some assumptions about players and teams. I’ll try to keep these as safe as possible:

# NCHC 2016-17: Returning Defense

This week I’ve been looking at returning 2016-17 NCHC talent. I’ve evaluated the goaltenders and forwards. Today, let’s tackle the defenders.

In the forwards article, I mentioned just how much offensive talent has left the conference this year. But with the exception of a trio of North Dakota juniors, defensive talent largely sticks around. Coupled with the solid goaltenders who remain, this could be a year of defensive chess matches in the NCHC – very exciting stuff.

Before looking at any data, however, we need to have a quick discussion about the analysis itself, because this category is always more subjective than the other two. When we’re evaluating defensive players in hockey, we tend to conflate their actual defending abilities with their offensive contributions.

Defensemen are the most difficult position to assess in hockey, at all levels. You can evaluate them in the same way you do forwards, but that only tells you who the most offensive-oriented guys are. Trying to determine the most defensive defensemen can be difficult, especially with the lack of data we have at the NCAA level. For so long, plus/minus was the standard, but the stats community has come to a consensus that +/- is unreliable and useless. At the NHL level, two-way blue line talent can be looked at through ice time, relative Corsi, player usage charts, etc., (see here, here and here) but we simply don’t have that kind of data in college yet. We’re stuck with shots, shot blocks, faceoffs, goals and assists.

We’ve tried to make do with what we have, knowing that we still need a better way. But in working with the extant data, we can do a pretty good job of evaluating who is helping the team score goals from an offensive perspective, and we might be able to infer some things about who is actually playing good preventative defense. We’ll return to this discussion at the end of the article, because there are a few more preferable indicators of good defense (and they’re not that hard to get at), but it would take some investment from the NCAA and the conferences.

For now, let’s play with the data we’ve got.

### Top Losses

NCHC teams lose 17 defenseman in 16-17, just one less than last year. Let’s take a quick look at the top departures before getting to returning players:

Team Player Year Total Shots Blocks Expected Points Actual Points Rating
SCSU Ethan Prow Sr 136 73 29.19 38 1.30
NDAK Troy Stetcher Jr 253 54 31.28 29 0.93
DEN Nolan Zajac Sr 198 77 18.13 20 1.10
UMD Andy Welinski Sr 214 49 20.52 19 0.93
NDAK Paul LaDue Jr 216 51 21.60 19 0.88
NDAK Keaton Thompson Jr 134 34 16.38 17 1.04
UNO Brian Cooper Sr 133 64 16.16 16 0.99
MIA Matthew Caito Sr 119 40 18.17 11 0.61
UMD Willie Corrin Sr 106 33 6.20 10 1.61
MIA Chris Joyaux Sr 58 40 10.85 6 0.55

Last year, I suspected Ethan Prow would be the best defender in the conference. That held up pretty well – Prow led blue liners in points, and overperformed statistical expectations by about 30%. Similarly, Nolan Zajac and Troy Stetcher – also in my top five – had good years. Matthew Caito was someone of a miss, though. Perhaps he had an off year, but he earned about 40% fewer points than I would have expected. Though quite a few top Miami defenders had poor years – perhaps something about their systems? Could be an artifact, too.

The Fighting Hawks lose the most at defense, as three juniors depart. Interesting, Western Michigan has zero returning defenders, a good sign for them in a year they’ll be seeking a new goalie. Also interesting is that some of the “best” defenders have very even ratings (close to 1.00), which differs from top forwards, who tend to have high ratings. This supports my theory posed last year, which suggests the best defenders cluster around 1.00, or “at offensive expectations.”

Let’s get to it – who are the top returning defenders in the NCHC? There are 47 returning this year, just one less than last year. Let’s look at the top d-men in a few of the more traditional ways – points, blocked shots, and blocks per game.

# NCHC 2016-17: Returning Forwards

Last year turned into a banner year for the National Collegiate Hockey Conference’s offensive players. Two seasons ago, the top 20 NCHC forwards combined for 703 points. Last year? 808 – nearly 15% more scoring. High-powered offense certainly made for more than a few exhilarating tilts.

Now as 2016-17 approaches, NCHC hockey returns but without 14 of the leagues 20 top forwards, six of whom departed early. Adios, Kalle Kossila. See ya, Danton Heinen. Peace out, Nick Schmaltz. Jack Roslovic? We hardly knew ye.

A huge vacuum of talent waits to be filled, but in this league the wait never lasts long. Who’s going to step in for all that lost offensive production? My goal here is to figure that out.

Earlier this week I looked at returning NCHC goaltenders, finding few surprises. That’s not so with the NCHC forwards. Much like with the goalie model Taylor and I developed, we have utilized the new data available from College Hockey News to look beyond the traditional scouting reports. With a more complete picture of the shot statistics available, we can get closer to understanding who’s really changing the game with their ice time, and who stands out as the most effective forwards in the league. I’ll spare you the gory methodology since it’s about the same as last year’s analysis.

Let’s get warmed up by applying that analysis to those NCHC forwards not returning in 2016-17.

### Top Losses

Looking only at guys who played in 50% or more of their team’s games, the NCHC loses 31 forwards, just a few more than last year. As I mentioned above, though, the list is top heavy, and some teams get hit harder than others.

St. Cloud State loses its top five point earners, for starters. The represents 61% of their forward’s scoring from 2015-16, and even for a strong program like SCSU, that’s a tough roster to reload. Denver and North Dakota each lose three of their top five scorers, though for Denver that includes underclassmen Heinen and Trevor Moore. For North Dakota, Nick Schmaltz leaves early, as does Luke Johnson. Most unscathed is probably Western Michigan – losing only 15% of their scoring from last season.

Teams losing their top-scoring forward include Colorado College (Hunter Fejes), Denver (Heinen), Miami (Roslovic), St. Cloud (Kossila), Duluth (Tony Cameranisi), and Omaha (Jake Guentzel). Only Western Michigan and North Dakota return their top forward. Woof.

Let’s warm up by applying the advanced model to the top 10 departing NCHC forwards by points earned:

# 2016 NCHC Prediction

Here it comes – Year Three of the NCHC. Once again, the eight-team conference looks to be the toughest and most talented in the nation. Most polls put 4-5 teams in the national Top 10, and, having sent 75% of the conference to the NCAA Tournament in March, the NCHC looks poised to return at least half the conference this year.

As the 2016 NCAA hockey season gets underway, it’s time to predict the final 2016 standings of the National Collegiate Hockey Conference. Per usual, I will be doing this using only actual statistical data based on each team’s past performance.

This only gets tougher as nearly every team in the conference has proven a national contender over the last few years. Already the teams are so talented that the marginal differences between each team are so slim – any team could beat any other team on any given night (yes, even CC).

Last year when I did this, the statistical method of predicting did slightly better than all of the NCHC hockey journalists. There’s no guarantee I will do as well this year though, so the better my odds I made a few adjustments to the model to try and get an even more accurate prediction.

I’ve collected individual-level data on all NCHC players from 2014-15, primarily goals, shots, shot%, save% and a derived possession-share (individual shots/all shots). That data was much more readily available thanks to better tracking by the NCHC and more in-depth shot statistics compiled by College Hockey News.

As we did last year, we’re going to adjust sh%, sv% and possession for each team based on what we know about roster changes, particularly about who is returning and who has left. To do this, I have to make some assumptions about players and teams. I’ll try to keep these as safe as possible:

Welcome to a long thorough analysis of my analysis – the NCHC model for predicting final standings. Before last season, I created a model to predict the NCHC final standings. In this post, I’m re-examining my methods and assessing their validity. Honestly, I thought I did OK last year, considering I did better than most of the expert writers and media. However, my model didn’t exactly nail the final standings, so there’s room for improvement. Today I’ll take a look at the components of the model and whether they worked as intended. Later this week, I’ll bring it all together for a 2016 prediction.

Be warned – this post is methods heavy. If you have an interest in NCAA hockey analytics, read on. If not, turn back now, and come back later in the week for my stats-based NCHC predictions. Still here? Ok, here we go.

First of all, kudos to the National Collegiate Hockey Conference for providing full-season and in-conference shot statistics for 2014-15. It’s a good start and a huge improvement over last year.

If you’d like to get familiar with the theory behind this model, I suggest reading my post from last year in which I created the NCHC model. Also, before we get started, the usual disclaimers: this analysis uses NCHC data taken from NCHC official records, and only considers intra-conference play during the regular season. Non-conference games, NCHC tournament games, and NCAA tournament games are not included.

### 2013-14 vs. 2014-15

Team '14 Sh% '15 Sh% Δ '14 Sv% '15 Sv% Δ '14 Poss. '15 Poss.  Δ
Colorado College 7.74% 6.81% -0.93% 89.39% 88.65% -0.74% 48.73% 42.10% -6.63%
Denver 9.08% 10.49% 1.41% 92.87% 90.65% -2.22% 44.81% 50.90% 6.09%
Miami 7.60% 8.99% 1.39% 88.90% 90.90% 2.00% 50.55% 55.20% 4.65%
Minnesota-Duluth 8.83% 8.95% 0.12% 89.83% 91.26% 1.43% 53.17% 53.00% -0.17%
Omaha 9.87% 10.15% 0.28% 89.22% 92.63% 3.41% 56.49% 46.50% -9.99%
North Dakota 11.06% 9.61% -1.45% 91.41% 92.75% 1.34% 48.38% 49.70% -1.36%
St. Cloud State 12.46% 9.18% -3.28% 91.14% 91.37% 0.23% 49.15% 53.70% 4.55%
Western Michigan 10.06% 7.38% -2.68% 90.11% 89.51% -0.60% 48.45% 49.80% 1.35%
NCHC Total 9.58% 9.02% -0.56% 90.42% 90.98% 0.56% 50.00% 50.00% 0.00%

The adage says the best predictor of future performance is past performance. Looking at this table, that holds. St. Cloud lost some shooting prowess, and Omaha had better goaltending, but not much jumps out otherwise. Performance was fairly steady across the three categories from ’14 to ’15 – except possession. Wild swings there, eh? Omaha lost nearly 10 percentage points in possession share. Meanwhile, Denver and Miami saw great improvements, which was a big reason for their good finishes. I’ll take a look at what might be driving possession swings a little later. For now, when comparing ’14 and ’15 performance, you can group the eight teams into four categories:

# NCHC 2015-16 Returners: Defense

This is part three in a series on returning 2015-16 NCHC talent. Earlier in the week, we created models to evaluate the relative on-ice performance of goaltenders and forwards. Today, let’s tackle the defenders. But before we do, we need to have a quick discussion about the analysis itself, because this category is always more subjective than the other two.

I picked an image of a talented NCHC defensive player (SCSU’s Ethan Prow) making an offensive move for a reason. Prow is a good two-way guy in a lot of ways, and I’ll show you why we think that in a minute. But when we’re evaluating defensive players in hockey, we tend to conflate their actual defending abilities with their offensive contributions.

Defensemen are the most difficult position to assess in hockey, at all levels. You can evaluate them in the same way you do forwards, but that only tells you who the most offensive-oriented guys are. Trying to determine the most defensive defensemen can be difficult, especially with the lack of data we have at the NCAA level. FOr so long, plus/minus was the standard, but the stats community has come to a consensus that +/- is unreliable and useless. At the NHL level, two-way blue line talent can be looked at through ice time, relative Corsi, player usage charts, etc., (see here, here and here) but we simply don’t have that kind of data in college yet. We’re stuck with shots, shot blocks, faceoffs, goals and assists.

We’ve tried to make do with what we have, knowing that we still need a better way. But in working with the extant data, we can do a pretty good job of evaluating who is helping the team score goals from an offensive perspective, and we might be able to infer some things about who is actually playing good preventative defense. We’ll return to this discussion at the end of the article, because there are a few more preferable indicators of good defense (and they’re not that hard to get at), but it would take some investment from the NCAA and the conferences.

For now, let’s play with the data we’ve got.

### Top Losses

NCHC teams lose 18 defenseman in 15-16, whether through graduations or defections. No player will likely be as missed as Denver’s Joey LaLeggia, one of the top five point earners in the league. North Dakota’s Jordan Schmaltz leaves with a year of eligibility, and Nick Mattson graduates – both contributed 20+ points. Colorado College will miss sophomore Jaccob Slavin and senior Peter Stokykewich, who combined for 139 blocked shots last year. WMU’s Kenny Morrison leaves a year early after a relatively fruitful 2014-15, but he certainly could have contributed significantly in the upcoming season. SCSU’s losses of Andrew Proncho and Tim Daly will be felt, too – Daly led the league in blocked shots.

UNO, Miami and Duluth remain relatively unscathed, however, losing only four defensive players between them, and only two who played a full season.

There are 48 returning defenders in the NCHC. As we did for the forwards, let’s look at the top d-men in a few of the more traditional ways – points, blocked shots, and blocks per game.

# NCHC 2015-16 Returners: Forwards

We’re still a few weeks away from the 2015-16 NCHC season, and the photo above sums up how I feel. I can’t wait for some ol’ fashioned hashtag college hockey. But in the meantime, let’s continue our look at the talent that will be returning to the ice. A couple of days ago, we took a more advanced analytic approach to goaltending in the conference. Today, let’s consider the offensive production side of things – forwards.

Much like with the goalies, Taylor and I have utilized the new data available from College Hockey News to get away from the usual points-goals-assists assessment. With a more complete picture of the shot statistics available, we can get closer to understanding who’s really changing the game with their ice time, and who stands out as the most effective forwards in the league. It’s not a perfect analysis, but it’s better than what was possible less than even a year ago. Progress is good.

So we’ll get there, I promise. But first, let’s take a look at who is not returning this year, and which teams have holes to fill.

### Top Losses

Looking only at guys who played in 50% or more of their team’s games, the NCHC loses 27 players. Most heavily hit is undoubtedly Miami, who loses three of the top four point earners in the conference – Austin Czarnik, Blake Coleman and Riley Barber, who is leaving early. St. Cloud’s Jonny Brodzinski also leaves early for the pros, taking 21 goals and 7.9 shots per game (!) with him. North Dakota (Michael Parks and Mark MacMillan) and Western Michigan (Colton Hargrove and Justin Kovacs) both lose a 50-plus-point pair of offensive leaders. Rounding out the top ten, say goodbye to Denver’s Daniel Doremus and Duluth’s Justin Crandall.

Every team lost a few key pieces, however Colorado College and Omaha escape graduation relatively unscathed in the forward department. Most depleted? Arguably Miami, though I could see a case for North Dakota or Denver, too.

Time to evaluate the returning forwards in the NCHC. For the sake of defining the discussion, and because the metrics we’re using are all based on shots, we’re only going to examine those players that took 50 or more shots last season. That will include pretty much everyone in each team’s top three lines, and it eliminates the regular scratches, cleanup lines, etc. This way, we’re more likely to compare apples-to-apples when we start looking at percentages and average performance.

Let’s first take a look at the returning talent in the traditional sense. We know the NCHC lost some big playmakers, but it wasn’t a total turnover. Some teams return a strong core of their point-producing players. Below, I list two metrics that are historically used to evaluate player contributions – points and goals scored.

# NCHC Fridays and Saturdays

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:

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.

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.

# 2015 NCHC Prediction: Testing the Model

In Part One earlier this week, I attempted to devise a model to predict last season’s NCHC standings, using puck possession (shots) and bounces (shot and save percentage) as the input variables. Not only did the formula reasonably predict the regular-season standings, but it appeared to express a strong fit to the data. Once again, that model is:

$\text{POINTS} = b + b_{1}(\frac{g_{f}}{sh_{f}})+b_{2}(\frac{sh_{a}-g_{a}}{sh_{a}})+b_{3}(\frac{sh_{f}}{sh_{a}})+e$

However, I only have one season of NCHC data to work with. It’s possible this is an anomaly, or it’s possible I fashioned a model to fit the standings that we already knew. What I really need is another couple of data sets (aka a few more NCHC seasons) to test this exact model. I don’t have that for the NCHC because there has only been one season so far, but I have something close.

Six of the eight NCHC teams played at least three seasons in the WHCA, so I applied this framework to those three previous seasons, building a regression model that considers all three seasons at once to get a model that is significant beyond the .0001 level. The results:

# A mathematical 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.