Early in the season, results can be very strange indeed.
On opening weekend, the preseason No. 5 team in the country, UMass Lowell, was swept in a home-and-home series to open the season by lowly UNH, a team picked to finish close to the bottom of its own conference. Instead, those Wildcats are now 4-0 to start the year.
And the strange results don’t end there.
Preseason No. 2 BU, with dreams of a Frozen Four appearance, was just swept — at home — by a solid if less-talented Minnesota State club that should run the show in the WCHA.
Preseason No. 6 Minnesota-Duluth has just one win from four games, including going winless this past weekend in a home-and-home with Bemidji State.
Colorado College has the best record in the NCHC. Clarkson is 3-0-1. Union is 0-5. You can go on like this.
But one thing to keep in mind is that systems matter in college hockey a whole hell of a lot more than they do in the NHL, and it takes time to get everyone up to speed on them – especially because, even if there’s not a new coach, you’re still going to turn over at least a few spots on your roster every year. That’s the nature of the game.
So a systems-y team like Lowell, which has one of the best records over the last six years of any team in the nation, is often going to struggle in the early goings. Indeed, over the last seven Octobers, they’re just three games above .500 and have an average goal margin of about plus-0.4. In all other months, they’re a nearly .700 hockey team that wins by more than 1.1 goals per game. That’s just how it goes with more teams than you might think.
The other thing to remember is that in college hockey, some teams are just more talented and get perennially good goaltending, while others do not. That makes it hard to get a true baseline for performance at this time of year, when percentages can be truly all over the map.
Just to return to Lowell — still only 1-3 this year and dropped from No. 5 to No. 19 in just two weeks — the problem is clear enough. They’re still a little above water in all-situations percent shots-on-goal percentage (statistically, that’s the best indicator of future team performance in college hockey) but they’re one of those teams that always has a high PDO. In fact, over the previous four seasons, theirs is the highest in the country, at 102.9.
But through the first few weeks of this season, they’re at just 97.9, which, hey, that kind of explains a lot. They have an okay shooting percentage at 9.7, but their team save percentage is just .882 (Rangers draft pick Tyler Wall, last year’s starter, is currently at .833 in less than 147 minutes), and that’s not gonna win you many games.
The River Hawks serve as a good object lesson in teams off to surprising starts, both good and bad: When you’ve played so few games and faced so few shots, it takes one good night shooting the puck, or a single shutout, to make your stats look unbelievable. And if you’re a little unlucky through two, three or four games? Well, that’s hockey. But it’s gonna have people asking questions.
As always, people don’t like to accept that luck has a huge role in any individual hockey game, and that the kind of luck that leads to a five-goal outburst or something like that often swings small-sample percentages wildly. Let’s put it this way: If Army could reliably shoot 17 percent and get .970 save percentages — those are just about their actual numbers through two games this year — we wouldn’t need to have a season. Hell, five teams right now have all-situations PDOs north of 107, and wouldn’t you know it, a bunch of those teams are also the ones with the really great records mentioned above.
So it’s worth considering: How likely are teams to keep up their percentage-based performances as the season moves along?
The labeled teams here are the ones with some of the biggest disparities between their four-year PDOs and those they’ve put up so far this season, and a few of them are interesting.
Wisconsin, for example, is a team I saw on Friday night and there are two reasons they have improved (though obviously not to the extent that they’re going to sustain a 106 PDO). First, they got Kyle Hayton, an elite goalie at the national level, after getting a combined .896 save percentage over the previous four seasons. That’s going to swing you a lot, and that’s a number that’s unlikely to come down too much. They’re also shooting 13.2 percent, which is too high. For the record, the national average shooting percentage over the past five years is about 9.2, and Wisconsin has the guns to outperform that a bit, but not to the extent they have.
On the other side of the ledger, you see St. Lawrence sitting with a national-low PDO of just 91.1. Part of that comes because Hayton left the program for Wisconsin (funny how that works), which resulted in the team’s save percentage collapsing. Meanwhile, Saints skaters are shooting 3.8 percent through four games. Part of the issue, of course, is that St. Lawrence isn’t that good to begin with, but they certainly have the quality to get percentages closer to the national averages.
The really interesting team here is Minnesota-Duluth, who went to the Frozen Four last year and, while they lost a lot to graduation and early departure, are generally always good regardless. Their big question mark coming into the year was goaltending, and given the team save percentage of .854, they haven’t even begun to answer them. They also were likely to struggle putting the puck into the net, though not to this extent, so here we are.
Even Boston College, with so-so goaltending (for Hockey East, anyway) and little in the way of the marquee offensive talent one might expect from the Eagles, isn’t so bad as to have a PDO two points below 100. They’ve been victimized by tough matchups (to an extent) and only against Providence on Sunday afternoon did pucks start going in for them, but if you had BC booked for some early struggles, they’re delivering.
This is meant only to serve as an illustration of why you can’t put a ton of stock into early-season results, either in terms of wins and losses or how often the puck goes in the net for or against you.
The things you thought about these teams going into the season is still true in the long term, but the unfortunate reality for the teams that are currently struggling is that these are results that matter insofar as teams getting into the NCAA tournament at the end of the year. Conversely, teams like Wisconsin that were likely to make an NCAA bid anyway get the benefit of a few extra points in their Pairwise number.
All these games count the same, and as everyone pulls back toward their long-term averages, we’ll be able to view these games for what they are: A handful of small blips on a much bigger radar screen.