National Preseason Rankings: No. 51-61
Every offseason, I enter a bunch of data into a massive spreadsheet in an effort to project the KRACH ratings for the following college hockey season. Below is a portion of these rankings, which will be published in multiple articles over the next two weeks. Obviously, ALL of these projections are based a typical 34-game schedule with non-conference games included, and that won’t be the case in 2020-21.
The model takes into consideration a number of factors, including:
Goals returning (including an exponent for expected development)
Save percentage returning
Shots for returning
Shots allowed from the previous season
Transfers (in and out)
Freshman impact (freshman impact is based on the number of freshmen entering the program, and their rating -relative to the average - as scored by Neutral Zone.
Note: This is a free preview of our paid content. The rest of these rankings will be for subscribers only
51. Colgate (-8 spots)
Last Year: 58.7 (43rd)
Projected: 28.7 (51st)
OK, why does the model think Colgate drops so much?
Well, first, the Raiders only return 34 goals (44.74%) which are fewest in the nation. The loss of all that offense doesn’t just make the model predict that the Raiders will score fewer goals, it also predicts that they’ll have the puck less, therefore, allow more shots.
We’ve predicted a goal differential of 64-103 (-39), while last season Colgate’s differential was 76-87 (-11). What ultimately hurts Colgate the most in our model is the lack of goals returning.
Now, Colgate’s recruiting class rates out slightly above the national average. We’re projecting that Colgate’s freshmen score 18 goals total. If that class scores more than that, then expect Colgate to fall less than we’ve projected. But, I think losing 56% of their scoring is going to be hard to overcome.
52. Union (+5 spots)
Last Year: 23.8 (57th)
Projected: 25.8 (52nd)
Union is interesting. The model predicts some improvement, but because we’re only using last year’s data, we aren’t accounting for any production from Jack Adams, a Detroit draft pick who missed all of last season due to injury. Adams had 22 points as a sophomore, including 10 goals. Just adding those 10 goals to Union’s projected goal total would jump the Dutchmen to No. 46 in the nation, and mark an 11-spot improvement.
The other important piece for Union is the return of goaltender Darion Hanson.
Union was outscored 67-112 last season (-45) and we projected that it will improve to 74-104 (-30) this season, but again, that doesn’t include any production from Adams.
53. Robert Morris (-2 spots)
Last Year: 29.4 (51st)
Projected: 24.4 (53rd)
The Colonials lose a workhorse between the pipes in Justin Kapelmaster, but Dyllan Lubbesmeyer has played very well when called upon the last few seasons. Robert Morris also brings in two top-notch recruits in defenseman Brian Kramer and winger Randy Hernandez.
Still, the data predicts a small drop.
Last season’s 84-104 (-30) goal differential is projected to fall to 86-109 (-33).
54. Brown (-2 spots)
Last Season: 28.4 (52nd)
Projected: 23.4 (54th)
As you’ll see will be the case with a few teams down at the bottom of our rankings, our model projects that Brown will be pretty much the same team it was last season. Brown was outscored 52-84 last season and our model predicts a 58-90 goal differential this season. The Bears went 8-21-2 last season and our model projected an 8-26-0 record.
Goaltending is a big question for Brown with the graduation of Gavin Nieto.
55. St. Lawrence (+4 spots)
Last Season: 12.3 (59th)
Projected 25.9 (55th)
St. Lawrence was outscored 64-130 last season (-66) and our model predicts a big improvement this season. Due to its returning goal scorers, we’re predicting SLU’s goal differential to be 80-103 this season (-23). The improvement in goals against is due in large part to an expected improvement in overall skill leading to more possession, and then less shots allowed.
As for records, St. Lawrence went 4-27-5 last season and our model is predicting an 11-23-0 season in “normal” 2020-21.
56. Princeton (-2 spots)
Last Season: 25.5 (54th)
Projected: 22.7 (56th)
As you’ll read about with Holy Cross below, the Princeton data doesn’t suggest a big swing one way or the other. They return about the national average in goals (3.63% less than the average). We’re projecting an improved save percentage, though. The Tigers started three different goalies last season but Jeremie Forget, by far, had the most success (.912).
Our model projects a 64-100 (-36) goal differential for the Tigers after posting a 66-100 goal differential last season.
57. Holy Cross (-1 spot)
Last Season: 24.5 (56th)
Projected: 23.7 (57th)
Holy Cross’ data doesn’t suggest a big swing one way or other. The goals returning to the roster are only two percentage points below the national average and the program’s projected save percentage is just about at the national average as well.
What that translates to is a projection that predicts Holy Cross will be very similar to where it was last season. The goal differential last season was 88-115 (-27), and this season our model projected that it would be 83-108 (-25). The Crusaders went 11-21-5 last season, and our model predicts them to go 12-22-0.
58. Dartmouth (-16 spots)
Last Season: 60.8 (42nd)
Projected: 20.4 (58th)
Chalk this one up as the first big surprise. I’m posting these in the order that the data dictates, and I can explain why the model predicts such a sharp drop for Dartmouth. Personally, I’m not expecting them to drop this much, especially with a new, and very smart, coaching staff. But, the numbers are what they are.
Dartmouth was hurt in the model due in large part to losing 21 early goals when Drew O’Connor signed early. If you add those 21 goals back onto the roster, Dartmouth’s projected KRACH ranking jumps to 48.8, and would project them to finish No. 43 in the nation.
And in reality, they lost more than 21 goals, because O’Connor would have improved entering his junior year. If he returned, our model would have projected him to score 24 goals this season.
You add all of that offense lost (28% more than the national average) to the fact that the Big Green will have a new starting goaltender, and our model believes that’s going to result in a big drop in goal differential. Last season Dartmouth was outscored 93-106 (-13) and our model predicts a differential of 66-138 for this season.
Again, is this model perfect? No. And I don’t personally believe that Dartmouth will fall this far, but I do agree that they will take a step back.
There are also a lot of unknowns, particularly at goaltender. We’re projecting that the Big Green post an .895 save percentage. If the starter — say it’s freshman Clay Stevenson — posts a .915 save percentage instead, then Dartmouth would jump up to No. 49 in our rankings. It’s still not great, but it would only be a seven-spot drop from last season despite losing a sophomore who was a 21-goal scorer to the NHL.
59. Mercyhurst (+1 spot)
Last Season: 7.9 (60th)
Projected: 18.9 (59th)
The loss of Garett Metcalf can’t be understated, because he kept a bad team in a lot of games last season. The Lakers are going to need someone — and my guess is that it will be freshman Kyle McLellan — to step into that massive void.
Elsewhere, Mercyhurst does return a little more than 80% of its scoring. The only problem is, they didn’t score a lot last season.
Still, our model predicts that the Lakers’ goal differential will improve a whopping 61 goals, from 68-158 (-90) to 90-109 (-29).
60. Canisius (-5 spots)
Last Season: 25.0 (55th)
Projected: 17.0 (60th)
Canisius returns 60.22% of its goals from last season, which is almost 12 percent less than the national average. The Griffins had a team save percentage of .890 last season but freshman Jacob Barczewski posted a .905 save percentage in his 28 starts. Our model projects that their save percentage will improve over last year, but it won’t be enough to bridge the gap with the lost scoring.
Canisius was outscored 93-122 last season and our model predicts that they’ll get outscored 78-131 this season (the model expects that, given the players lost, they’ll allow more shots this season which is why it’s projected that they’ll allow more goals despite improving overall save percentage). The model doesn’t account for ties, but did project that Canisius will finish with a 9-25-0 record, and last season the Griffins finished 10-26-6.
61. Long Island (n/a)
Last Season: n/a
Obviously, there’s no data on LIU, so they begin the season ranked No. 61 by default. But, there’s the potential for the Sharks to be better than last in their first season, and that’s all due to the work head coach Brett Riley put into assembling his roster.
Garett Metclaf is a more-than-capable goaltender who is used to being under siege, transferring in from a Mercyhurst program that allowed the most shots in the nation last season. In fact, all three of LIU’s goaltenders are transfers, so all of them have collegiate experience.
I wrote about Nolan McElhaney the other day, and I think he has the potential to be a stud for LIU if he can stay healthy and Mitch Meek brings a lot of experienced minutes to the defensive corps. as a grad transfer.
It feels like LIU has built its roster from the net out, which is obviously smart. Again, there’s no data to consider here, but I think they’ll have nights where they’ll cause problems for opponents. The question will just be whether or not they can score enough goals to cause those problems consistently.