College hockey’s overtime rules still don’t make sense and are worse than they were before

For many years — including nearly two decades after the NHL made the switch to a 4-on-4 (and later 3-on-3) overtime — college hockey maintained its traditional tiebreaking formula, with all games* that ended in a tie proceeding to five minutes of OT and then ending in a tie for national record keeping purposes.

(*except for the Beanpot... more on this later)

From there, conferences had a mix of ways to break ties for conference points — including adding an extra overtime session of 4-on-4 or 3-on-3 after the end of the OT, and/or going straight to a shootout. These results didn’t impact the Pairwise rankings that select the NCAA tournament, but they did impact conference standings, and added some entertainment factor in at the end of tie games.

While it was a little strange and confusing that different conferences had different rules, this system largely served the purpose of keeping things fair and easy to understand for Pairwise purposes.

(One oddity under the old system was that in-season tournaments such as the Beanpot had the option of playing continuous overtime, guaranteeing that a game would have a winner and a loser for Pairwise purposes. This was legislated out in 2019-20, requiring the Beanpot to play an ‘official’ 5 minute overtime that counted in the national rankings before moving on to a just-for-funsies series of 20 minute OTs for advancement/the trophy.)

Many observers, myself included, advocated for years for college hockey to update its overtime rules to align with the NHL, and drop the 5-on-5 followed by a tie system. While I personally have no problem with ties, 1) 4-on-4 and 3-on-3 hockey is fun, and fun things are good; 2) this is basically the standard at every major level of hockey now, and there’s no real reason for NCAA hockey to be the only major-ish competition with a rule set stuck in the late 1990s.

But as it worked out, it wasn’t until 2020-21 that a new system was adopted. The problem is that the new system is somehow more jumbled up and confusing than it was before.

College hockey adopted standardized 3-on-3 overtime for this season at the conclusion of 60 minutes of tied hockey. A (perfectly reasonable) argument against adopting this standard is that the vagaries of 3-on-3 or shootout hockey in the short sample size of a college season is a bit unfair to weigh heavily in the national rankings. So what they instead worked out was a solution that is, to me, sub-optimal, and difficult to comprehend.

(1) 3-on-3 overtimes ‘count’ toward a team’s win-loss record in the official history books. However, for pairwise ranking purposes, a 3-on-3 OT win counts as just 55% of a “win” and a loss is 45% of a “loss.” The net result of this is that overtime, while fun to watch, has an incredibly small impact on the final rankings and is not much different in the grand scheme of things from the game ending in a tie.

(2) Tie games after 3-on-3 go to a shootout, but those results have no bearing at all on the national rankings.

(3) This is the part that I really hate - In order to accurately document a team’s record, you basically have to both list its win-loss-tie record inclusive of 3-on-3 OTs, as well as then break down how many OT wins/losses and shootout wins/losses it has. For instance, a team that has played six games, won 2 in regulation, lost 1 in regulation, lost 1 in overtime, won 1 in overtime, and won 1 in a shootout, has their record listed for posterity as 3-2-1, but to be correctly documented in the standings, you have to show that they are 2-1-3 over 60 minutes, 1-1 in OT, and 1-0 in shootouts.

I follow this stuff very closely and I find it to be somewhere between confusing and impossible to read and understand the standings.

(3a) And now, according to Jimmy Connelly of USCHO, it looks like the NCAA selection committee is basically not even going to take 3-on-3 overtime results into consideration at all as it selects the NCAA tournament field except as a tiebreaker, even though the results are considered “wins” for historical record-keeping purposes.

While I get the concerns of coaches and programs who think it’s unfair to put too much weight on 3-on-3 overtime, at the end of the day it either needs to count or it doesn’t. The right rules for OT for college hockey should consist of one of the following:

(1) Overtime counts. Whether you want it to be 3-on-3, or 5-on-5, or whatever — a win is a win and a loss is a loss, for both your record and for tournament selection purposes. Then shootouts can be held just for conference points.

(2) Overtime and shootouts count in the records, but regulation wins are also added in as a new, heavily-weighted factor in the pairwise rankings — giving a boost to teams who take care of business within 60 minutes. This is probably the fairest solution.

(3) Overtime doesn’t count. The result after 60 minutes determines your national record and ranking, and then conferences can do whatever they want for an extra league point to entertain the crowd. 2-on-2 overtime. Slam dunk contest. Spelling bee. Whatever.

In a sport that struggles to connect with the casual fan it’s hard to see what is accomplished by having such a complicated system that is confusing even to dedicated followers. Right now, 3-on-3 OT kinda-sorta counts but not really, and then the shootout doesn’t count at all. It’s a whole lot of to-do for very little impact. Let’s just standardize the system and make it make sense for everyone.

My preferred solution would be the second one, and for college hockey to be a little bit innovative about how it does OT: ten minutes of 3-on-3, which counts, and then the game ends in a tie if nobody scores (conferences can do shootouts if they want).

As anyone who has watched much 3-on-3 knows, 10 minutes of 3-on-3 would end an overwhelming majority of games with a winner and a loser. Given that college only plays ~2 games a week, it’s not an insane ask on the players the way it would be to adopt something like this in the NHL.

There are those who will say putting such a heavy weight on 3-on-3 is unfair to teams that might not be built to play that way, which is true. One counterpoint is the ability to make regulation wins — or RPI only factoring in regulation — a new factor in the Pairwise rankings, giving teams an opportunity to move up on the basis of their regulation results.

But my larger counterpoint to the argument against 3-on-3 is basically that that life isn’t fair and neither is hockey. Games that have a ton of penalties because they were assigned the ref who loves to hear himself blow the whistle aren’t fair to teams that are built to play well 5-on-5, but the vagaries of how officials call games absolutely impact the final NCAA rankings. Think back to 2017, when a falsely disallowed BC goal in the Beanpot consolation game was essentially the difference between BC qualifying and not qualifying for the NCAA tournament. (The good news is that no, I’m definitely not still bitter about that — not at all.)

3-on-3 hockey is a normal part of the sport at this point and something teams should practice and be good at. If we don’t want it to count, then don’t make it sorta count! Let’s either count it, or don’t.