Sunday, May 5, 2024

System could use change

John Markell's team had just been smacked around Joe Louis Arena by Northern Michigan.

Dutifully, the Ohio State head coach withstood a press conference during which he had to explain why his players seemed devoid of urgency in their 4-1 loss to the Wildcats in the CCHA Super Six consolation game.

As would be expected, once the reporters' questions ceased, Markell made a beeline out of the Joe's makeshift interview room.

But seconds later, he popped back through the door and made another beeline - this time to a computer in the back of the media lounge.

A computer. This was a man who had a locker room full of hockey players waiting to be chastised for a lackluster performance. But above that, he was more compelled to see the numbers spit out by a computer program designed to pick the NCAA Tournament.

"Are we in?" Markell asked the kid wearing a red "O" on his hat, typing on the keyboard.

"I don't know," the kid responded.

Walt Kyle, Northern's head coach, soon joined Markell near technology row. The coaches, whose teams were slashing the hell out of each other minutes earlier, were now discussing numbers rounded out to the thousandths with about as much enthusiasm as Math 1825 students.

After several minutes of thinking through scenarios, Kyle simply offered: "Tell you what, I hope we both make it."

Such is life for college hockey coaches in the world of Pairwise Rankings. You think about power-play schemes and forechecking assignments all week, and then once the game is over, you race to a computer and see what the result of the game actually means.

Basically, the Pairwise is a system devised by USCHO.com to predict the NCAA Tournament field and seedings. It factors in several criteria, including head-to-head competition and the confusing Ratings Percentage Index.

It's similar to the Bowl Championship Series in college football, with one big exception. The gridiron numbers only impact one bowl - the national title game - while the Pairwise determines the entire hockey tournament field.

If the Pairwise's numbers say you're in, congratulations. If not, have fun watching on TV. This kind of objectivity is something college hockey prides itself on.

It's simple and easy to understand, but it doesn't seem right to rely on a computer to make or break a hockey team's season. I understand there are many more important things in the world, but it bugs me when teams are viewed as numbers when the postseason arrives.

Just because Team X has a better RPI than Team Y doesn't mean it is the superior team.

The committee did add bonus points this year, rewarding quality nonconference wins. It adds a spice to the Pairwise, but it's still a numerical system. Human reason has to step in somewhere.

The Pairwise generally does a good job at selecting 10 at-large teams for the 16-squad tournament. But why should the NCAA Selection Committee be roped down by a computer?

Human rationale makes a lot more sense than a pile of numbers anyway.

"I'm a Northern grad, so I'm not smart enough to figure that stuff out," Kyle said, half-kidding.

You know what? Good for him.

I'd rather coaches like Kyle not understand how the Pairwise works. Leave that junk to the cyber-monkeys and let hockey guys concentrate on the game itself.

Our hometown Spartans, of course, have no legitimate gripe with the system. Not this year. They shot themselves in the foot too many times.

MSU head coach Rick Comley will tell you his team was one of the five or 10 best in the country at the end of the season - and he's right. The Spartans finished the year on a 15-5-1 run and beat some decent teams.

However, several abominable losses - most of which were several months ago - dogged MSU (23-14-2) all season. The Spartans lost to Lake Superior State and Bowling Green, whose combined record ended up as 8-44-4.

But enough about the Spartans. They know why they're not going dancing this weekend. What's tough to figure out is why certain teams are.

St. Cloud State got a bid despite a 17-15-5 record. Other teams, such as Denver (21-14-6), Minnesota-Duluth (22-15-5) and even MSU, had better records and didn't get in. Despite losing its first-round league tournament series, St. Cloud was buoyed by a high strength of schedule and some bonus points.

Nothing against the Huskies, but they shouldn't have gotten in with a record only two games above .500. Every bubble team had pluses and minuses, but the computer only sees things in black and white. An empowered human committee would be able to see St. Cloud's situation for what it is.

I bet Kyle's Wildcats, who went 22-17-2 and weren't invited, would agree.

The Pairwise is a valuable guide, and the newly introduced bonus points are a good start to make the system more subjective. But as long as computations dictate everything, we'll all be missing out on something.

James Jahnke is running an NCAA hockey pool with a grand prize of 45 Timbits. Send him your brackets at jahnkeja@msu.edu.

Discussion

Share and discuss “System could use change” on social media.