Should all players be treated the same? Probably not.

A very experienced NHL coach once said something to the effect of:

“The most difficult players to coach are those in the middle of your lineup. Your best players will always be your best, and your depth players are usually just happy to be there. Catering to the second and third liners is the toughest thing, because they’ll often have a different opinion of themselves than you.”

When I first heard this, I thought it made a lot of sense from a psychological point of view. In every team I’ve been a part of, there were three groups of players: those who were on the powerplay, those who wanted to get on the powerplay, and those who know they’ll never be on the powerplay. Usually the second group experiences the most friction with the coaching staff, and I am speaking from personal experience.

But I think this idea also has some implications in terms of analytics. From what I have gathered from tracking games at the NHL and various amateur levels, elite play drivers (relative to their leagues) tend to do well in CorsiFor% day in, day out, regardless of deployment, Quality of Competition or Quality of Teammates.

For a lack of better word, their puck possession metrics are “sturdy” – whether you’re talking about Sidney Crosby, Patrice Bergeron, Joe Thornton, Pavel Dastyuk or some of the players I’ve had the pleasure of working with in person, you can generally expect a nice, consistent level of performance from them. Oftentimes they’re great. Occasionally they’re just okay, but they always hold their own.

Another type of consistency altogether can be found in the bottom-line players on most teams. The grinders, muckers and enforcers in this world don’t generally move the needle much when it comes to puck possession. Their coaches may decide to be satisfied with the other talents they bring to the table, or choose to swap them out with someone better at the first available opportunity. In any case, there isn’t much room for surprise.

The biggest challenge for coaches looking to use analytics in determining roster makeup, then, is how to deploy middle-six forwards and middle-pair defensemen – players with very nice skillsets and a potential to really drive the play, but who are liable to get caved in if used with the wrong linemates and in the wrong situations.

Here I’m talking about guys like Nail Yakupov, David Desharnais, P-A Parenteau, Dion Phaneuf, Tom Gilbert or Patrick Wiercioch – players who could help any of the 30 NHL teams, but who need a soft touch when it comes to maximizing what they are good at and covering up what they aren’t so good at.

These players are usually the type who find themselves in the doghouse – they are not good enough to excel in absolutely every situation, but they are good enough for people to (rightfully) harbor expectations on their performance.

Sometimes a mediocre second-line center makes for a great third-line winger. Sometimes a top-pair defenseman getting overwhelmed can star as a sheltered second-pair D-man. Sometimes a healthy scratch can keep up with a team’s best player, allowing other good players to shore up weaknesses elsewhere. Eliminating some of that guesswork can pay big dividends.

There are some decent statistical ways to investigate and express this phenomenon (tracking the variance of individual player CF%s, for instance), and it is something I want to dig into a bit more as the year goes on. But mostly I see this as the impetus for a philosophical discussion that members of a coaching staff should have amongst themselves.

I don’t think it’s ever a bad thing to ask your players to be consistent in effort and preparation, but only your best (and worst) players can really be consistent in their results. For everyone else, it’s up to coaches, armed with the right information, to put them in the best position to express their unique talents.

***

Jack Han is the Video & Analytics Coordinator for the McGill Martlet Hockey team. He also writes occasionally about the NHL for Habs Eyes on the Prize. You can find him on Twitter or on the ice at McConnell Arena.