I had a conversation on Discord recently that started with a pretty simple question: When I scout a player, am I looking for specific traits or skills and evaluating how they use them over a series of shifts? The answer is yes, but probably not in the way people traditionally think about scouting.

If you asked me to scout a player by watching a game and then giving him a numerical grade for his skating, hands, passing, shooting, hockey IQ, and all of the other individual categories you normally see on a scouting report, I’d probably be terrible at it. That’s just not really how I see hockey.

I see ten skaters and two goalies playing a violent game of blitz chess on skates.

Everyone is moving, everyone is reacting to everyone else, and every decision changes what is available to the other nine skaters on the ice.

A player’s skating matters because of what it allows him to do within that environment. His hands matter because they might allow him to hold onto a puck for an extra half-second under pressure and create a passing lane that wasn’t there before. His hockey IQ isn’t something I can easily separate from his passing because recognizing where a teammate is going to be is part of what makes the pass possible in the first place. The individual skills obviously matter, but I’m much more interested in how they interact and what they allow a player to accomplish.

I think anyone who has watched enough hockey can walk into a rink, watch a few shifts, and identify which players are good.

The best players usually make themselves obvious. They’re faster, they have the puck more, they make plays other players can’t make, and they stand out. I don’t think that’s the difficult part of scouting.

The difficult part is figuring out how good that player will be three, four, or five years from now. More importantly, what does a player who eventually reaches the NHL have that another very good player doesn’t? That’s the question I’ve tried to answer with my tracking data.

When I watch a player shift-by-shift, I’m manually tracking the sequences that happen while they’re on the ice, but the numbers themselves aren’t really the evaluation. I’m also taking notes on why something worked or didn’t work.

Was a player able to escape pressure because he’s an explosive skater? Did he recognize where the pressure was coming from before he received the puck? Did he manipulate a defender into opening a passing lane? Or did he make a play because he had five feet of open ice and nobody challenged him? Those can all result in the same successful play in my spreadsheet. They don’t necessarily tell us the same thing about the player.

Offensive transition involvement is a good example. Every time a player’s team successfully crosses a blue line with possession, I track whether that player was involved in making it happen. That involvement can come from carrying the puck across the line, passing it across the line, or receiving a pass on the other side.

Let’s take Landon Dupont since his Summer Scouting profile is already out. If Everett successfully crosses the blue line 100 times and Dupont is involved in 30 of those entries or exits, his offensive transition involvement would be 30%. I also track what happens when he gets involved. If Dupont attempts to become involved in 100 transition plays and Everett successfully maintains possession on 58 of them, he’s 58% efficient. The other 42% of those attempts end with the opponent gaining possession.

Now we have two pieces of information that tell us different things. Involvement tells me how frequently a player is being asked, or choosing, to influence transition. Efficiency tells me how successful he is when he does. Neither number means much without watching the player.

That’s why I’ve never viewed tracking data as a replacement for scouting. I use it to make my scouting more accountable. If I write that a defenseman is excellent in offensive transition, I want to know what that actually means. How often is he involved? How often does it work? How is he doing it? What happens when pressure arrives? Does he make the same plays against better competition? Does his approach change when space disappears? The numbers help me identify what is happening. Watching the player helps me understand why.

I can make this considerably more complicated, and throughout the year I will, as I explain different concepts, but when I’m explaining what I’m looking for to someone who doesn’t spend an unhealthy amount of time watching junior hockey, I usually boil offensive and defensive impact down to a handful of questions:

Can you enter the offensive zone with possession? Can you get the puck into the “house,” or the dangerous areas of the offensive zone, either by carrying it there or finding a teammate with a pass? If your team loses the puck in the offensive zone, can you help get it back? Can you prevent the other team from crossing blue lines with possession? And when you’re in your own zone, can you get the puck and get it out with control?

Obviously, playing hockey at an NHL level requires considerably more than answering those six questions. Different positions require different skills; different players accomplish those things in different ways, and there are physical and mental tools that determine whether they’ll continue to be able to do so as the game gets faster.

But those questions give me a starting point, and then I want to understand how the player is answering them. A player might be excellent at entering the offensive zone because he’s faster than everyone else on the ice. Another might do it because of his edgework and ability to change direction. Another might not be particularly fast at all, but constantly manipulates defenders, recognizes available space, and attacks it before the defense can react.

All three players might generate a successful controlled entry in my dataset. Still, the scouting evaluation is based on projecting which of those methods will continue to work when everyone on the ice is bigger, faster, and smarter.

That’s ultimately why I track what I track. I’m not trying to turn hockey players into spreadsheets or create a number that tells me who will become an NHL player.

I’m trying to give myself more evidence to answer the hardest question in scouting: not just whether a player is good right now, but why he’s good and whether the things making him good today can translate across his developmental path to the NHL.