NHL General Manager Paul Holmgren said on Thursday that the team wants to build a more “automated” way to analyze player performance and “improve the way we can use analytics” in the NHL.
“We want to have some automation in the hockey world,” Holmgreens stated at the NHL’s annual GM meetings in Anaheim.
“We have an automation team working on the analytics team at our front office, which is a great team, and we’re also going to have our analytics guys and our hockey ops guys.
But we want to get them all together, and the way to do that is to automate.
That’s what we’re looking at now.””
That’s going to be part of our analytics team, because analytics is what we do,” Hologren continued.
“It’s not an artificial intelligence thing.
It’s a process where you go through a bunch of data and look at that, and you take that and make sure that you have the right kind of data.
So we’re going to automate that process.””
You’re going, ‘OK, this is what I’m seeing,’ and that’s going into the computer and it’s going back and forth.
So you’re going through that process all the time.
And then the computer comes back and you’re like, ‘That’s not what I saw.’
So we want automation in there, and it needs to be a human-like process.”
Holmgren explained that the analytics staff would be responsible for using analytics and their “human” counterparts, who are going to “use their judgement to make sure they’re right.”
The NHL’s analytics team is a new one that has been created to “help” the organization analyze players on a regular basis and, if the team believes they need more “intelligent” data, they can hire a “data scientist” or “human expert” who will help them do so.
“You’ve got to have a team of experts that’s capable of going through a lot of data, analyzing that data, and making sure that they’re accurate and they’re making sure they are not creating false positives or false negatives,” Holgren said.
“So that’s what this team is going to do, and they are going, they’re going ‘OK,’ and then they’re just going to come in and do their work.”
“So that is a lot different than just having a data scientist that’s not a professional hockey player, or someone that’s got a little bit of experience in analytics,” Holgerunner added.
“They’re going into our system, they have a computer, they run through their data, then they do a little analysis, and that is then going into their computers, and then those are then used to create an automated process that’s able to analyze that data.”
So we’re not going to take our data from a team that’s a data nerd, and our data is going into a computer.
We’re going have to have this team that understands how to create and apply this data.
That is going back to us and then we’ll bring that data into the system and it will be run through the machine.
“With all of the new teams entering the league in the coming years, it appears the NHL is still a work in progress with regards to how to properly utilize the technology and improve the way in which teams use it.