Arnovitz: Blazers GM Neil Olshey Talks Analytics

Kevin Arnovitz of ESPN.com interviewed Portland Trail Blazers GM Neil Olshey at the 2013 MIT Sloan Sports Analytics Conference. Here's a link to the videos, which are embedded below.

Here's a partial transcript, in which Olshey admits to entering the conference with a "very remedial" understanding of analytics that he hoped to improve upon.

What are you doing here?

Getting better. It's a great place to learn. This is not something I can replicate with my own background in the game of basketball. I have Ben Falk who works with me, Ryan Parker who works with us. It's important that I support them to know that we do embrace analytics and we're invested in it but I also have to increase my education curve. I came to this with a very remedial knowledge of analytics. Any chance I get to accelerate my learning curve, I have to take advantage of it.

First exposure to advanced stats?

We did a little bit of it in Los Angeles with the Clippers later during my tenure. That was more unit analysis to give the coaches insight into the best roster composition we could have on the floor. Who was being productive, who wasn't. We didn't really use it a lot in the draft or free agency. When I got hired in Portland, I realized Ben Falk was the smartest guy in the room and it was probably a good idea if I listened to him once in awhile.

How does process work between Neil Olshey and Ben Falk?

It works both ways. There are things Ben brings to us that I wouldn't even have the educational background to ask the question to acquire. There's also things, through the lens that I look at players or our team, I go to him and say is this something you can help me with, create a model with draft analysis or free agency or roster composition. Who performs better with whom, how we are with guys on the floor, whether we are better or worse. That leads me to what we should be looking for on the free agency or trade markets.

>What happens if numbers conflict with a traditional scout's take?

I would say 95 percent of the time if you have a scout or front office person who really knows what they are doing, and same with the analytics guy, I would say 95 percent of the time they are on the same guy. It's very rare that there's a crazy outlier that one or the other believes in or is invested in and the other person is completely off the reservation.

Why draft Damian Lillard if numbers preferred other players?

Everybody wants to quantify 100 percent of the decision-making matrix and you can't. We've got live scouting, we've got workouts, we've got tape, we've got psych, we've got intel, we've got background, we've got the interview, we've got analytics. We've got to look at that pie chart and assign a value to each of those things. At some point, once you've quantified as much as you can quantify, there's an intrinsic feel for players and the game of basketball that can only come from being around it for a certain period of time. Whether you played, coached, worked with players, scouted players, at some point you know there's just an 'it' factor. It's the girl sitting at the counter in the soda shop in the 1950s and the next thing you know she's a movie star.

At some point there's got to be an element of gut instinct on players. You try to shift the odds in your favor and decrease your margin of error and you do that by hiring competent professionals with a proven track record. You hire analytics people to give you a different lens to view players through.

Rising importance of analytics

Clearly when you have an owner that created Microsoft, he's going to hold you to a higher standard than someone who developed a real estate empire. They want to know that you're exhausting every possible resource to improve their team and build their organization. I don't think you can leave any stone unturned. The competitive balance in the NBA, the new Collective Bargaining Agreement, any way you can get an edge. [Rockets GM] Daryl [Morey] brought this up the other night. This went from, 'any way you can get an edge on your competitors' to, 'if you're not doing this you're falling behind.' At some point it's not about progress it's about staying par with the course.

Assessing a player's make-up

We're really just trying to shift the odds in our favor, how can we decrease our margin of error? From the draft standpoint, there are a lot of unknowns. It's hard to quantify. They play at a different level, they are at schools at different times, are they late bloomers, are they early bloomers. What kind of system do they run in college? What kind of system do you run in the NBA? Personality-wise, are there any red flags? I think at some point you can get only so far, you can get into the red zone but having a feel for players is what separates front offices that consistently knock it out of the park with their draft picks and have an incredible draft record and teams that are all over the map.

Mark Cuban argues that evaluating a team requires the decision-makers to be in close proximity to the team during the season

I agree. I'm at home whenever the team is at home. That's when players are more relaxed and accessible, you have breakfast together in the dining room. You're out on the court, maybe you shag a ball for them. It's things you can't do on the road. [On the road,] they're at shootaround, they're on the bus, they're back, they're taking a nap, they're back, you don't want to interfere with the process.

I also think having a great working relationship with the coach, as I do with Terry [Stotts], I trust he and his assistants to keep their finger on the pulse of the team in a more organic nature. If I start hanging out in the locker room, I interfere with their process and it puts players on edge a little bit because it's not commonplace. I can do that at home because I'm always around, and they are used to that and their guard is down, and I can get to know them and see what's going on with our team. You've got to let it breathe a little bit too, let your team go, get some peace on the road, focus on their task. If there's an issue you hop on a plane and catch up with the team and if there's an issue with a player you catch up with the player. Once you've established that baseline relationship and you know the coach has his finger on the pulse of the team, there's also something about hovering. With your kids, you can't constantly be looking over their shoulder, at some point you have to let them grow a little bit, make mistakes. It's better to have players come to you when they trust you rather than constantly trying to inundate them with your presence.

You said Ben Falk "Can't play dead in a cowboy movie"

How many guys in old Hollywood play dead? He gets shot, he lays on the ground, he has no lines. That's the end of it. Ben as a basketball player isn't talented enough to play dead as an extra in a cowboy movie but in terms of what he does, I think he's the best in the league and I think he's brilliant. What was your first impression of Ben? He looked exactly the way I wanted an analytics guy to look. I knew I wasn't the smartest guy in the room as soon as he sat down. But I also knew as a basketball player he couldn't play dead in a cowboy movie.

-- Ben Golliver | benjamin.golliver@gmail.com | Twitter