[TrueHoop] A Stat Geek prepares for the Draft

This isn't directly related to the Rockets, but I feel its relevant to the GARM because the Rockets GM is also an analytics guy. I think the way we approach the draft and assess prospects might be very similar to what this guy from the Nets is talking about.

http://myespn.go.com/blogs/truehoop/0-32-350/An-NBA-Stat-Geek-Prepares-for-the-Draft.html

[rquoter]
Ken Catanella is the Nets' Coordinator of Statistical Analysis, and one of the few people in the world with a full-time NBA job in advanced basketball analytics.

Catanella has two things that a lot of similar experts do not: the knowledge of basketball that comes with having played as a professional, and hands-on experience doing rigorous multivariate analysis on Wall Street. Following graduation from Amherst College in 1997, Catanella analyzed arena finance for professional teams, and valued publicly traded companies.

At Amherst he was both a player and later an assistant coach, and then played professionally for the German Bundesliga's Cologne 99ers, where he also later served as the German team's Assistant GM.

While earning his MBA at Duke in 2004 and 2005, Catanella assisted the men's basketball teams, developing analytical tools and systems for Coach Krzyzewski's staff. At the same time, Catanella interned for the 76ers' front office.

Two days before the draft, it is obviously a very busy time for Catanella at the moment. The Nets have picks 10, 21, and 40 -- which means they have to know about practically every part of the draft.

And don't forget that there could be several trades as the draft unfolds. Once the draft starts, Catanella says there is no time for sophisticated analysis. So the team imagines virtually every conceivable scenario, so that as opportunities arise, they are ready to make quick decisions they can live with.

On top of all that, Catanella recently had his first child. He agreed to talk to TrueHoop about his job.

Busy time for you, huh?
I'm enjoying it. This is the most diverse set of draft picks. No team I have worked for has had anything like it, and it makes this a really interesting draft.

It seems to me like nearly every player in this draft has some question mark or another.
I don't know about that. There are some pretty good players this year. You can come up with questions about just about any player in any league. A lot of that is just a glass half-full vs. glass half-empty thing.

So, I know a lot of teams wrestle with how to integrate new-breed analytical work into their decision making process. Can you give me glimpse of how that works for the Nets?
Taking a step back -- my background is in valuing companies for investment banks or mutual fund companies. It's making a projection based on past performance, trying to answer the question "what here really indicates likely future success?"

Figuring that out really starts in conversations with the coaching staff, and trying to get a sense of what kinds of stats could be important. I did a lot of that when I was on Coach K's staff, and I'm lucky enough to work with Lawrence Frank.

Then, based on those insights, we do a bit of regression, which helps to project how different players' careers might play out in the NBA.

What goes into the mix is not any one magic number, though.

There are typical efficiency stats. There are anthropometric ratings like wing span, body fat, vertical leap, standing reach and the like. There are strength of schedule ratings. There is looking at who they were on the court against, and who they were on the court with. There are +/- numbers. The numbers really run the whole gamut. You try to find anything that you can measure that might be helpful, and put it all in the mix.

Do I sense that you are not even looking at the same sets of numbers for every prospect?
It depends on the position. And it can get very detailed.

We chart, essentially, every game that every draft prospect has played on video, and we track just about every category you can imagine.

Closing out the shooter, winning loose balls ... all that stuff?
All that and much, much more.

I work with one of the best teams in the League, as far bouncing ideas off people. Lawrence Frank is from that Jeff Van Gundy, Pat Riley school that really values metrics, and thinks they are worth their time.

Every team is dealing with the same player pool, and thanks to the luxury tax and salary cap we have very similar resources. So the little advantage we get from having a coaching staff that lends insight to how we gather information to assess players ... those are the types of advantages that can make a little difference.

Also, scouts are very valuable tools, too. I never want to underestimate them. We can work together by getting data into their assessments early, seeing a players' characteristics. Trying to get a full picture, using playing stats, evaluation stats, and the statistical part of the psychological evaluation ...

It's an opportunity, on this team, to have a healthy dialogue whether we're talking about the draft, free agency, trades. Kiki [Vandeweghe] and Rod [Thorn] have seen every kind of player before. Working with the coaches, the scouts, and the numbers, we can try to get a sense of how any player might fit with our team.

Lawrence Frank, for instance, will actually take us on the court, and walk us through a play. That way we can really understand what the player we get will be asked to do, and we can go and look for somebody who can do that.

Sometimes some statistical assessments will recommend players that are very different than traditional scouts. For instance, John Hollinger's new projections don't rank Derrick Rose so highly as a point guard prospect. How do you handle that?
I read everything. I read John Hollinger's stuff all the time. I got a chance to talk to him in Boston at that conference, and learned that we use a lot of the things he uses.

Those kinds of things can be indicators, negative or positive.

We never look at any of these things and make a final decision. The traditional scouts have another take, which is often more valuable.

I try to relate what I see in the numbers to what I have learned from my basketball background. So when the numbers show something, it's normal to ask: Is this really important? Will this help us win a playoff game? We're always trying to figure out, how will this thing we have discovered affect our team when we run this play.

That specific kind of thinking is invaluable.

When you're just going through numbers, you're left with a lot to learn. So when the numbers turn something up, before taking it to the scouts we'll go to the film, and see what's happening. After you watch whatever it is playing out over ten games, you can get an idea -- do these numbers make sense? Are they relevant?

Sometimes you find there is not a whole lot there.

And sometimes you'll find that there's a player who seems to be really effective, but is not on anybody's draft board, or is very low on people's draft board. There were a couple of guys like that last draft, and now I think other teams are starting to notice, too.

What role do these stats play in final decisions? Of those three picks you have, how many would you guess would be different players if you didn't work for the team?
That's a real hypothetical. I'm not sure. But I would think that some themes we have discussed, focused on, and debated, have played a role. What we see in data ends up being what people can see on film. It is generally noticed already. But perhaps overlooked, or nor weighted much.

Like, is this guy good at choosing shots? Is he a good rebounder? Where does he take his shots from? How does he perform against NBA-type players? The new era stats help us have a rich dialogue on those kinds of topics. Basically, it helps Kiki and Rod with information that they might find intuitively, or not.

In baseball, the "Moneyball" trend clearly led to valuing on-base percentage higher than before. Have basketball's new statistics touched off any visible trends?
One thing is that teams do really not talk to each other about this kind of stuff. So it is hard for me to know what kinds of decisions they are making.

But you can see how they spend their money, and who they draft.
Yes, and there are clearly a lot more measurements around. If we're ranking the best player, for instance for the draft, we have 20 or more metrics now. Or if we're looking to sign a role player, we might use five different measurements unique to that position.

The big thing, is what John Hollinger started talking about years ago. Efficiency levels. He really opened a lot of people's eyes, including my own. Team defensive efficiency, team offensive efficiency, points per possession, individual efficiency ... people were doing some of those kinds of things before, but now there is a lot more, and you see that playing out across the league.

For instance, something we have seen is players like Paul Millsap and Carl Landry. I'm guessing that traditional scouts thought they might be a little short for their positions, but they both have proven to be efficient in the NBA. Is that a kind of player that we can attribute to some new-breed stats?
There are some smaller frontcourt players like that, I think, who are an inch or two shorter than a traditional scout might like. Some of them have been overlooked, through the years, to a certain degree. Discounted because of certain physical attributes. There are ideas about his is how an NBA player should look, this is how he should run, and this is how he should jump.

But people are getting past that to a certain degree. That doesn't happen by accident. I don't know for sure about those players you mentioned, but I would guess that those guys would not have been drafted as high as they were drafted without someone assigning a value to what they do.

Also, it seems that charting all your own video, which you say the Nets do, is just about the only way to assign any kind of numbers to how well a player plays defense.
Exactly. The boxscore is lacking on that front. With the right people doing the charting, and the right methodology for collecting all the data, that can be powerful.

It's objective and subjective. The people charting the information have to know how to view certain plays. And you have to factor in how a certain play was coached. Was the defender told to hug Dirk Nowitzki on the pick and roll? If Nowitzki scores, whose fault was that?

Watching the tape, I guess you have to someone differentiate between a three made over a closing defender doing a great job, and a three made over a closing defender trying to look like he made an effort.
Certain defensive stats are more difficult to judge, no question. The key is discern the important basketball pieces of the puzzle. Rod has seen multiple generations of style of play, and players, come and go. He is comfortable dealing with all the variables that go into this.

One thing that has always impressed me: We have all fallen in love with this or that top college player who has not flourished in the NBA. Some people seem to have a good ability to tell which college characteristics will translate well to the NBA, and which will not.
We have a system for measuring those kinds of things. There are statistical measures. There are more than a handful of numerical components, where you can compare this or that to past generations. And the longer we do this, the more years of information in our system, the smarter that system gets.

Some of our scouts get to see 120 college games a year, too, in person. They are working with a body of knowledge that most fans could never have.

For them, there is not a lot of emphasis on numbers. If there is a player with a certain level of quickness, for instance, or somebody with an NBA body, they've probably seen five like it before, and have an idea what to expect. The data might have information on another 200 players like that one, and we can meld those two kinds of information.

You must have some players who are kind of "your guys" that you are hoping will become Nets.
Oh, by the end of the draft process, we all do. We all have players we like. But we come to consensus. No one holds grudges, and we all have the same goal. The scouts are tremendous and see things I hadn't thought of. And later, I might say "you know what? He was right." At the end of the process, we end up with our draft board all set-up, and we all know where we rank players as a franchise.

So, when you're actually in the war room on draft night, what is your role?
Good teams have gone through almost everything well ahead of that night. All the evaluations, the rankings, the conversations.

Trade offers for current NBA players can catch you by surprise. But by planning for a lot of those, I might have the relevant information in my head, or be able to pull it up quickly from our databases -- which is why you build those databases, by the way, for nights like that.

All those discussions, however, if done in the correct way, the rankings and the like, there is not time for that level of dialogue. The discussion and analysis has to be done ahead of time.

So when is that done?
It's a long process, coming up with where to put each player, and then imagining all the scenarios and preparing for them. It takes place over many days. But it's typically complete the day before the draft.

Good luck.
[/rquoter]

 

This interview is specifically talking about how a stats-oriented team is doing evaluation for the draft. I thought it was very detailed and insightful. We're a stats-oriented team, and the draft is a few days away. So I think it's relevant.

 

Actually, if you think there is a real, concrete similarity between the business practices of Aetna and the Rockets, and a discussion of it would help us understand our team better, then I think it has a place in the Rockets GARM. At least as much as some gossip thread about one our player's personal lives.

But if this thread bothers you that much, I'm not going to cry if a mod locks or moves it.