Announcement

Collapse
No announcement yet.

Jay Satur's 2 part Q/A with Alex Rucker. Remember him? Raps' Analytics guy.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Jay Satur's 2 part Q/A with Alex Rucker. Remember him? Raps' Analytics guy.

    http://blog.raptors.com/q-a-with-alex-rucker/

    Part 1:

    The opportunity to sit and chat with Raptors Director of Analytics Alex Rucker is one I’ve been looking forward to for a while. Advanced stats isn’t a topic we’ve heard a lot about with relation to the Raptors over the years and Rucker has been a big part of its recent growth within the organization.

    I met him inside the Raptors front office headquarters at ACC recently to talk tweets, trends, summer transactions and a how a whole bunch of cameras have kept him busier than ever.

    Jay Satur: Your twitter account has offered up some fascinating insight on the organization without giving too much away. One tweet that really caught my eye was in mid-July when you said “I’m now in my 4th year with the team, and the past month has been the best month…and it’s not close.” Why so happy?

    Alex Rucker: The last three years for me has been sort of an interesting journey. I was heavily focused on the coaching stuff and then there’s kind of been a three-year progression of both refining what we do analytically and also working more and more closely with the transactional side with Bryan [Colangelo] and now Ed [Stefanski]. So for me as an analyst, the time period of signing Jonas Valanciunas, bringing him over, trading for Kyle Lowry, bringing on board Landry Fields, I mean these are guys that analytically are all excellent additions to our squad. Jonas has obviously been in the works for about a year, but there was a significant shift to these guys that I feel very strongly will improve our team in a variety of different areas so it was great to see that all come together.

    JS: This will be your first full year for you and the analystics team, but you’ve been working with the Raptors in some capacity since 2009, how has your role evolved over time?

    AR: I’d spent the previous eight years kind of refining my craft as an analyst, but there is a massive difference between being a public analyst with access to public data and then working with an organization where there’s an exponential increase in data available to you. Whether it’s coaches play calls on offence or defence, or specific things the coaching staff or video guys are tracking, kind of folding that into your analysis, there’s all these new data streams. So the last two or three years I feel like as a professional, as a basketball person, as an analyst, I’ve grown massively. Our analytic team, led by myself and Keith Boyarsky, we’ve really benefitted from that additional flow of information and being able to fold it into what we do.

    I think it was also a new thing three years ago for this organization. It was one of those things where Jay was aware of this world of analytics, Bryan was aware of it, but it wasn’t something that had been really embraced or taken advantage of because it’s a journey, right? With baseball you can see that in the 80′s, there was all this public analysis being done. If you use the ‘Moneyball’ book/movie, it took 15-20 years for it to get engrained in front offices, so it’s not an unusual thing this new way of looking at the game, if you will, it kind of takes a while to trickle in, for people to get comfortable with it, for people to understand it, because it’s an unfamiliar way of looking at the game to someone who has grown up inside that world.

    JS: How have you seen the use of analytics grow over the course of your time with the Raptors?


    Coach support is one of three big picture areas that Rucker’s team assists with. (Ron Turenne/NBAE)
    AR: There are three ‘big picture’ areas. The first is pro player evaluations; looking at NBA players, free agents, transactions, contracts, the guys on our team. Then there’s prospects, the draft and that’s a whole separate world and frankly requires a very different analytic methodology. The third is what I call ‘coach support’. The other two are focused on transactions, player value and role at the strategic level. At the coaching level, it’s much more about executing every individual possession a certain way. That’s three very dissimilar areas and growth and progression has occurred differently in each of the three.

    At the player transaction level, I think it’s really more been a function of better understanding what Bryan’s vision was and better identifying players that fit that vision analytically. There’s always kind of a lag time with the science piece of building some new design, model or methodology and then testing it, making sure it works and then ‘OK, you want these things? Here they are.’ So I think that it’s largely been a function of that, the development of it, so the things they’ve wanted, we now have much more of them. To identify certain things we’re now better able to do that.

    Then at the college level it’s kind of the same thing. Before coming here, my focus analytically had 100 per cent been on pro players. I had not devoted a lot of energy into college or European prospects because honestly, the data’s ugly. NBA data is much more rigorous and reliable. College data, there’s a lot of issues with it. European data even more issues and there’s less of it, so we hadn’t delved into that world. That’s been a significant challenge. I think we’ve gotten to a much better place, but scouting and drafting will always be a lot more of a grey area, I don’t know that it will ever get as scientific as the pro player stuff.

    JS: Has this been accelerated by working with a coaching staff that’s been a little more rooted in analytics in the past?

    AR: Dwane’s definitely very open. His staff is very inquisitive. They ask a lot of questions. They challenge our assumptions and question the things we do and that’s great. The first thing I tell anyone, whether it’s working for Bryan or the coaches, is that I’m going to offer suggestions and I want you to challenge them.

    In the academic world, it’s like submitting to peer review. I want these guys to attack our ideas. I want them to criticize them, to dig into them. If they don’t stand up to scrutiny, they don’t belong in the conversation. That’s what we aspire to and Dwane and his staff have done a great job with that. We have very passionate conversations about a whole host of topics and whether it’s Dwane or Tom [Sterner] or Micah [Nori], that’s the stuff that helps us get better and better supports them.

    JS: Advanced stats is something that’s become commonplace now in public basketball discussion, how do you view that public analysis given your advantage in terms of that volume of data available to you?

    AR: The disparity in available data is a huge deal and I’m always sympathetic to that. I’ve mentioned this before, but there is basically no good data about defence in the public sphere. If you as a public analyst want to understand defence, good luck. The data just is not there and the attempts that do get made are mostly unhelpful in terms of understanding defence.

    I’ll read articles about defensive rating or steals or blocks. Those all indicate something, I don’t know that it’s correct to say that you can take those things together and say that it gives you an idea about a guy’s defence. For some guys yes, for some guys no, but then the problem is that it’s not a correlation that’s a meaningful thing.

    I think that in the public sphere, you can make a lot of progress in terms of offence. I think there’s enough data to do some really interesting, really good analysis of certain things on offence. Lots of areas for growth there. But on the defensive side, with the current data, you’re not getting there.

    JS: How beneficial has the SportVu system been to what you do?

    AR: The SportVu data system, which to recap is six cameras tracking all 10 players on the court, the three referees and then the ball in three dimensions, has been the single biggest step forward in terms of analytics in my lifetime basically. It’s orders of magnitude better than what we had before in terms of data. I guess the easy example is we went from about 800 bits of data in an NBA play-by-play to about 800,000.

    Now the opportunities for growth and for enhanced understanding using that data are almost limitless. If you can ask a question about the game, there’s probably a way to capture that using SportVu camera data. The other things all help, but that is the single biggest advantage that we enjoy over the public sphere. Having all the intent pieces from the coaches and the management is awesome, but having this incredibly rich array of data that SportVu gives us, there’s nothing to compare it to.

    JS: This is fairly new technology, but how many teams currently use it?

    AR: Two years ago in their beta year, there were six teams, this past season, there were 10 teams and my understanding is they’re shooting for half the league this coming year. From our perspective, every additional team is more data, so that’s great. If they stay with 10, that’s still frankly a lot of data for us to continue to exploit and explore. But it’s been a phenomenal step forward.

    Part two coming soon, with Rucker’s thoughts on Lowry, Fields and Valanciunas.
    TORONTOOOOOO RAPTORSSSSSS

  • #2
    Couple of things jump out at me. "...but there is a massive difference between being a public analyst with access to public data and then working with an organization where there’s an exponential increase in data available to you."

    And "...there is basically no good data about defence in the public sphere. If you as a public analyst want to understand defence, good luck. The data just is not there...I’ll read articles about defensive rating or steals or blocks. Those all indicate something, I don’t know that it’s correct to say that you can take those things together and say that it gives you an idea about a guy’s defence. For some guys yes, for some guys no, but then the problem is that it’s not a correlation that’s a meaningful thing."

    So when Bryco and company target a particular player, it may be that the stats indicate it is a very favourable acquisition, within the context of the type of offence/defence the Raptors want to run, even though a lot of fans go "WTF?"

    And the comments about the lack of correlation between publicly available defensive stats (steals/blocks) and a players actual defensive effectiveness might be pertinent to any discussion on James Johnson. I don't know that it is, but if he really was a defensive stopper, maybe he wouldn't be gone, or maybe there were other huge downsides to keeping him.

    Looking forward to part two.

    Comment


    • #3
      Puffer wrote: View Post
      Couple of things jump out at me. "...but there is a massive difference between being a public analyst with access to public data and then working with an organization where there’s an exponential increase in data available to you."

      And "...there is basically no good data about defence in the public sphere. If you as a public analyst want to understand defence, good luck. The data just is not there...I’ll read articles about defensive rating or steals or blocks. Those all indicate something, I don’t know that it’s correct to say that you can take those things together and say that it gives you an idea about a guy’s defence. For some guys yes, for some guys no, but then the problem is that it’s not a correlation that’s a meaningful thing."

      So when Bryco and company target a particular player, it may be that the stats indicate it is a very favourable acquisition, within the context of the type of offence/defence the Raptors want to run, even though a lot of fans go "WTF?"

      And the comments about the lack of correlation between publicly available defensive stats (steals/blocks) and a players actual defensive effectiveness might be pertinent to any discussion on James Johnson. I don't know that it is, but if he really was a defensive stopper, maybe he wouldn't be gone, or maybe there were other huge downsides to keeping him.

      Looking forward to part two.
      Pretty interesting stuff. If you consider that most of us have used the stats that are avaiable to us to make or dispute an argument on the mainpage or forum this kind of says that those stats don't necessarily represent the truth.

      Comment


      • #4
        Puffer wrote: View Post
        Couple of things jump out at me. "...but there is a massive difference between being a public analyst with access to public data and then working with an organization where there’s an exponential increase in data available to you."

        And "...there is basically no good data about defence in the public sphere. If you as a public analyst want to understand defence, good luck. The data just is not there...I’ll read articles about defensive rating or steals or blocks. Those all indicate something, I don’t know that it’s correct to say that you can take those things together and say that it gives you an idea about a guy’s defence. For some guys yes, for some guys no, but then the problem is that it’s not a correlation that’s a meaningful thing."

        So when Bryco and company target a particular player, it may be that the stats indicate it is a very favourable acquisition, within the context of the type of offence/defence the Raptors want to run, even though a lot of fans go "WTF?"

        And the comments about the lack of correlation between publicly available defensive stats (steals/blocks) and a players actual defensive effectiveness might be pertinent to any discussion on James Johnson. I don't know that it is, but if he really was a defensive stopper, maybe he wouldn't be gone, or maybe there were other huge downsides to keeping him.

        Looking forward to part two.
        +1

        I was thinking the same about Johnson.

        Comment


        • #5
          I found it interesting how Rucker's point about the lack of reliable Euro data seems to directly counter BC's Euro-heavy draft & free agent record. I wonder what access Toronto and the rest of the league have to the more high profile international competitions, e.g. Olympics, FIBA, etc?

          Comment


          • #6
            dunphyj wrote: View Post
            I found it interesting how Rucker's point about the lack of reliable Euro data seems to directly counter BC's Euro-heavy draft & free agent record...
            BC's Euro-heavy draft record may reflect his connection with European coaching staff and GMs. His Dad wasn't afraid to take on staff from Europe and Bryan has maintained those connections and made more of his own. In that sense, over the years he may have had an in that a lot of other NBA executives didn't,

            Comment


            • #7
              CT2010 wrote: View Post
              Pretty interesting stuff. If you consider that most of us have used the stats that are avaiable to us to make or dispute an argument on the mainpage or forum this kind of says that those stats don't necessarily represent the truth.

              Rucker makes the argument that "defensive rating or steals or blocks. Those all indicate something, I don’t know that it’s correct to say that you can take those things together and say that it gives you an idea about a guy’s defence. For some guys yes, for some guys no, but then the problem is that it’s not a correlation that’s a meaningful thing", yet any 'stat head' would say the exact same thing. If one actually delves into the defensive-centric statistics available (or any really) they aren't even close to being based on just those numbers anyways (although they may be part of a larger package). Taking a step further, if we look at +/- or on/off court stats they don't even consider individual values for numbers like these.

              Steals/Blocks may very well not = good defense, BUT alot of public stats tell us this very thing anyways. So I personally don't find it a very compelling argument about the worth 'public stats'. Nor is using a term like 'public stats' very telling either, as there are many different mainstream methods that provide very different results/analysis. (PER, WP, Adjusted +/-, Win Share just to name perhaps the most common).

              I've mentioned this before but I'm really curious about knowing the method(s) Rucker uses. After all, if he's been working with this organization since 2008 and the team has a win record of about 38% over that time whats going on? Even more specifically, until this year, Toronto hit near historic records in defensive weakness. Is this a situation where:

              - management/coaching staff hasn't, to date, been listening to his analysis?
              - some or part of his analysis isn't working?
              - the organization is using bits of Rucker's work but not enough of it?
              - random chance or specific plans (eg. tanking) played a significant role in the results?

              Can his analysis be compared to, or similar to, one of the more common 'public statistics' available or is it completely outside the realm of what they do?

              Comment


              • #8
                Part 2 of the interview:

                http://blog.raptors.com/q-a-with-alex-rucker-pt-2/

                Comment


                • #9
                  Nilanka wrote: View Post
                  Another great article. wish it had been longer. Looks like Rucker and his team were instrumental in both the Fields and Lowry acquisitions. According to him the Raps had their eye on Fields well before Steve Nash was getting the come hither look from NY.

                  Comment


                  • #10
                    Puffer wrote: View Post
                    Another great article. wish it had been longer. Looks like Rucker and his team were instrumental in both the Fields and Lowry acquisitions. According to him the Raps had their eye on Fields well before Steve Nash was getting the come hither look from NY.
                    I would love to be lurking in the background of the "war room" to hear exactly how these internal discussions go down. Very interesting stuff.

                    Comment

                    Working...
                    X