Announcement

Collapse
No announcement yet.

Comments Thread For: Teofimo Lopez Is Proving The Power Of Manager David McWater's Boxing Analytics

Collapse
  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    While he was giving very vague descriptions of his analytical computations in the article, I imagined this guy McWater using a folded paper fortune teller to determine the next great "can't miss" prospect.

    Comment


    • #12
      Analytics can be interesting in boxing. In baseball it takes away the whole homerun thing and looks at other valuable assets like a player’s contact rate, defense, etc.

      What people must realize when speaking of analytics is that all data is flawed. There is no foolproof stats even in baseball. So it is going to be the same in boxing. It is going to be even more flawed and will not be perfected any time soon. In the MLB, baseball teams pay big bucks for statisticians. It actually did not start with the A’s as many people are lead to believe. It started with the Brooklyn Dodgers in 1947. They were the first team to hire a statistician to gather information about their team, what pitches are thrown at them by what pitchers, homeruns, hits, strikeouts, etc. The Oakland A’s version was made popular because a book and a movie was made about it. So everyone sees it as some innovative idea by the A’s.

      It has been around for a long time and has its uses in any sport, or really anything. It has a lot of flaws, but it can work. It can take a while for it to work. The Dodgers didn’t instantly win it in 1947 when they implemented stats into their team. They actually won their first championship in 1955, 8 years after first implementing it. And they have been a very successful team ever since.

      The A’s tried it in 2002 and won the World Series in 2002. Difference between the A’s 2002 and Dodgers 1947 is the information available and resources they had was completely different. The Dodgers had pen, paper, and calculators. The A’s had computers and software that was dedicated to stats. Data mining is even better now than it was in 2002. And way better than it was in 1947.

      Originally posted by mobetter View Post
      So it takes a statistician to know that gold medalist winners have a greater chance to succeed in boxing, geesh
      Yes, and Lopez is not an Olympic Gold winner. But he is projected to have a similar career to past Olympic Gold Medalists.

      Comment


      • #13
        Originally posted by gingerbreadman View Post
        As someone who deals with statistical analysis in my job and in my PhD to a lesser extent, I'd have a couple of worries about this approach - firstly, boxing doesn't happen nearly as often as any sport, so your sample size is by necessity either very volatile ('swingy') or else is going to be predicated on an observer's guesstimation of certain metrics, which is what coaches, pundits and fans already do. Secondly, boxers learn outside fights. In Moneyball, Michael Lewis argues that you couldn't identify Barry Zito as a top pitcher until he'd perfected the disguise on his curveball in his senior year of college; the same things happen in boxing gyms, because boxing is a technical sport. I'm not saying this sort of analysis can't give you an edge, I'm just very sceptical of its use relative to sports in which it's used more routinely.
        I've always been big into analytics, and love that it is finally making its way into boxing. However, I am 100% on board with your sample size critique. In baseball, you are getting 600+ PA a season to evaluate a hitter; in boxing youd be lucky to get 25 rounds a year from the top fighters. Moreover, statistics are not as readily available in boxing as they are in other sports, and even the ones we have (compubox) are somewhat dodgy.

        Still, I think there is a place for it (I know there is a strong anti-information faction out there that hates these types of things). I believe the best bet would be to utilize movement trackers, similar to the NBA, in gloves and on the body which can give real time data. Then cross reference that with successful outcomes. Additionally, I think using analytics to evaluate judges would be EXTREMELY beneficial. In what other sport do we so readily accept the outcome as some sort of 'beauty is in the eye of the beholder' mentality as we do in boxing. I believe analytics could create a more uniform approach to judging fights, instead of this constant shifting of the goalposts. This could in turn reduce perceived corruption, or at the very least reduce bias given towards the more famous fighters.

        Comment


        • #14
          Originally posted by DeeMoney View Post

          I've always been big into analytics, and love that it is finally making its way into boxing. However, I am 100% on board with your sample size critique. In baseball, you are getting 600+ PA a season to evaluate a hitter; in boxing youd be lucky to get 25 rounds a year from the top fighters. Moreover, statistics are not as readily available in boxing as they are in other sports, and even the ones we have (compubox) are somewhat dodgy.

          Still, I think there is a place for it (I know there is a strong anti-information faction out there that hates these types of things). I believe the best bet would be to utilize movement trackers, similar to the NBA, in gloves and on the body which can give real time data. Then cross reference that with successful outcomes. Additionally, I think using analytics to evaluate judges would be EXTREMELY beneficial. In what other sport do we so readily accept the outcome as some sort of 'beauty is in the eye of the beholder' mentality as we do in boxing. I believe analytics could create a more uniform approach to judging fights, instead of this constant shifting of the goalposts. This could in turn reduce perceived corruption, or at the very least reduce bias given towards the more famous fighters.
          What you said with movement trackers is something that I’ve seen before in boxing. I’m that YouTube video that attempts to show that Pacquiao really beat Mayweather, the author of that video used movement tracking to show where Mayweather and Pacquiao were going throughout the fight and how much activity there was. It would be interesting to see it in real time.

          I also remember hearing a long time ago about a PSI tracking device placed in hand wraps and gloves to track PSI on a landed punch. It would be interesting to see these things being implemented in a real match.

          Comment


          • #15
            Just looking at these comments and extremely impressed with the intelligence of my fellow boxing die hard fans. You guys are really breaking down baseball and it's analytics. Pretty cool for a boxing forum
            b morph likes this.

            Comment

            Working...
            X
            TOP