Announcement

Collapse
No announcement yet.

Where can I find the best player ratings?

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

  • Where can I find the best player ratings?

    Hi there,

    Despite never having picked up a bat in my life (I was brought up in England), I have taken an interest in Sabermetrics.

    Basically I'm a professional gambler, and recently I read the book Fantasyland on vacation. It got me thinking, perhaps I could use Sabermetrics to evaluate the relative strengths of two opposing teams and calculate the % chance of either result?

    Clearly the likes of Batting Average, ERA are stats that can be misleading. A pitcher could give up 6 singles and 2 homers, but could concede either 2 runs or 8 runs based on how these are distributed in the game (which seems almost entirely down to 'luck').

    Anyway this is presumably all very basic to you guys - what I'm looking for is a website that applies some sound Sabermetrics to form a rating for each MLB player, and updates it daily. Hopefully I can interpret these ratings and make a couple of bucks

    Any advice would be appreciated.

  • #2
    My advice is don't do it!

    You want to apply saber principles for selecting players in a high-stakes fantasy league, go for it. Ideally you can identify which breakouts are likely repeatable and which are not, etc. Same for down years. Over a full season with the same player, you have a reasonable chance that your saber seed will bear fruit.

    But to bet on a single game, or several single games is folly from a sabermetric standpoint. One of the saber-faced elephants in every statistical room is sample size. An individual game's results are meaningless in relation to the meta-truths sabermetrics uncovers over large sample sizes.

    Also, many of the advantages of saber wit are found at the margins. Think of it this way, even if I told you that I had a coin that had a 51% chance of heads, would that be a enough of a statistical advantage to justify one to bet a lot of money on heads over a relatively small (say even, 100 trials)? No - the threat of the randomness outweighs the advantage of the tip. I wouldn't be any more likely to place a bet on ten coin flips with a 49/51 chance than a 50/50 (maybe 1% more likely, ).

    Sabermetrics is about maximizing returns over the long haul, and making proper strategic decisions in context. It's not about clairvoyance.

    Perhaps, the simplest response to this question is, were this a viable strategy, the statistical mean income of the regulars to this forum would be a lot higher...
    THE REVOLUTION WILL NOT COME WITH A SCORECARD

    In the avy: AZ - Doe or Die

    Comment


    • #3
      Stat One

      Check out my book, Stat One, for insight to using P/E Averages to gauge an offensive player's value.
      Visit www.statonebaseball.com to learn why traditional statistics are ultimately flawed...and why P/E Averages are not!

      Comment


      • #4
        Ditto digg.

        Since you are new, I'd suggest you go here:
        http://tangotiger.net/wiki/

        And also try wikis noted here:
        http://tangotiger.net/wiki/index.php...Websites#Wikis
        Author of THE BOOK -- Playing The Percentages In Baseball

        Comment


        • #5
          Thanks for your reply.

          Originally posted by digglahhh View Post
          My advice is don't do it!

          You want to apply saber principles for selecting players in a high-stakes fantasy league, go for it. Ideally you can identify which breakouts are likely repeatable and which are not, etc. Same for down years. Over a full season with the same player, you have a reasonable chance that your saber seed will bear fruit.

          But to bet on a single game, or several single games is folly from a sabermetric standpoint. One of the saber-faced elephants in every statistical room is sample size. An individual game's results are meaningless in relation to the meta-truths sabermetrics uncovers over large sample sizes.
          Surely sample size isn't a problem in Baseball? There are about a hundred games a week. Any small edge on your bets will surely bear profit over the course of the season?

          Originally posted by digglahhh
          Sabermetrics is about maximizing returns over the long haul, and making proper strategic decisions in context. It's not about clairvoyance.
          You could easily replace "Sabermetrics" with "Betting" in that statement. Someone might ask me who will win tonight's game between San Diego and Houston. I would reply that I don't know. However I might know that there is a 60% chance of San Diego winning and 40% chance of Houston winning. If I could back San Diego at Evens then I would have an edge. Enough of these bets over the course of a season, with a sensible staking plan, would return a good profit.

          Originally posted by digglahhh
          Also, many of the advantages of saber wit are found at the margins. Think of it this way, even if I told you that I had a coin that had a 51% chance of heads, would that be a enough of a statistical advantage to justify one to bet a lot of money on heads over a relatively small (say even, 100 trials)? No - the threat of the randomness outweighs the advantage of the tip. I wouldn't be any more likely to place a bet on ten coin flips with a 49/51 chance than a 50/50 (maybe 1% more likely, ).
          I would absolutely bet on a coin that had a 51% chance of Heads. It's all about stake management. I wouldn't put my life savings on one spin, of course. I would have a 49% chance of going bankrupt. But I would place a small portion of my bank on it. Given an unlimited supply of these 51% coin tosses, there is no limit to how much money you would make.


          Originally posted by digglahhh
          Perhaps, the simplest response to this question is, were this a viable strategy, the statistical mean income of the regulars to this forum would be a lot higher...
          It's obviously going to be difficult, so I agree with your scepticism in a way. But there are people who consistently make profit from a range of sports, including Baseball. It CAN be done!

          Comment


          • #6
            Originally posted by digglahhh View Post
            Think of it this way, even if I told you that I had a coin that had a 51% chance of heads, would that be a enough of a statistical advantage to justify one to bet a lot of money on heads over a relatively small (say even, 100 trials)? No - the threat of the randomness outweighs the advantage of the tip.
            Actually, it very much depends. The cost certainly might outweigh the benefit. That depends on the pot odds, implied odds, and reverse implied odds. Also, position, table image, blind structure/size in relation to average stack size.....

            Much more to it than this, but A-K vs. 10-10 is often where the "smart money" goes, despite what the odds calculators will tell you.

            Oh, snap.....

            Comment


            • #7
              We gotta play some poker, Chris!

              Originally posted by csh19792001 View Post
              Much more to it than this, but A-K vs. 10-10 is often where the "smart money" goes, despite what the odds calculators will tell you.

              Oh, snap.....
              Say we're both all in, before any cards are out, and then we turn over those cards. Two over-cards vs. a pocket pair.

              Q: Who groans? (Assuming everybody else folded)
              A: Both of us

              And that's kinda my point.... The statistical advantage is not enough to make you feel comfortable in the face of random distribution - if we're three hours into a game, we'll both feel like tools for wasting the previous three hours all in pursuit of putting our fortunes in the hands of an ostensible coin flip....

              A very small probabilistic advantage isn't statistically significant in the context of an individual betting on something, the outcome of which is largely dependent on luck and random variation.

              An individual, unless the OP is Sultan of Brunei wealthy, does not have adequate resources to assume that betting on a tiny margin will net him a return. Think about it this way - in a casino, the odds in greatest favor of the house odds are in roulette; a casino is virtually guaranteed to profit on the game over time, thanks to the 0 and 00 slots. So why didn't I cash out my savings account and run a casino (I live in Queens, just like Giovanni Ribisi's character from Boiler Room too...). I could hit a run of luck in the bettors' favor and my collateral would be gone in a night... That's my point. One doesn't have the resources to bet on a scale that would yield definite returns on marginal edges.

              I'm not saying that one should or shouldn't bet on baseball, just that some esoteric saber knowledge doesn't make any individual trial any more predictable. Besides, even if you had the best tool ever to know who the best players were, you could always count on their Luddite managers to prove why their IQ scores rival Rey Ordonez's seasonal RBI totals...

              In conclusion:

              1. Yes, people do make money betting on sports. People skilled enough to do so don't usually get into it by soliciting the advice of strangers on a message board and taking their word for it.

              2. If said strangers had discovered such foolproof technique to predicting said events, they wouldn't have the time (or energy) to get up from underneath Elisha Cuthbert long enough to sign on to said message board.
              Last edited by digglahhh; 04-24-2008, 01:56 PM.
              THE REVOLUTION WILL NOT COME WITH A SCORECARD

              In the avy: AZ - Doe or Die

              Comment


              • #8
                Originally posted by Stat One Author View Post
                Check out my book, Stat One, for insight to using P/E Averages to gauge an offensive player's value.
                Why don't you explain what it is and let the jury decide?

                I'd advise you to bring your flame-ratardant suit, I don't predict a pretty outcome.

                You completely neglect offensive context, relative stats, park advantages, and just about every other variable that one should ideally attempt to control for.
                Last edited by digglahhh; 04-24-2008, 02:06 PM.
                THE REVOLUTION WILL NOT COME WITH A SCORECARD

                In the avy: AZ - Doe or Die

                Comment

                Ad Widget

                Collapse
                Working...
                X