No announcement yet.

difference between custom draft guide and projections?

  • Filter
  • Time
  • Show
Clear All
new posts

  • difference between custom draft guide and projections?

    I am in a 4x4 AL only roto draft and am wondering why I get different values between the custom draft guide and the weekly projections. I have been assuming that R$ is based on the eight standard categories yet when I use the same 8 cats in the draft guide different values are generated. I am curious as to what the difference is. Thank you in advance.
    12 team AL 5x5

  • #2
    How many teams are in your league? I think the projections are based on a 12-team AL only league (and 13-team NL only).
    If at first you don't succeed, find out if the loser gets anything.
    -Bill Lyon


    • #3
      helloneumn's correct.


      • #4
        12 teams AL only. that's why I would assume the results should be the same.
        12 team AL 5x5


        • #5
          I get Arod $43 on projections and $39 on custom draft
          Crawford is $34 on projections and $36 on custom draft to name two.
          I might be doing something wrong but it looks right.
          The only difference I can see would be the offensive budget defaults to 66% on custom. What does player projections use?
          12 team AL 5x5


          • #6
            Ah, ok. tcb, the issue is that the projections and the valuator use different valuation methods. I believe the projections use the SGP method, so the hit/pitch split is basically irrelevant there.... $ values are based on the marginal value of each HR, RBI, W, K, etc.

            If you're 'forcing positions', that will skew things further, as will the 'top players' vs. 'balanced' settings.

            In short, you're seeing values that are a bit different on the top end, but they aren't intended to match up. If you used Rotolab, you'd see a subtly-different 3rd set of values.

            I wouldn't sweat it.


            • #7

              Does the SGP method give you a better model in a straight draft, or the PVM method?
              Any thoughts?



              • #8
                My off-the-cuff answer is that the PVM method plays better with the positional scarcity approach. But if you've got good SGP data (always a big "if"), that's more accurate overall.

                So, I have no strong preference.


                • #9
                  I think HQ should switch to a tiered pricing method so that instead of a $4 player, he's a $1-$5 tier. Instead of a $23 or a $21 player, they are both $20-$24 players, and so on.

                  Would eliminate these sorts of worries, and would be demonstrably as accurate, if not more so.
         / @NickRichardsHQ