19 Responses

  1. Mark
    Mark at |

    How do you handle bench spots for determining replacement level? In that case, it seems that the method you describe here could undervalue a guy like Wilin Rosario (high points/PA, low PA).

  2. Mark
    Mark at |

    I definitely agree about value being in flux. Kris Bryant isn’t earning anybody points crushing dingers in Iowa, but he could be an instant impact player once he gets to Wrigley.

    Chad Young’s ottoneu values over at RotoGraphs use a replacement level in terms of points/PA, with the final PAR calculated as (points/PA – replacement)*PA. It would seem more appropriate to use points/G over points/PA, but unfortunately the mass Steamer download doesn’t include a column for G on the hitter side.

    Piggybacking off your Bryant example, if I use the Fan projections (which do include G in the download) and ottoneu scoring, he comes in as the #16 3B for 2015 using raw point totals. The Fans say he’s the top 3B for points/G, but his 114 G projection drops him to #7 by (points/G – replacement)*G.

    I don’t have a perfect answer either. My gut feeling is that the deeper the bench, and the later the draft round, the more the points/G effect matters. But I’m not sure how much more.

  3. Scott
    Scott at |

    Tanner, it seems that steps 4-9 would need to be repeated whenever projections are revised. That’s a fair amount of data-entry work (and an inducement not to update projections). Is there no way to automate that?

  4. Alex
    Alex at |

    I’m trying to replace outdated Steamer rankings with more recent one’s. When I try to repeat earlier steps, it makes me reinput every single formula all over again. Is there an easier way to do this? Thanks.

  5. Alex
    Alex at |

    For example, I want to replace the steamer pitcher rankings from 2.15 with the one’s from 2.17. I DL the 2.17 data, format it in the proper way – player ID ahead of name, in a table, etc. I move the sheet into my main sheet with all of my formulas, name it Steamer_P (just like the one it was replacing) and delete the old sheet. When I look in my Pitcher Ranks however all of the columns are either 0’s or #REF. When I re-type in each formula, it notices the new sheet but there has to be a quicker way to update data, correct?

    1. Tanner
      Tanner at |

      Hi Alex, this is a really good question because I do try to design things so they’re easily reusable (from one season to the next or so you can just get updated projections and “drop them in”). Instead of deleting the current projection tabs, I would leave them in place but just delete out the projection data (leave the column headers). Then paste in the new projections (so you’re not moving the whole tab, you’re just copying and pasting the stats, that way the formulas looking to this table will remain in tact).

      I have a very similar example written up for this here. It’s specifically part of a Standings Gain Points series, but the same concepts apply.

  6. andrew
    andrew at |

    Hi Tanner,

    I was hoping you could help me out with my next step in a fantasy points league. I play in a league with 1350IP max. I have my projected totals for each player based off from my league’s scoring system. I have sorted out the 12C, 12 1B, 12 2B, 12 3B, 12 SS, 36 OF, and 24 UTIL, along with 108 SP. I created the Margin points based on subtracting the lowest point total batter (Yan Gomes 279 points in my league) from the other batters. I did the same for pitchers subtracting the 108th pitchers total from the other pitchers points, getting my margin points for all players. Then I added the margin points for the league and divided the total by 2844, to figure out the $ amount for each margin point. Then I multiplied the margin point of each player by that number to get their auction $ amount (all players in the leagued added up to $3120 just like it should).

    Here is where I’m stuck. In my league Clayton Kershaw is projected for 792 points over 220IP my $value on him is $46. Craig Kimbrel is projected for 480 points across 65IP and my $ value is $19. Kimbrel is way more valuable because he gets me more points per IP, but my $ value on Kimbrel is way less.

    How can I take IP into my calculations to get better $value for pitchers?

    I just feel like my $ values are off.

    Is my best plan to spend money on my closers/holders and then the best bats. Spend $1 on starting pitchers and just stream based on match ups?

    Sorry for the long post!

    Love the site

  7. andrew
    andrew at |

    Thanks for the response! I’ll have to fiddle with it.

    My plan is to stream pitchers throughout the year.

    I have another question related to 5×5 roto leagues that I’m having a tough time finding any background on.

    Is it a plausible strategy in a 12 team 5×5 league to not use SP throughout the year, stocking up on hitters, and closers during the draft. I would bottom out in W and K, but lead in WHIP,ERA,Saves and the hitting categories.

    When somebody 1st mentioned this strategy to me I thought it can’t work. Then I started thinking that it really could work with the elite closers and in a league without an innings pitched minimum.

    Any thoughts?

  8. andrew
    andrew at |

    Thanks Tanner, great thoughts as always

    1. Tanner
      Tanner at |

      You’re welcome! Sorry if I got a little heated. Can you tell I don’t like punting?!?!?

  9. andrew
    andrew at |

    No problem at all! It’s nice to find people that are passionate about fantasy baseball.

  10. Carl
    Carl at |

    For people in points leagues, how do you project blown saves? None of the projection systems I use project this category and yet it is used in my points league. I was thinking of just using a blanket number. For example, say every pitcher’s blown saves with equal 10 percent of their projected save total.

    Obviously that is very imprecise but I can’t think of a better way and don’t want closers to be overvalued if they aren’t being penalized for blown saves.

    1. Tanner
      Tanner at |

      Hi Carl,

      Wow, interesting thought. I have never bothered to project blown saves, mostly because I don’t think it can be done reliably AND because I don’t think there’s much of a spread there (the best closers are going to blow three or four and the worst seven or eight?).

      But the point that you don’t want to overvalue relievers because of it is worth thinking about. I think your simple solution of giving everyone 10 percent is a great solution to the problem.

      Playing devil’s advocate, it still may not matter much. Closers are so unusual. We can sit here and try to value them properly all day. But when push comes to shove, in a draft, their values don’t matter. When and where to draft a reliever is all game theory and just reacting to when everyone else will do it.

      With that said, I’m a stickler for trying to make things as accurate as possible. So I do like your simple solution to the problem.



Comments are closed.