• Content count

  • Joined

  • Last visited

Community Reputation

0 Neutral

About yrforyg

  • Rank

Previous Fields

  • Favorite NFL Team
    Green Bay Packers
  1. I am a mac user who has been using the 'classic' windows draft dominator for many years (through vmware fusion/parallels/etc). I have been testing out the new mac-native draft dominator app (version 1.0.10). While the layout will certainly take some getting used to, I think the design is (overall) not bad and starting to grow on me. That said, I have some issues that I hope can be addressed: 1. Speed / Performance I really like the panel in the lower-right part of the screen that displays the current team that is on the clock and the available players. However, one big problem is that this panel is verrrrrrry sloooooow to respond to changing position filters. In a real draft, with limited time for each pick, I am often racing to keep current with all the picks and while this panel could be a helpful way to stay on top of things, the slow filter speed is just killing me. I am using a new, fast, retina macbook pro with lots of ram, and use this machine to play much more cpu/ram/graphic-intensive games, so it's definitely got the necessary horsepower. The app just seems to be poorly optimized in this regard. I don't know if this is specific to the mac build, but it needs attention. 2. Versioning Version 1.0.10 for Mac is the only version I've downloaded and used, so I don't have the full app upgrade experience yet, but from what I can tell, the only way to upgrade the app (when the next version is available) will be to go out to the website, download another zip file, unzip it, and move the new app bundle to replace the existing one. This is pretty clunky in an age when most apps automatically check and initiate upgrades from within the app itself. 3. Player data The app does not make it obvious when/if the player data has been updated. It would be great if the app could provide a proactive pop-up or other permanent visual cue indicating the date/time when data was last downloaded/updated. Yes, you can find this by clicking into the cloud menu, but why not just stamp it somewhere in the main UI (navigation bar/menu bar/etc). 4. Auction I don't know how to interpret the price data that the app provides during auctions. It seems clear that the "retail" price calculated becomes more and more discounted through the course of the draft as budgets diminish. How are these prices calculated? Is the basic recommendation to bid all players up to their current retail value? How do positional needs/team composition factor into the auction pricing - or do they? All things considered, although I am a big fan of the classic dominator, it has been clear for many years that a new app was needed. I'm hopeful about this one, but it does seem quite raw so far and I haven't decided yet if I'm ready to rely on it for this year's drafts. I hope there are more app updates planned in July and August, and hope the Mac platform isn't on the back burner.
  2. phrasing...
  3. What FAAB % for Asiata 10 team ppr? Worth dropping L Miller, MJD or V Cruz for him?
  4. Could one of the FBG staff maybe let us know for sure?
  5. I have not seen it yet -- apologies if I've overlooked it.
  6. I have been experimenting with the current (trial) version of Draft Dominator (15.0a - trial) to try to better understand the effect of various configuration options on the best value / DVBD calculations. One finding that has been somewhat surprising is that the baseline selected appears to have no impact on the recommendations ... almost. The almost part is what I found particularly strange. I tested by using a 12-team mock draft (16 rounds, start 1QB, 2RB, 2WR, 1 TE, 1DST, 1Flex[RB/WR]). DD was set to draft other teams by ADP. Strength of schedule and weekly weights were disabled, and auto look ahead was enabled. I drafted as team 5 of 12. All teams were set to use the 'Consensus' ADP list. I ran through the mock draft 5 different times, using a different baseline option for each attempt: 1. Joe's secret 2. worst starter 3. custom - qb16, rb38, wr46, te18, pk0, dst0 4. custom (a manual approximation of joe's secret) - qb11, rb42, wr24, te6, pk3, dst3 5. custom - qb0, rb0, wr0, te0, pk0, dst0 I completed a mock draft using each of the above, always selecting the player recommended by the best value window. All methods produced the same team: 1. Jimmy Graham 2. Giovani Bernard 3. Andre Johnson 4. Chris Johnson 5. Rashad Jennings 6. RG3 7. Cam Newton 8. DeAngelo Williams 9. Tavon Austin 10. Kenny Stills 11. Danny Amendola 12. Greg Jennings 13. Cincinnati D/ST 14. Roy Helu 15. Phil Dawson 16. James Starks ...Except, for the Joe's Secret method. Whether selected explicitly, or using the custom approximation, this produced a very slight variation in the latest rounds: 1. Jimmy Graham 2. Giovani Bernard 3. Andre Johnson 4. Chris Johnson 5. Rashad Jennings 6. RG3 7. Cam Newton 8. DeAngelo Williams 9. Tavon Austin 10. Kenny Stills 11. Danny Amendola 12. Greg Jennings 13. Cincinnati 14. Steven Hauschka 15. Jacquizz Rodgers 16. James Starks This was puzzling to me -- I would have expected baseline to have no effect on DVBD whatsoever, since it's supposed to be just finding the difference between two players (potential selection vs. projected dropoff player). I took a closer look to see what was going on here. In round 14, where the discrepancy occurs, here's what's happening Under Joe's secret, both Hauschka and Helu are available (pick 14.08 / 164 overall). Both have the following numbers in the best value window: Steven Hauschka (Dropoff 0, Drop Value 0, Need Mult 1, DVBD 0) Roy Helu (Dropoff 0, Drop Value 0, Need Mult .3, DVBD 0) Here I'm guessing that DD looks at DVBD, finds both equal, then looks next at the need mult, and recommends Hauschka because of greater need at PK. Under the other methods, like above, both Hauschka and Helu are available, but the numbers are slightly different now: Steven Hauschka (Dropoff 0, Drop Value 0, Need Mult 1, DVBD 0) Roy Helu (Dropoff 1, Drop Value 1, Need Mult .3, DVBD 0) I'm not sure exactly what's happening here, but Helu is recommended. My guess is the DVBD number displayed is rounded to 0 for display purposes, but actually slightly greater than 0 for Helu. In both cases, Helu (RB45) is presumably being compared against RB46 (Jacquizz Rodgers). Helu is projected for 71.9 pts and Jacquizz is projected for 71.0 pts. But the VBD number for each seems to be rounded differently, depending upon which baseline formula is used. Under Joe's secret, both have the same VBD number. Under the other methods, Helu's VBD number is 1 greater. This is all fairly minor, but I thought it was interesting. And if I'm right that this is all due to rounding discrepancies, it makes me wonder if perhaps the rounding issues might have a more pronounced effect under other circumstances.
  7. 6. We have 4 people that do preseason offensive projections: Dodds, Wood, Henry and Tremblay. Sigmund does provide projections during the season, but he already does a TON for us in other areas during the preseason. I don't see us changing this for 2013. He is too valuable creating and managing our content. Thanks David -- but what's confusing to me is that Bloom apparently DOES provide preseason projections. After all, like you, he's rated by FantasyPros for draft accuracy <http://www.fantasypros.com/nfl/accuracy/draft.php>, not just in-season or weekly accuracy.
  8. 3. This is one of those situations where we fully understand what people want, but can't provide. We project for the season, not for individual games. So we can't let people cut it anyway they want because we would have to assign probabilities and projections to the infinite possible buckets. Thanks for this reply David. I think you may have misunderstood the request though -- your projections are just fine. Here's how it works now (as I understand it) -- you (or the other projectors) project a total number of points allowed (yards allowed, etc.) for the entire season for each Team Defense. Draft Dominator then translates this full season projection into a per-game projection (points allowed per game, yards allowed per game). So far so good. Now, Draft Dominator then translates those stat projections into scoring projections, on the basis of the scoring configuration that is set up by the user. For most positions, the scoring criteria are extremely customizable (e.g. 6 pts per TD, .04 pts per PYD, etc). But this is not customizable for defenses. I'd like to be able to say "Defenses allowing between X and Y points per game earn Z fantasy points" (and/or "Defenses allowing between X and Y yards per game earn Z fantasy points"). Right now, only Z is customizable, but X and Y are fixed. So this is not a question of projections at all, it's simply a straightforward change in how DD translates projected stats into projected fantasy points. Hope this clarifies. Thanks again for taking the time to address these -- much appreciated!
  9. How do you suggest handling changes in league membership -- e.g. one past member is replaced with a new member whose tendencies are unknown?
  10. Might not be a bad idea to sticky this so folks don't keep creating new feature request threads
  11. 1. I would like a Mac version that does every single thing the Windows version currently does. 2. Include adp numbers for ESPN in the adp.csv file distributed with draft dominator. You already publish these on the website (http://subscribers.footballguys.com/apps/adp.php), so it should be extremely simple to put these in the csv file. 3. Rather than having hard-coded ranges for D/ST scoring (e.g. 0 points allowed, 2-5points allowed, less than 100 yards, 101-150 yards, etc) allow all ranges to be custom specified by the user so that they can precisely match the league's scoring system for D/ST. Same thing for Kickers. 4. Add a feature that auto-runs several mock drafts using a variety of different VBD baselines, then compares the strengths of the teams that are selected and (on the basis of the outcome) recommends the optimal VBD baseline to use for selecting the strongest team. 5. Allow adp lists to be weighted (e.g. 50% consensus, 50% expert) 6. Provide draft projections by Sigmund Bloom in addition to Dodds.
  12. So what's the verdict? Is he this year's A. Foster?