Agree Agree:  0
Likes Likes:  5
Results 1 to 10 of 10
  1. #1
    Contests
    Awards Showcase

    James7's Avatar
    Join Date
    Mar 2009
    Location
    Gaithersburg, MD
    Posts
    6,248
    Blog Entries
    3

    Bracket Prediction Contest: Your Feedback Wanted!

    Thanks all for helping make the first year of this little contest I designed a success. I know there's some room for improvement, so I want to hear from you what your thoughts are regarding it.

    First: Already on tap, the plan is to move away from combined scoring and just score ATP and WTA separately, awarding one medal to the winner of each. There will be no combined medal/standings.

    Now I'm opening up for some comments about other things. But I want to provide a little bit of my perspective as I do. The intent of the bonus category is to make it a different kind of contest. One in which there is still action at the end of the event, even for those who may not have any predictions remaining at that point. I think it succeeds in that regard. I had originally envisioned something where picks could be revised at a pentalty at any point during the actual run of the slam, but that became too complicated to think of a way in which it would work well.

    We have progressive scoring per round, so picks at each round of the tournament score one point more than the last. I like the gradual ramp compared to some other contests because I do think I want even first round match predictions to matter quite a bit here. Even if you get the (often big name) winners, you can lose out big by having your early picks be a mess. There's 64 points possible in the first round versus just 7 for the final winner of the tournament.

    We have had 4 unique winners at each tournament this year, and bonus scoring is never enough to overcome terrible predictions. The predictions of winners have been very solid. In an ideal world, I think maybe I would ask for game point predictions once those matchups were decided, but I think that it is too big of an ask to require input basically every day at the end from so many. As-is we ensure essentially even footing, with a quirk of making your rough set/game count count predictions based on who you expect to be there.

    I should note that we have basically a 'normal' distribution of bonus scores overall (excluding some who entered only one of two brackets). So generally, people are performing pretty evenly and consistently there.

    So that's my perspective here. In bullet form here are my ideals:
    • Not a generic bracket only.
    • Allows for late tournament action when other contests are otherwise 'locked in'.


    So what are your thoughts?

    Thanks!
    Last edited by James7; 09-12-2016 at 11:20 AM.
    I disapprove of this message

  2. #2

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    Congratulations feedback!!








    PS: I don't have any big feedback, at least for now. I just stopped playing as I have no real sense what is going on with the tours, and it has left me too frustrated. Heck, maybe think about medals for ATP, WTA and Combined? I'd also suggest maybe a bigger reward for those who get SF and F picks correct? IN any case, it's a terrifically put together and executed contest. Thanks for all of your work, James!
    Old News= Madison Brengle. New News- It's All About Amanda Fink Chichi Scholl, FULL CIRCLE OF LIFE MADISON BRENGLE BABY!!!!!!!!

  3. #3

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    I like the bonus as I've stated before. It gives you a shot even if you're not in an ideal position.

    Again, the only thing I might consider is adding a multiplying factor to the bonus as follows:
    x1 if you have none of the players in the match
    x1.5 if you predicted at least one of the players disputing the match
    x2 if you have both players

    The rest I'd leave the same.
    Meet again we do, old foe...

  4. #4
    Contests
    Awards Showcase

    James7's Avatar
    Join Date
    Mar 2009
    Location
    Gaithersburg, MD
    Posts
    6,248
    Blog Entries
    3

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    Quote Originally Posted by Drop-shot View Post
    I like the bonus as I've stated before. It gives you a shot even if you're not in an ideal position.

    Again, the only thing I might consider is adding a multiplying factor to the bonus as follows:
    x1 if you have none of the players in the match
    x1.5 if you predicted at least one of the players disputing the match
    x2 if you have both players

    The rest I'd leave the same.
    Interesting idea there.
    I disapprove of this message

  5. #5

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    Being totally biased as someone who came in 3rd I think in one of these this year, I was surprised that only one medal was awarded for this. Maybe it's because I've barely won anything around here, but placing in the top is pretty hard, so not sure if it's on the table, but awarding 3 medals is something to consider. Granted, I'd play either way, but it's a suggestion. I also think a Combined works along with WTA and ATP.

    For bonus points, I think the way it is now, works. I think any changes should be kept simple.

    In terms of late play. I'm not sure about the idea of revising picks, that feels like it could get messy and aids people who messed up early and penalizes those who make good initial picks. But I play on GrandSlamFantasyTennis.com and it's probably my favorite because I like the format. I'm not sure if you can pull ideas from there. Essentially, we pick our teams of 8 with some restrictions and those are locked in from the beginning and can't be changed. Then we also get points for picking matches accurately daily from the 4th round and later. It allows you to pad your already good picks or make up for your woeful choices, but I've never had a time where my terribly constructed team was saved by good picks in the late rounds, just improved my standings a bit. So, maybe that's something we could do here, pick the late rounds either all at once or the upcoming matches each day.

    But seriously, I'm sure I'll participate no matter what you decide. It's great that we're doing something for the whole bracket now during the Grand Slams.

  6. #6
    Contests
    Awards Showcase

    James7's Avatar
    Join Date
    Mar 2009
    Location
    Gaithersburg, MD
    Posts
    6,248
    Blog Entries
    3

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    Yes, it's a difficult challenge to win here for the effort put in up front. But many of the existing contests are. Retaining the combined may be doable, but that's something to discuss with the contest staff. Don't want to go overboard as most contests are only giving out one or two medals.
    I disapprove of this message

  7. #7

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    I like contests that require you to live with your choices from the get-go. I don't participate in the contests that offer mid-event trades. The value of a bonus system is to allow for further differentiation of the field of contestants, to spread out the values.

    As a point of comparison, YCWJ makes a pretty good stab at that, although I think the wooden spoon bonus values are a bit too high/rewarding

    That said, I'm not sure that's really needed for this event. I haven't seen a compelling argument based on data that suggests that the point spread needs to be further diversified. If it does, I would like to see a change from the total number of games in the later rounds. That strikes me as a crap shoot. When it comes to that element of the contest, I'm basically imagining a score between two players who may not even be playing. It's pretty much a complete guess. If that's what makes the difference between winning the event or not, I think it's misjudged.

    My $0.02.
    TIHBS TIHBW

    Avatar --> Politician 1: Philosopher King

  8. #8

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    Quote Originally Posted by Drop-shot View Post
    I like the bonus as I've stated before. It gives you a shot even if you're not in an ideal position.

    Again, the only thing I might consider is adding a multiplying factor to the bonus as follows:
    x1 if you have none of the players in the match
    x1.5 if you predicted at least one of the players disputing the match
    x2 if you have both players
    This is a great idea, and better than anything I could have come up with. My minor "lament" has been that someone could get the winner correct and only "barely" gain on someone who is a prescient points picker. That being said I am also sensitive to the amount of work James has to do. If adding multipliers makes it tremendously difficult - then I am against it :-).

    The picking interface? It's just fine; I have no qualms whatsoever with it.

    While I am COMPLETELY fine with the way things are...I cannot help but compare this to the annual NCAA bracket pools I fill out. I have been involved with many over the years, and there have thus been a plethora of varied scoring systems, as well as changes made to make them more interesting. Here are some items to throw out for consideration...
    Scoring:
    There was the standard 1 point per game pool (dismissed doesn't reward late round matches enough). Then it was 1,2,3,4,5 based on round. Then it was 1,2,4,8,16,32 (dismissed because it rewarded back end TOO much). It got us to where we are now, which might be of interest here (and allow bonus points to stay the same). Early on it doubles (1,2,4,8) but then stops doubling and goes up by half of what it should. It rewards the back end, but doesn't eliminate someone who is good early on.

    Another thing that might be interesting, but I suspect is too much work since it can't really be automated, is a reward for picking an upset. A couple ways this could be done:
    - Double points for picking a seed to lose in the first or 2nd round
    - double points for picking a seed (or higher seed) to lose in 1st-4th round

    Anything beyond that requires ranking all 128 and that gets way to complex.

    Anywho - thanks - I truly enjoyed this one and look forward to it next year as well as the consideration for feedback...

    -Mike
    Last edited by fastbackss; 09-15-2016 at 08:22 AM.

  9. #9
    Contests
    Awards Showcase

    James7's Avatar
    Join Date
    Mar 2009
    Location
    Gaithersburg, MD
    Posts
    6,248
    Blog Entries
    3

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    I'm not sure I want to give up on the twist inclusion entirely.

    The main issue being that just generic bracket contests already exist. And my main problem with them is illustrated below:
    ATPScoreByRound.png

    In the most recent case, the final standings based on just predictions were already "done" by round 4. This tends to be the case when bracket entries is the only factor. There's basically no activity because things get limited by design later in the tournament. And I don't think making tournament winners count for oversized portion of the score more is the answer. My bias is that the ability to have predicted first round match results is a better test of predictive score than knowing which big names will be alive at the end.

    Just trying to vocalize some of my thought process here. I appreciate all the comments and am still trying to think about it. I realize there's an inherent conflict in saying early rounds are important while also wanting late bracket action.
    Last edited by James7; 09-15-2016 at 09:21 AM.
    I disapprove of this message

  10. #10

    Re: Bracket Prediction Contest: Your Feedback Wanted!

    Quote Originally Posted by James7 View Post
    I'm not sure I want to give up on the twist inclusion entirely.

    The main issue being that just generic bracket contests already exist. And my main problem with them is illustrated below:
    ATPScoreByRound.png

    In the most recent case, the final standings based on just predictions were already "done" by round 4. This tends to be the case when bracket entries is the only factor. There's basically no activity because things get limited by design later in the tournament. And I don't think making tournament winners count for oversized portion of the score more is the answer. My bias is that the ability to have predicted first round match results is a better test of predictive score than knowing which big names will be alive at the end.

    Just trying to vocalize some of my thought process here. I appreciate all the comments and am still trying to think about it. I realize there's an inherent conflict in saying early rounds are important while also wanting late bracket action.
    The twist is great but it also reduces the contest to a game of chance since let's face it, upsets will happen, and at that point the game total is meaningless since it's between two players who you did not predict to play each other. So the only solution to that is to have game prediction occur after the match-ups are set, but that goes against having people do more predictions as the tournament progresses. So what's the solution...I don't know.
    Towel Avatar, do your thing!

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •