No announcement yet.

Request to change TBP system

  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Another more radical proposal: the top teams could be more fairly ranked and sorted with a change in the qual match schedule near the end of an event.

    One option for qualifier events would have teams play a random schedule for 3 matches and then the top 1/4 teams could be scheduled to face each other with mid-ranked partners for the final 2 qual matches, resulting in less random and more fair head-to-head types of matches at the ends of quals. At Worlds this could be particularly easy to do by releasing a "final day" match schedule the night before based on the rankings up to that point.

    Or an even more radical concept: after qual matches, have the top 6 teams face off in three 1 v 1 matches, with their relative scores determining the final top 6 rankings. This would add a new stage of matches between regular quals and regular elims (kinda-sorta like the round-robin stage that FRC has at worlds before their finals, though with many obvious differences).
    Last edited by Cheer4FTC; 05-14-2019, 10:02 AM.


    • #17
      I am a rookie FTC coach, but I have mentored on an FRC team the last three years.

      It seems to me that who you are assigned with in FTC plays a much larger role than FRC for three reasons:
      1) You have far fewer matches - so each one counts for more.
      2) There are two members per alliance instead of three. So the amount of impact each team has is 1/4th instead of 1/6th.
      3) There are many fewer teams that get to play in elims. Only 4 captains instead of 8.

      I *love* that FTC is trying to keep the scoring where it is easy to explain. I *hate* explaining district points to people new to FRC. They almost feel intentionally complicated. Thank you for your clear explanation of what your goals would be for any replacement for TBP.

      While only tangentially related to TBP, I would *love* to see FTC go to 8 alliances with a quarter-final, semi-final, and final. On the downside, this would make the tournament day longer. On the positive side, this would allow twice as many teams to play in elims and would reduce the impact of TBP as more teams would get to be alliance captains. It would increase the number of teams that need to be scouting, Also, brackets are exciting with the opportunity for upsets.

      Rookie Coach FTC #16072


      • #18
        Alan, most qualifying tournaments are between 20-28 teams, so 8 alliances in all tournaments would not be a necessary improvement. The 80-team world championship division is a great place for 8 alliances. I think most big state tournaments use 2 reasonably-sized divisions rather than one big division. There should probably be a threshold set (48 or more perhaps) for when to go to 8 alliances.


        • #19
          Our district FRC events (typically 32 or less teams) have 8 alliances and with 3 teams on each alliance, that means most of the teams are playing in elims.


          • #20
            Hi Folks,

            Please follow the link to our blog post on TBP. Feel free to email me with your thoughts. jhalloran(at)firstinspires(dot)org. Thanks!!



            • #21
              Having read through the thread several thoughts come to mind:
              • It is easier to teak a system then to replace it. And most people will understand such faster.
              • It should be considered a breach of Gracious Professionalism to even suggest a team do less than their best in competition.
              • Scoring for one's opponent should probably be banned as it does not give an idea of their strength. It does throw off such results.
              • Either way, the legality of scoring for the opponent should be specified in the game rules. (I would think it good to prohibit intentional or repetitive scoring for the opponent.)
              • I would highly favor putting emphasis on the autonomous portion of the match in tie breakers. This doesn't require a change in scoring at the moment. It does encourage the most real world benefits of the program..
              • I think 10 to 20% of teams should move into a face off.


              • #22
                Originally posted by ftcsachse0 View Post
                ... I would highly favor putting emphasis on the autonomous portion of the match in tie breakers ...
                The old objective for TBP being based on losing alliance score was to motivate stronger teams to help weaker teams [prior to a tournament] so as to yield higher TBPs for stronger teams at tournaments. The new objective is to prevent blowout matches. I agree that TBP should be entirely weighted, or otherwise heavily weighted, on your own alliance's autonomous performance (i.e. how well your alliance performs during the actual robotics portion of a match), but unfortunately this does not satisfy the new objective of TBP.

                The deeper issue is that in order to be a legitimate sport FIRST needs to conduct legitimate tournaments. At worlds you had eighty teams in qualification rounds and only four alliances advancing to elimination rounds. This falls short of a legitimate tournament, especially considering the high degree of random factors that can affect match outcomes in qualification rounds. Further, having a sufficient number of alliances in elimination rounds helps alleviate the flaws in TBP.


                • #23
                  I know this post is old, but thought I'd add a simple thought. Why not just give a weighted score? AKA your points scored gives you a metric of .1 or .01 while the other teams points could have a weight of .075/.050 or .0075/.0050 respectively. So it is always in your best interest to score your own points, but if you have strong competition it is going to help your case vs another team that is scoring about the same as you.


                  Team 1 through 5 matches scored a total of 537 points. Their opponents scored 400. 537x.1 + 400x.05 = 53.7 + 20 = 73.7 TBP/RP/Whatever you wanna call it.

                  Because it is weighted towards your points you won't have an incentive to score points for your opponent, while you would get a bonus for having stronger competition. I don't think it's perfect, but it might be closer to a good middle ground.