Announcement

Collapse
No announcement yet.

Strength of Opponent

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Strength of Opponent

    How is CS calculated? Is there a specific formula that can be used?


    Thanks

  • #2
    Re: Strength of Opponent

    CS is calculated by adding the total number of ballots won by your opponents.

    Here is the language from the Tab Manual (which you can view on the AMTA Website, on the Tabulation Page).


    Determining a Team’s Combined Strength

    For each team involved in the tiebreaker, make a list of that team’s four opponents. Next to each of those opponents, list the number of ballots won by that team. Add these four numbers to determine the total number of ballots won by a team’s opponents, a.k.a its “combined strength.”

    Example: Team 1509
    Opponents Record
    1001 8 ballots won (8-0)
    1323 1 (1-7)
    1414 4 (4-4)
    1178 5.5 (5-2-1)
    Total: 18.5

    Note that while the term “combined strength” represents the same thing as in years past, the number that represents a team’s CS will now be lower. The old method doubled the record of each opponent (because there are two ballots per round). The results of calculating CS should be exactly the same under both systems. If a team has a larger CS number it is more likely to prevail in the tiebreaker.

    Comment


    • #3
      Re: Strength of Opponent

      That is surprisingly simple, probably should have figured that one by looking! Thanks for your help!

      Comment


      • #4
        Re: Strength of Opponent

        Here's something I've always wondered: why is combined strength, rather than another measure, used as a tiebreaker? Do you think there's a better system of tiebreaking out there? Combined strength can be frustrating because it's determined by how well your opponents do, so it's out of your team's control.

        Comment


        • #5
          Re: Strength of Opponent

          [quote author=Columbo link=topic=4902.msg193599#msg193599 date=1297827533]
          Here's something I've always wondered: why is combined strength, rather than another measure, used as a tiebreaker? Do you think there's a better system of tiebreaking out there? Combined strength can be frustrating because it's determined by how well your opponents do, so it's out of your team's control.
          [/quote]

          Macalester uses point differential as the highest priority tiebreaker for their tournaments. That can be frustrating too though, because then teams are rewarded for being paired against an easy team first round and winning by a huge margin. Being paired against tougher teams is also out of your team's control.

          Comment


          • #6
            Re: Strength of Opponent

            [quote author=Columbo link=topic=4902.msg193599#msg193599 date=1297827533]
            Here's something I've always wondered: why is combined strength, rather than another measure, used as a tiebreaker? Do you think there's a better system of tiebreaking out there? Combined strength can be frustrating because it's determined by how well your opponents do, so it's out of your team's control.
            [/quote]



            If you end up in a tie with ballots, the only way to fairly compare who's ballots were "harder to earn" is to look at how tough the competition was. Yes, one team may have a lower CS because the teams they hit didn't do well, however since the team went against easier teams they should have taken more ballots. So the team that ends up being ranked higher did in fact take the same amount of ballots, but against harder teams. If the other deserved to be ranked higher, theoretically, they should have been able to take more ballots from the easier teams. Of course this is all still subject to the inherent subjectiveness of mock trial.

            Comment


            • #7
              Re: Strength of Opponent

              [quote author=self-authenticating link=topic=4902.msg193603#msg193603 date=1297828429]


              If you end up in a tie with ballots, the only way to fairly compare who's ballots were "harder to earn" is to look at how tough the competition was. Yes, one team may have a lower CS because the teams they hit didn't do well, however since the team went against easier teams they should have taken more ballots. So the team that ends up being ranked higher did in fact take the same amount of ballots, but against harder teams. If the other deserved to be ranked higher, theoretically, they should have been able to take more ballots from the easier teams. Of course this is all still subject to the inherent subjectiveness of mock trial.
              [/quote]

              Your reasoning may apply to situations where one team has a CS of 19 and the other has a CS of 12, but it doesn't apply to all situations, and maybe not even the majority of them. Imagine two teams, both of whom end up with a 7-1 record. Team 1 earned a tough +10, -2 split with an opponent who finished 6-2, and positively obliterated its other three opponents to the tune of 20 or so points per ballots, though those opponents ended up with records of 2-5-1, 3-5, and 4-4. Team 2 split +2, -15 with an opponent who ended up 4-3-1, and squeaked by its other three opponents, whose records were 4-3-1, 4-4, and 3-4-1, to the tune of 1-5 points per ballot. Team 1 has a record of 7-1, a CS of 15.5, and a point differential of over 100. Team 2 has a record of 7-1, a CS of 16.5, and a point differential of around +30. Which team do you think is better?

              It's obviously Team 1, but Team 2 gets the victory if all you look at is combined strength. What's more, it isn't the case that Team 1 "should have been able to take more ballots from the easier teams." Team 1 took all the ballots from the easy teams it faced, and had a hard-luck split with a team that finished 6-2. Team 2's only lost ballot came by a much sketchier margin against a much worse team. Yet it still wins the CS battle because it squeaked by mediocre teams rather than blowing out bad ones.

              What's more, the first round at almost every tournament is paired randomly. I can tell you from experience that if you get randomly paired in the first round against a team that eventually ends up 0-8, your only recourse is to go undefeated. It's especially fun when it happens at the national championship tournament. And sometimes, even going undefeated isn't enough. At least two teams have gone 8-0 in their division at the championship tournament and lost a tiebreaker for the final round.

              Point differential certainly isn't perfect, and you can think of a situation analogous to the one I proposed above that would give the victory to the wrong team based on PD. Ideally, I'd like something akin to the fractional win system that they use in the advanced tab summary for the Beach Party--something that combines the advantages of the two systems. But if I'm forced to choose between straight CS and straight PD as a tiebreaker, I'd choose PD. It at least puts a team's destiny in its own hands.

              Comment


              • #8
                Re: Strength of Opponent

                In addition to what self-authenticating wrote, there's another huge factor in CS' favor: it's quick and easy to calculate, and, as a result, is less error-prone. I'm all for getting awards ceremonies started sooner, and without tabroom errors.

                The old primary tiebreaker was "SOO," which was notoriously time-consuming. I won't go so far as to say it was difficult, but it has multiple steps and is therefore somewhat confusing. SOO is still in the tiebreaker arsenal, but now falls behind head-to-head, CS, and OCS, so it's much more rare that it's used these days.

                Take a recent tab summary with lots of tied teams, read the tab manual on how to do the SOO breaker, and try it out sometime. Time yourself, too. Then, do the CS instead, and tell me which one you like better =)
                I post in my personal capacity, not on behalf of AMTA.

                Comment


                • #9
                  Re: Strength of Opponent

                  Just out of curiosity, Dan: who is the other team to go undefeated and lose the tiebreak at nationals? I know Mac did it (was it 2002?) But I didn't know it had happened twice.
                  Minnesota Mock, doing it right since 1987

                  Comment


                  • #10
                    Re: Strength of Opponent


                    When it comes down to it, scoring mock trial is not a perfect science as the entire "sport" is subjective and luck can certainly play into it a great deal. Although it may not necessarily be fair, as you pointed out using point differentials isn't always fair either. To go even beyond tie breakers, even scoring on individual ballots themselves are sometimes nonsensical. Someone pointed out that Yale took a ballot by +40 one round and tied the other. There are always absurdities and, of course, it always helps to have luck on your side!

                    Comment


                    • #11
                      Re: Strength of Opponent

                      Nobody is advocating a return to SOO. That was a nightmare.

                      Comment


                      • #12
                        Re: Strength of Opponent

                        [quote author=TheHamborgler link=topic=4902.msg193606#msg193606 date=1297830559]
                        Just out of curiosity, Dan: who is the other team to go undefeated and lose the tiebreak at nationals? I know Mac did it (was it 2002?) But I didn't know it had happened twice.
                        [/quote]

                        Mac was one team, though it was before my time, so pre-2001. The other was Cal-Berkeley in 2004, which lost an 8-0 tiebreaker to Columbia if I recall correctly.

                        Comment


                        • #13
                          Re: Strength of Opponent

                          [quote author=TenaciousDRB link=topic=4902.msg193604#msg193604 date=1297830128]
                          Your reasoning may apply to situations where one team has a CS of 19 and the other has a CS of 12, but it doesn't apply to all situations, and maybe not even the majority of them. Imagine two teams, both of whom end up with a 7-1 record. Team 1 earned a tough +10, -2 split with an opponent who finished 6-2, and positively obliterated its other three opponents to the tune of 20 or so points per ballots, though those opponents ended up with records of 2-5-1, 3-5, and 4-4. Team 2 split +2, -15 with an opponent who ended up 4-3-1, and squeaked by its other three opponents, whose records were 4-3-1, 4-4, and 3-4-1, to the tune of 1-5 points per ballot. Team 1 has a record of 7-1, a CS of 15.5, and a point differential of over 100. Team 2 has a record of 7-1, a CS of 16.5, and a point differential of around +30. Which team do you think is better?

                          It's obviously Team 1, but Team 2 gets the victory if all you look at is combined strength. What's more, it isn't the case that Team 1 "should have been able to take more ballots from the easier teams." Team 1 took all the ballots from the easy teams it faced, and had a hard-luck split with a team that finished 6-2. Team 2's only lost ballot came by a much sketchier margin against a much worse team. Yet it still wins the CS battle because it squeaked by mediocre teams rather than blowing out bad ones.

                          What's more, the first round at almost every tournament is paired randomly. I can tell you from experience that if you get randomly paired in the first round against a team that eventually ends up 0-8, your only recourse is to go undefeated. It's especially fun when it happens at the national championship tournament. And sometimes, even going undefeated isn't enough. At least two teams have gone 8-0 in their division at the championship tournament and lost a tiebreaker for the final round.

                          Point differential certainly isn't perfect, and you can think of a situation analogous to the one I proposed above that would give the victory to the wrong team based on PD. Ideally, I'd like something akin to the fractional win system that they use in the advanced tab summary for the Beach Party--something that combines the advantages of the two systems. But if I'm forced to choose between straight CS and straight PD as a tiebreaker, I'd choose PD. It at least puts a team's destiny in its own hands.
                          [/quote]

                          Wouldn't using PD as a tiebreaker make those ridiculous +40, -2 rounds even more harmful to overall results? Sorry if I'm missing something here, but it seems to me that's more likely than the situation you described. It happens in pretty much every tournament.

                          If you take the regional we were in for example, we ended up with a 5-3 record, but having run into W&L A and GW A and having some weird splits with them resulted in us having a negative point differential. Using CS we came out at the top of the 5-3 teams. If we had used PD we would have lost out to 5-3 Davidson, whom we took two ballots from by significant margins in the first round and whom had a significantly lower CS than us at the end of the tournament. Had we been using PD it would have been clearly to our advantage to deliberately drop both ballots in round 2 (or throw round 1 right off the bat) and dodge GW/W&L because strength of schedule wouldn't even be a concern. Obviously that would be immoral, but I'm not sure we want to build any incentive into the scoring system for teams to be doing that.

                          It just seems like the problems with CS arise from bad judging, whereas using PD results in less representative results.

                          Prioritizing strength of schedule makes it worth it for a contending team that may not be quite an elite program to still go all out every round and try to face the toughest opponents, rather than trying to drop ballots early to run into soft competition and sneak through later on.

                          I could be completely wrong, I don't really get too into the number crunching. Feel free to educate me.

                          Comment


                          • #14
                            Re: Strength of Opponent

                            [quote author=TenaciousDRB link=topic=4902.msg193604#msg193604 date=1297830128]
                            Point differential certainly isn't perfect, and you can think of a situation analogous to the one I proposed above that would give the victory to the wrong team based on PD. Ideally, I'd like something akin to the fractional win system that they use in the advanced tab summary for the Beach Party--something that combines the advantages of the two systems. But if I'm forced to choose between straight CS and straight PD as a tiebreaker, I'd choose PD. It at least puts a team's destiny in its own hands.
                            [/quote]

                            Ultimately, that's the problem in selecting any tiebreaker -- no matter which one you pick, there's always a situation where you can make the argument that the tiebreaker reaches the "wrong" result. The problem I (and others) have with straight PD is it doesn't so much put a team's destiny "in its own hands" as much as it puts it in the scoring quirks of the particular judges. I'm more comfortable looking at ballots in the aggregate as opposed to whether or not individual judges are either high scorers and/or are using the full scoring range.

                            Secondly--and especially at Regionals where there's more variation in the skill levels of the teams--PD tends to have an inverse relationship with the strength of opposition (much as in your hypothetical above.) You more often see the huge PD numbers when you're blowing out the less skilled teams as opposed to winning close matches with average to above-average teams. I'd be more comfortable with putting more stock into PD at Championship, where one could make the argument that all of the teams are of comparatively equivalent strength to begin with.

                            Notably, in your hypo, it seems to me that you really hang the distinction between the two teams on the fact that the first team had a better quality loss than the second team. Assessing the "quality of losses" was one of the benefits of the old SOO breaker, which, as I mentioned above, has been largely retired for good reason. But looking at the two teams' schedules in the aggregate, the second team *did* have the tougher schedule.

                            Geez, it really is Old Home Week in here if I'm debating tiebreakers with fellow old mockers. Will there be a discussion of neckties soon?
                            I post in my personal capacity, not on behalf of AMTA.

                            Comment


                            • #15
                              Re: Strength of Opponent

                              [quote author=Spidey link=topic=4902.msg193611#msg193611 date=1297831087]
                              If you take the regional we were in for example, we ended up with a 5-3 record, but having run into W&L A and GW A and having some weird splits with them resulted in us having a negative point differential. Using CS we came out at the top of the 5-3 teams. If we had used PD we would have lost out to 5-3 Davidson, whom we took two ballots from by significant margins in the first round and whom had a significantly lower CS than us at the end of the tournament.
                              [/quote]

                              I don't think that's quite correct- head to head overrules CS, so you would have beaten them anyway.
                              I still see your point, but in this specific example you would have advanced instead of Davidson.
                              Attorney 1: Objection!
                              Presiding: Response, counselor?
                              Attorney 2: I disagree!

                              Comment

                              Working...
                              X