Idea for Quarterfinals; Mix up the Order of Corps

[Tez]

Not sure if I miswrote something or you misread something. I meant for the shows to get progressively better throughout the night, not the other way around. ie. Pioneer on first and Blue Devils on last (just picking two corps)

Oh... really sorry about that... I just mis-read your post.... I actually really agree with you.

[Tez]

Why are one corps scores dependant in any way on the performance of other corps? Shouldn't the judging be evaulating each corps based only on their own performance?

It is the concept that a corps might be scored based on the percieved score of the "winning corps to be seen later" isn't the best way to evaluated performances in the first place that might be somewhat aleviated by have a random draw at the only contest that has all of the corps in it in the first place.

As to the argument that some audience members only really want to see the winners (or near winners), then those folks should wait until semifinals, and come in late.

Yes, a judges does evaluate a corps on it's own performance, but that is only part of it. How can you honestly say that they cannot directly compare corps to corps when the levels are so similar? Here is my prior post where I gave a specific example:

I really don't think going on "last" necessarily has any advantage, the advantage is that when corps compete closely to ones who they are close in level to, they are scored more accurately and fairly. So yeah, the way we have it now is easier on the judges. But shouldn't it be like that? Why make it harder on the judges by making them "remember" how a corps performed several hours ago, when comparing to the one they just saw? Well, you'll say, it's supposed to be about the sheets. The sheets, are just a guideline, nowhere on the sheets does it describe what the difference between handing out a 17.4 and a 17.6 is. The sheets are really more of a guide for the corps, and they tell the judges generally where a group should be scoring. When we are talking about a group of corps who are all very similar in achievement level (like 7 corps within 4 points)... every little 10th of a point a judge decides one way or another makes a big impact on the scores. I don't in anyway see how judging by the sheets will help here. If any one of the 12 judges last year at finals would have scored the Blue Devils just one tenth higher in any subcaption, we would have had a tie. Think about it this way.... what if two corps (let's say Crown and BD) are scored by the brass judge and he determines their performances were almost identical in achievement.... but he feels that BD had just the slightest of edges.... going by the sheets, they were easily both "box 5." How can you tell me that the judge should not directly compare these two corps? How could he do this accurately if they performed 4 hours apart? This is why I believe that a random draw would actually create less accurate scores for the corps.

The only way you could accurately score corps and NOT directly compare them would be if the sheets had a different achievement description for every increment in score a judge could possibly give. This is why spreads are everything in drum corps judging.

[Bruckner8]

I really don't think going on "last" necessarily has any advantage, the advantage is that when corps compete closely to ones who they are close in level to, they are scored more accurately and fairly. So yeah, the way we have it now is easier on the judges. But shouldn't it be like that? Why make it harder on the judges by making them "remember" how a corps performed several hours ago, when comparing to the one they just saw? Well, you'll say, it's supposed to be about the sheets. The sheets, are just a guideline, nowhere on the sheets does it describe what the difference between handing out a 17.4 and a 17.6 is. The sheets are really more of a guide for the corps, and they tell the judges generally where a group should be scoring. When we are talking about a group of corps who are all very similar in achievement level (like 7 corps within 4 points)... every little 10th of a point a judge decides one way or another makes a big impact on the scores. I don't in anyway see how judging by the sheets will help here. If any one of the 12 judges last year at finals would have scored the Blue Devils just one tenth higher in any subcaption, we would have had a tie. Think about it this way.... what if two corps (let's say Crown and BD) are scored by the brass judge and he determines their performances were almost identical in achievement.... but he feels that BD had just the slightest of edges.... going by the sheets, they were easily both "box 5." How can you tell me that the judge should not directly compare these two corps? How could he do this accurately if they performed 4 hours apart? This is why I believe that a random draw would actually create less accurate scores for the corps.

Why should one corps performance be compared to the other? I claim the corps should never be compared to anything except the sheets. But a huge contingent of the DCI community (incl insiders, corps, fans) demands "rank." I've heard conversations in critique (staff saying to judge) along the lines of "Couldn't you make a MF-ing decision?" (I watered it down...family show!) I take it you feel the same, b/c you put stock in the judges' abilities to recall a certain prior performance, so that he can "rank it properly."

If they follow the sheets to the letter (ie, more science than art), the order would not matter, and we'd get more fair outcomes. And yes, sometimes the outcome would be different, anytime the corps are very very close! DUH!

The question for all judged events should be: How can we make it as fair as possible?

If everyone agrees that having the same corps going on last is an advantage, then it's our responsibility to NOT do that! Every reason that has been pointed out against random order (keep everything the same) is valid, except if you include "fair." People justify the "fairness" by claiming that it's a reward for other success. That's fine, as long as you admit that you're inherently making each show slightly unfair in the process. And that's cool, BTW. Just quit urinating on me and telling me it's raining,OK?

(POP QUIZ for those who know me, in person or as a DCP Veteran: Can't you see how this is also part of the Competitive Inertia?)

[Liam]

(In reply to Tez -- two posts up ...)

Thing is, though, that there are specific historical examples where judges were able to do this -- I listed a couple a few posts up.

Also, BOA does this all the time -- both prelims, which happen over 2 days, and finals are random order. I realize BOA and DCI aren't exactly the same thing, just saying that actual historical example have shown that judges are able to handle it. Speculation that they can't is just that -- speculation.

Edited July 10, 2009 by Liam

[j.morgus]

The only way you could accurately score corps and NOT directly compare them would be if the sheets had a different achievement description for every increment in score a judge could possibly give. This is why spreads are everything in drum corps judging.

The newest judges sheets I have seen are over 20 years old, and I haven't seen them in almost that long, so I can't argue on your interpretation. I am saying that in a perfect world, a judge should be able to say that if in a given caption a corps' material was worth X points, and they performed it at 95% of the best possible performance, then they should score .95X. If an other corps happens to also score .95X in that caption, then so be it. The spread should be calculated by the tabulator, and not be a concern of the judge. If there is a tie, then there is a tie.

Corps very well might not be judged this way, but ideally, a corps' score should be the same regardless of what competition they face at a given show. It should be as if they were the only ones there, and not be assigned based on the need to assure the spread.

Edited July 10, 2009 by j.morgus

[dbc03]

The newest judges sheets I have seen are over 20 years old, and I haven't seen them in almost that long, so I can't argue on your interpretation. I am saying that in a perfect world, a judge should be able to say that if in a given caption a corps' material was worth X points, and they performed it at 95% of the best possible performance, then they should score .95X. If an other corps happens to also score .95X in that caption, then so be it. The spread should be calculated by the tabulator, and not be a concern of the judge. If there is a tie, then there is a tie.

Corps very well might not be judged this way, but ideally, a corps' score should be the same regardless of what competition they face at a given show. It should be as if they were the only ones there, and not be assigned based on the need to assure the spread.

That's fine. But here in the real world judges aren't machines that can pick a number out of the air for each individual corps and have it be perfectly accurate.

Spreads are more important than strict interpretation of the sheets

[j.morgus]

That's fine. But here in the real world judges aren't machines that can pick a number out of the air for each individual corps and have it be perfectly accurate.

Spreads are more important than strict interpretation of the sheets

More important how? More important to the winner? To the fan? To the third place corps?

Maybe more important in assuring a placement? But should placement be the concern of an adjudicator? Spreads and placement only become more important if we work from the assumption that the object of the endeavor is to win. If we look at this from the viewpoint that the competition is the endeavor, then adherence to the judging criteria is "more important" than picking the winner.

[Tez]

If everyone agrees that having the same corps going on last is an advantage, then it's our responsibility to NOT do that! Every reason that has been pointed out against random order (keep everything the same) is valid, except if you include "fair." People justify the "fairness" by claiming that it's a reward for other success. That's fine, as long as you admit that you're inherently making each show slightly unfair in the process. And that's cool, BTW. Just quit urinating on me and telling me it's raining,OK?

But I just made a case that it is more fair if we don't have a random order. And everyone does NOT agree that going on last is an advantage. At least I don't and I'm sure there are others. I just think it allows judges to makes better decisions, which equals more fairness.... plus it's better for the audience.

Now if you want to make a case for random orders based on the fact that some corps never get to perform at night or in front of large crowds... I think that is completely valid. I just don't agree that it would make the COMPETITION more fair.

[bluesop08]

Why should one corps performance be compared to the other? I claim the corps should never be compared to anything except the sheets. But a huge contingent of the DCI community (incl insiders, corps, fans) demands "rank." I've heard conversations in critique (staff saying to judge) along the lines of "Couldn't you make a MF-ing decision?" (I watered it down...family show!) I take it you feel the same, b/c you put stock in the judges' abilities to recall a certain prior performance, so that he can "rank it properly."

If they follow the sheets to the letter (ie, more science than art), the order would not matter, and we'd get more fair outcomes. And yes, sometimes the outcome would be different, anytime the corps are very very close! DUH!

The question for all judged events should be: How can we make it as fair as possible?

If everyone agrees that having the same corps going on last is an advantage, then it's our responsibility to NOT do that! Every reason that has been pointed out against random order (keep everything the same) is valid, except if you include "fair." People justify the "fairness" by claiming that it's a reward for other success. That's fine, as long as you admit that you're inherently making each show slightly unfair in the process. And that's cool, BTW. Just quit urinating on me and telling me it's raining,OK?

(POP QUIZ for those who know me, in person or as a DCP Veteran: Can't you see how this is also part of the Competitive Inertia?)

It would make sense in the essence that a corps would have to work their way into a later slot through past performances before they would have a shot at the title. Even if they are good enough to actually win, they may not if they go too early and a judge rates them low/mid to give breathing room on both sides of their score for other corps to be placed. I know that this happens for a fact during All-State/ All-Region Band try-outs (in Texas), new judges are instructed to place the first performers in a more "mid-level" position on the scores to leave room for kids before/after, making it extremely difficult for the first performers to compete. This is also why the order is random, and that with each etude the competing order is changed.

[Tez]

The newest judges sheets I have seen are over 20 years old, and I haven't seen them in almost that long, so I can't argue on your interpretation. I am saying that in a perfect world, a judge should be able to say that if in a given caption a corps' material was worth X points, and they performed it at 95% of the best possible performance, then they should score .95X. If an other corps happens to also score .95X in that caption, then so be it. The spread should be calculated by the tabulator, and not be a concern of the judge. If there is a tie, then there is a tie.

Corps very well might not be judged this way, but ideally, a corps' score should be the same regardless of what competition they face at a given show. It should be as if they were the only ones there, and not be assigned based on the need to assure the spread.

You make some good points here... and I guess in a perfect world that would be the way corps are scored.... but don't you agree that assigning a number based on the difference in performance from the previous groups is easier than figuring out if the group was at a 95% achievement level? And if it is easier, the judge can do it more accurately.....right?