2010 - The most BORING DCI championships ever!

[Michael Boo]

judging a large show is tricky.

and i'm not accusing him of this, but sometimes you see a recap like that and you wonder if the subs ended up that way to keep from having a bottom-line tie. Not saying he did it that way, but I've seen it done, and it sucks.

All I know based on my own experience is that my mind doesn't work fast enough when I'm judging a marching band show to make that happen, even if I wanted to do so.

It's human nature to wonder such things, though. When you get down to the bottom line, there's almost nothing a judge can do that isn't going to create suspicion somewhere from someone. The real wonder is how we continue to get people to become judges and subject themselves to that. You've got to have a thick skin to be a judge.

[Jeff Ream]

you do. imagine being Friday's drum guy...9.9/9.8 with 5 corps to go.

and...Crown beating Cadets in drums at finals. I never thought it, but have heard many people say "would that have happened if the judge didn't go help Ryan when he went down?"

[bccadet09]

Are you kidding me? Bluecoats came into quarters seeded 2nd overall and this was the most boring World Championships ever? Respectfully I disagree.

I think this season was more exciting in terms of competition. It got so far to see Blue Knights beating SCV one time, and Blue Stars getting their next highest score in their history....and many more.

[littlebirdy]

boo hoo. Do you want the judges to put out fake scores just to appease you?

[Bob J]

Are you kidding me? Bluecoats came into quarters seeded 2nd overall and this was the most boring World Championships ever? Respectfully I disagree.

I think this season was more exciting in terms of competition. It got so far to see Blue Knights beating SCV one time, and Blue Stars getting their next highest score in their history....and many more.

Let me clarify my post.

I am only talking about the consistancy in placings from quarters to semis to finals. Places 1-17 in semis were the same as quarters. Placings 1-12 in finals were the same as semis and quarters. When looking ONLY at the overall placings of the corps in these 3 shows on 3 consecutive nights things look very consistent/boring. While I'll admit I do like to log in to the web to see the results of the shows each night, and I do get excited when I see corps B beat corps A when corps A won last night, I don't think the judges should fool with the numbers to make it exciting for me. I am just pointing out the statistical oddity that occurred at the end of this season.

[Jeff Ream]

Let me clarify my post.

I am only talking about the consistancy in placings from quarters to semis to finals. Places 1-17 in semis were the same as quarters. Placings 1-12 in finals were the same as semis and quarters. When looking ONLY at the overall placings of the corps in these 3 shows on 3 consecutive nights things look very consistent/boring. While I'll admit I do like to log in to the web to see the results of the shows each night, and I do get excited when I see corps B beat corps A when corps A won last night, I don't think the judges should fool with the numbers to make it exciting for me. I am just pointing out the statistical oddity that occurred at the end of this season.

by finals tho....this year anyway...i only truly feel only one corps should have moved up from where they were the 2 nights before ( and IMO, should have been that way both nights).

99% of the time, you're only going to see so much improvement in those last final days. Yeah it does happen from time to time, but not as much as you think

[glory]

...Quite honestly, I am coming around to the idea that we just need to drop scores and move to ordinals. Let's let the judges not have to worry about leaving spreads between corps and instead focus on which group is better.

I'm 100% with Mike on this one. If we can draw any conclusion from this discussion and all those like it, it is that incremental distinctions of scoring are problematic at best and unscientific in general. The distinctions are often too fine for a scoring system so simple and so subjective. Better to rely on the ordinals for ranking and averaging. That's really the only distinction that matters.

HH

[skevinp]

But of interest to me is how the sub-captions just happened to add up that way.

Recap of World Class Finals

Although Bluecoats were 3rd in Brass and that's how the corps placed overall, the brass line was 5th in Musicianship and 2nd in Technique. That sure isn't slotting. And although Carolina Crown was 4th in Brass and 4th overall, the line was 2nd in Musicianship and 5th in Technique. Ditto.

The Cadets were 4th in both sub-captions, but the numbers added up that they were 5th in Brass and 5th overall. Blue Stars, Boston Crusaders, Madison Scouts and Blue Knights all had different placements between their two Brass sub-captions.

Although it might look suspicious to some, a peek at the sub-captions shows there was clearly some serious considerations going on in the mind of the Brass judge, Steve Calhoun.

I don't know. You do see substantial variances in sub-captions in a number of places, and yet there never seem to be any ties in the captions even though they are tightly packed in some spots. That leads me to believe that they are very aware of what the total caption number will be and may work backwards to generate or adjust the sub-captions to some extent. If they just scored the sub-captions and let the chips fall where they may, working with numbers that tight, some of them would turn out to be ties.

I'm not passing judgement on whether working backwards from the score itself is right or wrong. I'm just saying this is a reason that I don't think variations in sub-captions are sufficient to disprove that slotting is a factor.

[Michael Boo]

Regarding the idea of ordinals, I have a big problem with that, having covered the world of figure skating for many years. There are too many instances where the ordinals called the shots, even though the scores within had the groups placing higher. I specially remember a performance by Soviet skater who went from first or second to right off the podium at the World Championships after each skater that followed him, even though their scores weren't as good as his. This happened due to how the ordinals worked out.

The following is from the text of my out-of-print book, "The Story of Figure Skating."

In figure skating, higher points are generally better than lower points, but only if they translate to lower ordinals. What ordinals mean is: The score a judge gives each skater is important only when compared to the scores the same judge gave to other skaters.

Huh?

Let’s say that Judge A gives Julie a Technical Merit mark of 4.4 and a Presentation (artistic) mark of 4.8 in the long program (Free Skate). Then Julie gets an 9.2 from Judge A. That doesn’t sound very high, since the maximum possible score in amateur competition is 6.0 in Technical Merit and 6.0 in Presentation, for a total possible score of 12.0. (This score would represent perfection in the eyes of that judge.) But even though Julie’s seems to be a fairly low score, if it’s the highest score that Judge A gives to all the skaters in the particular competition, then Julie wins Judge A’s “card” with an ordinal of 1.

The skater with the lowest total number of ordinals wins, so an ordinal of 1 is better than an ordinal of 2 or 3 from any judge.

Continuing in the same contest, even if Judge B gives Julie a higher combined score of 10.2, putting Julie in first with an ordinal of 1, that judge’s 10.2 is worth no more than Judge A’s score of 9.2.

Fans may express disapproval at Judge A’s lower score and approval of Judge B’s higher score, but since both judges have Julie on top, the point difference is meaningless.

Now, if Judge C gave Julie a combined score of 9.8, but gave Rebecca a 9.6 and Marianne a 10.0, this judge’s ordinal for Julie would be 2. Julie may not realize it when she sees her scores go up, but Judge A’s low score of 9.2 (and ordinal of 1) is more in her interest than Judge C’s 10.0 (and ordinal of 2).

Until all the scores are in, no one really knows what the individual scores mean until one sees how each score relates to every other score each judge gave.

Here’s a simple chart to explain the above scenario. Keep in mind this is just for the purpose of explaining the concept. There are typically seven judges at many events and nine judges at the really major events, (the large number reducing the risk of a tie), and usually far more than three skaters at even the smallest competitions.

Judge A Judge B Judge C

Julie 9.2 10.2 9.8

Rebecca 9.0 9.4 9.6

Marianne 8.8 9.8 10.0

Here’s the above chart translated into ordinals.

Judge A Judge B Judge C

Julie 1 1 2

Rebecca 2 3 3

Marianne 3 2 1

If two of the skaters had tied score-wise, the lower ordinal would have gone to the one with the higher Presentation mark. (It used to be that the Technical Merit marks broke a tie, but the ISU decided that too much emphasis was being placed on jumps. This is how the tie between Oksana Baiul and Nancy Kerrigan was broken—in Baiul’s favor—at the 1994 Winter Olympics.)

It so happens that in this case, if you were to add up all the scores for each skater, the total scores would reflect their placement. But in actual competition, when seven or nine judges are used, there’s more possibility that by adding up the scores, you would see lower (worse) scores end up with lower (better) ordinals, which is why the point is made that the raw scores are meaningless.

However, that’s not all there is to ordinals. The skater with a majority of first place ordinals will win that segment of the competition. The skater with a majority of second place ordinals with be in second, etc. Julie has a majority of first place ordinals, so she’s in first place. Rebecca has a majority of third place ordinals, so she’s in third place. That puts Marianne in second place.

Here’s where it gets weird. Ordinals have sometimes knocked well-placing skaters out of medal contention; not because they skated poorly or because others skated much better, but because the dynamics of the ordinals—the way they related to each other on all judges’ sheets—were such that the ordinals of skaters performing later (and not as well) managed to throw everything out of whack.

Translation: Sometimes the ordinals of a skater performing at the end of a contest will cause a skater who was on earlier to slip in the rankings, even though a side-by-side comparison of the ordinals of those two skaters (apart from all the other skaters) will show that the earlier skater had lower (better) ordinals. But, it’s not just how the ordinals of two skaters compare, it’s how the ordinals of all skaters together compare.

Sometimes, a skater will see their position drop two placements after the announcement of just one other score, a score whose ordinals were not as good as what they had been awarded.

It’s possible that a single new score can turn everything upside-down. It’s all in how the ordinals stack up, and not just near the end of the competition, but at the very, very end.

This is why computers are used to figure out the ordinal rankings. In the days before computers and calculators, this was all done by hand. Imagine how time consuming that must have been, since each new score could juggle the order, and every ordinal from every judge had to be factored in.

[MikeN]

Boo - interesting read!

One major difference that I would propose, though, would be that the ordinals would be given without benefit of scores - just the ordinal.

Harkening back to our earlier discussion about showing captions vs performance order and where to draw the line, I tallied it up for the three Championship shows - 60% of the captions given were exactly on their final placement. Expand that to one up or one down (meaning the corps they went on right after or right before) and the number jumps up to 90%. Which suggests, to me, that the judges right now are more or less looking at the corps only in context of the one that went on before. A natural thing to do, I would submit, but in terms of total show I think it grants a natural advantage to later-performing corps.

Maybe removing the "numbers management" game would free them up to take a bigger-picture approach?

Or not... We're all talking off the tops of our heads here. (Well, those of us who haven't written books on it...)

Mike