Score observation...

[Scerpella]

This post again begs the question, have we reached another tipping point?

Has DCI's direction finally begun to shoot itself in the butt? This is the 4th thread in a week essentially saying the same thing. Or is it just the parting shots of yet another subsection of formerly rabid loyal (older) fans as they decide the activity has now become incomprehensible to them?

I'd wager its the latter, and bound to happen under DCI's current structure. DCI appears to not be concerned as they regularly part with fans now alienated with the new direction. (Look for electronics changes to drive away even more next year)

It appears the people who control this activity are more concerned with personal financial stability than anything else and are making moves that serve that rather than the activity. Tap into the band market and these folks will get the financial rewards they believe is coming to them. Who cares about current fans and the money they spend when there is a much larger and comparatively richer target base a few steps away in the world of school bands which are of course financed through public funds?

Drum Corps has been the bas tard stepchild of music long enough. Time to get paid!

[cantgetenoughcorps]

I would tend to agree about alot of years, but I do think recently it has changed. In 2006, I seem to recall 8 or 9 corps switching places between semis and finals. And while I wonder if it wasn't just done because people were always complaining that it is never done, it did make for an interesting night. Although I have to admit, I think the judges were closer to being right on semis.

Also, the last 2 seasons it seems like the judges have been mixing it up a little more.

[ShortAndFast]

The 2010 Cavaliers come out with their show unfinished...their closer isn't on the field, and they aren't even playing it standing still. And because of the difficulty of the show, the show is quite dirty.

Meanwhile, the Blue Knights come out on fire. This is their best show in 20 years, it's completed, and because it isn't as complicated as the Cavaliers' show is, it is being marched quite cleanly. The brass is great, the percussion is great, the guard is great....they are most likely shooting for a top five, even top three placement when it is all said and done.

In this scenario, how much money would you put on the Blue Knights beating the Cavaliers? I know I wouldn't stake more than a few beers on that result, because you and I both know that the Cavaliers are more than likely going to win this show.

June 16, 2007: the Bluecoats beat the Cadets at their first show. July 2, 2007: the Bluecoats beat the Cavaliers and Blue Devils. August 11, 2007: Top 3 is BD, Cadets, Cavaliers. Bluecoats finish 7th.

Is this not exactly the situation you're describing? Bluecoats came out with a strong product and the eventual top corps needed a fair amount of tweaking to get to their shows into their final form. Seems like the judges were perfectly willing to reward Bluecoats when they felt they had a better show on the field than the big guys.

[BlueR36]

Best topic I've read through in a long time.

It's sad when you can look through the schedule when it's released in Winter and just "know" who's going to win at which shows throughout the entire season. I would be that if you took just your own corps (Bluecoats for me) and predicted the placements for each show through the season, just about all of us would get more than half of them right in March. And if we updated our July and August predictions after June's shows, we'd likely all get at least 75 percent right.

That doesn't happen in other "Major Leagues" as DCI is portraying itself as ... there are always good teams that are suddenly bad and vice versa every year in football, baseball, basketball and hockey ... but never in drum corps.

There was one year a couple seasons ago that the 12th-place corps never beat anyone in front of them and never lost to anyone behind time. They were 12th the season before and stayed in 12th all year. Yawn.

It is a little disheartening, sometimes. And boring. I actually have some friends and relatives who used to follow drum corps back before and shortly after I marched who don't care anymore b/c "it's always the same corps every year."

[kdaddy]

Not that many people are into statistics here, but I did a simple linear regression analysis for all the Div 1 corps in 2006, 2007.

The typical result is that corps improve by about 0.35 pts per day regardless. "Day of season" accounts for 95% of the variability in score. All other factors (performance, judging panel, weather, etc) account for 5%

The day to day fluctuations are typically less than a point. So, if two corps are separated by 1.0 or more, there is almost no chance that their relative placements will change.

Can you explain the "performance" and "judging panel" variables a little more?

I'm not at all surprised by your results, and I'd imagine that you'd get similar results if you ran the regression on 1990 data.

I'd like to see some sort of comparison of judges' treatment of particular corps, maybe some sort of paired samples test.

[BlueR36]

June 16, 2007: the Bluecoats beat the Cadets at their first show. July 2, 2007: the Bluecoats beat the Cavaliers and Blue Devils. August 11, 2007: Top 3 is BD, Cadets, Cavaliers. Bluecoats finish 7th.

Is this not exactly the situation you're describing? Bluecoats came out with a strong product and the eventual top corps needed a fair amount of tweaking to get to their shows into their final form. Seems like the judges were perfectly willing to reward Bluecoats when they felt they had a better show on the field than the big guys.

I think, however, that is just about the only example there is other Bluecoats being in fourth, falling to sixth, and jumping back up to fourth during finals week 2006; and other than Phantom jumping up to tie for first in 1996.

Edited July 25, 2008 by BlueR36

[kdaddy]

It's interesting that this example is from a year when there was a 6 judge panel, as opposed to the current 9 or 11. More judges should translate into less variance in the final score just based on the law of averages.

That's a great point.

Though we're likely to see the smaller panels from now on, given travel costs and economic situation of DCI, among others. What about some analysis of data from the small panel contests we saw earlier this season? :)

Also, though I realize that judging is performed with the current system in mind, what about looking at scoring for the past couple years with different weightings of the captions? That'd be an interesting exercise, though based on some strong assumptions.

[GMichael1230]

(Whispered Aside: Of course....if you're judging design instead of performance, I can understand why scores are so....static.)

Bingo !! You get a gold star and that pretty much sums it up as of late.

Geoffrey

[kdaddy]

That's a great point.

Though we're likely to see the smaller panels from now on, given travel costs and economic situation of DCI, among others. What about some analysis of data from the small panel contests we saw earlier this season? :)

Also, though I realize that judging is performed with the current system in mind, what about looking at scoring for the past couple years with different weightings of the captions? That'd be an interesting exercise, though based on some strong assumptions.

And along those lines, though a small deviation from the topic, what was the makeup of those smaller judging crews of old?

Here's guessing that judging crews become a lot more regional for much of the season as DCI tries to cut down on travel costs. I say the crews will be either smaller, regional, or both. Were the judging crews back then national or regional? What effect might this have on scoring? Would it be a bad thing? It could cut down on consistency from night to night, as a lot of nights would be "first reads" for the crew. Or it could create even more consistency and slotting, as the fresh regional judges may rely more on the previous night's scores found on the internet...

[Top Cat]

Can you explain the "performance" and "judging panel" variables a little more?

I'm not at all surprised by your results, and I'd imagine that you'd get similar results if you ran the regression on 1990 data.

I'd like to see some sort of comparison of judges' treatment of particular corps, maybe some sort of paired samples test.

Performance variability relates the fact that we are dealing with live (basically) amature performers. No two shows are performed exactly alike. Depending on the specific variances (could be from any individual, any small group of performers, sections, etc.) and how perceptable those variances are, the score should change.

Judging panel variation comes in that in no two shows will all the judges be the same and most likely if they werte, they would be on different captions. So the experience level, coupled with the inherant differences in each persons scoring skills, you will not get the exact same reaction.