Opinion on why you can't compare scores with out full panels

[JKT90]

I see posts talking about how did these guys jump those guys tonight? etc...

Here is the best explanation I have heard.

If your corps is weak in one caption and stronger in another and at the show the stronger caption is not judged but the weaker one is, then your score will reflect an artificially low score. Conversely, if the stronger caption is judged but the weaker one is not, the score will reflect an artificially high score.

So if a corps on one night receives a 68 and the next night they get a 65, this is what we're looking at...if going up against another corps and their strong caption is judged and your strong caption is not judged then you get flip flops in placements...Until all corps are judged by full panels, with every caption being judged we can't get a good read...Now, heading into July every show should have full panels and we'll start to see a clearer picture.

I have also had this explained to me...scores don't matter, but the gaps do.

1 to 2 tenths - corps are basically the same - it's any given night scenario

3 to 5 tenths - some differences, but corps are similar

6 to 8 tenths - significant differences in design or achievement

9 tenths and above - the corps are not in the same ball park

Given a 5 panel judge set up - if percussion is good and guard is weak and the night before there was a percussion judge but no one to judge guard, scores jump up, if the next night there is no percussion judge but guard is judged, scores go down...so we don't look at the score, we look at the gap (the tenths in points in each caption) to get a read. We may have gotten jumped, but in every caption we are 1 to 3 tenths difference from corps x (any given night type thing), but 2 points difference in guard (not even in the same ball park), that easily explains a few point swing in scores night over night.

Explained to me this way, makes perfect sense.

It's not a conspiracy, it's the make up of the judge panel on any given night.

Edited June 29, 2015 by JKT90

[geluf]

The early-season panels also heavily favor the design captions, not the performance captions (note that, without fail, both GE captions are judged at every show), which is done on purpose in order to help the designers know what is working and what isn't, early-season. The idea is that *everyone* is dirty in the performance captions early on so it's less helpful to have a judge make broad statements amount stuff you already know, as opposed to later when the nitty-gritty can be addressed.

[Herk]

Add to all that the almost certainty that, no matter what captions are judged from show to show, there will be different individuals judging those captions. So the basic rule still applies...even though everyone ignores it: You cannot compare scores from different shows.

It's fun to make comparisons between different corps, performing at the same or different shows, with different judges. That's why people do it. But most people know that none of it really matters. The true read will come in San Antonio.

[RedJazz1900]

The power of the tenth is a good reminder for a lot of people. I'm a high school band director and every year at our state bandmasters convention they have judging sessions for marching band.

.1 to .2 apart - given the night, the scores could flip flop. This is essentially a "tie"

.3 to .5 apart - small differences in performance, but caption is still comparable.

.6 to .8 apart - starting to see clear differences between the two groups in the caption. At this point the scores are different because of performance and DESIGN.

.9 or more - different ballpark/box.

Now, I have always used these rules when judging specific captions. I think this rule applies better to captions versus the overall score.

I think this is a good reminder for us all as we look at recaps.

[geluf]

Add to all that the almost certainty that, no matter what captions are judged from show to show, there will be different individuals judging those captions. So the basic rule still applies...even though everyone ignores it: You cannot compare scores from different shows.

This is the most important thing.

Some judges keep scores on the low side. Some err on the side of too-high vs. too-low. The raw number is meaningless without context. Spreads matter, and they are all that matters, and you can't have a spread without the same panel judging the same corps on the same night.

Sure, some gaps are wide enough that you can make a general ballpark call, but everyone performing in the same "group" you just can't.

This is also where the whole "West Coats Inflation!" thing got going. It's less of a thing now, but back in the early 2000s the west coast shows had fewer corps in the same tier competing against each other, so the numbers were wonky because you had to have the right spreads. It always settled down after San Antonio (and it wasn't contained to the West Coast. It happened at any show where only one or two top groups were competing with a lot more lower-tier corps...people just fixated on West Coats because it featured the perennial Love-To-Hate Blue Devils.)

Now BK has regained it's edge and Phantom is a regular out there early season so it's all but vanished. But shows out West used to look like:

BD/SCV

Mandarins

PC

...sometimes Seattle?

That's

Numbers are jedi mind tricks. Watch the spreads.

Edited June 29, 2015 by geluf

[jwillis35]

I think anything can be compared. The question is, "how do you interpret the data?" Performance captions are better left to head-to-head meetings. But if a corps is in the 13s, 14s, 15s, etc. in a certain performance area then that is usually a good indicator of what other judges might see or hear.

As others have stated, you can usually get a better idea of show potential from early-season GE and Visual scores. Take the top 5 corps. None of them have a weak brass or percussion section. All of their guards seem strong. We can assume they will be marching quite well and clean come finals, and on and on. So the differences will be show effectiveness (GE), visual design, demand, and just being consistent across the board.

If you take a look at the 6-8 slots, as of right now, or the 9-18 slots, the performance captions will play more of a role, along with design, in determining placement. That's because so many of those corps are still filthy dirty. Who will clean? Who will get their brass and percussion to rise to the highest level possible for them?

So, to me, design tends to determine more with the top groups since they all perform very well (with shades of grey in there). The more fun battles to watch will take place from 6 - 18 and beyond.

For straight-up GE, I still believe Bluecoats have an advantage come August.

[MikeD]

The power of the tenth is a good reminder for a lot of people.

No, I thnk it is the Power of 10, not tenth.

[Cappybara]

I see posts talking about how did these guys jump those guys tonight? etc...

Here is the best explanation I have heard.

If your corps is weak in one caption and stronger in another and at the show the stronger caption is not judged but the weaker one is, then your score will reflect an artificially low score. Conversely, if the stronger caption is judged but the weaker one is not, the score will reflect an artificially high score.

So if a corps on one night receives a 68 and the next night they get a 65, this is what we're looking at...if going up against another corps and their strong caption is judged and your strong caption is not judged then you get flip flops in placements...Until all corps are judged by full panels, with every caption being judged we can't get a good read...Now, heading into July every show should have full panels and we'll start to see a clearer picture.

I have also had this explained to me...scores don't matter, but the gaps do.

1 to 2 tenths - corps are basically the same - it's any given night scenario

3 to 5 tenths - some differences, but corps are similar

6 to 8 tenths - significant differences in design or achievement

9 tenths and above - the corps are not in the same ball park

Given a 5 panel judge set up - if percussion is good and guard is weak and the night before there was a percussion judge but no one to judge guard, scores jump up, if the next night there is no percussion judge but guard is judged, scores go down...so we don't look at the score, we look at the gap (the tenths in points in each caption) to get a read. We may have gotten jumped, but in every caption we are 1 to 3 tenths difference from corps x (any given night type thing), but 2 points difference in guard (not even in the same ball park), that easily explains a few point swing in scores night over night.

Explained to me this way, makes perfect sense.

It's not a conspiracy, it's the make up of the judge panel on any given night.

Someone who understands!! Yay :) a lot of people posting on here make some good points too.

[jonwoody]

Everyone makes great points and I understand the differences and reasons why so many have different opinions. I still feel the only way for anyone to know for sure how good or bad a corps is doing is to judge same corps from same judges on same nights. To me it's the only way to get a true and fair accounting and only way to avoid endless conflicts as to who is better. Non-full panels give a great idea as far as the direction a corps is going. But nothing beats a full panel to give you very true reading of what is going on.

[soccerguy315]

Everyone makes great points and I understand the differences and reasons why so many have different opinions. I still feel the only way for anyone to know for sure how good or bad a corps is doing is to judge same corps from same judges on same nights. To me it's the only way to get a true and fair accounting and only way to avoid endless conflicts as to who is better. Non-full panels give a great idea as far as the direction a corps is going. But nothing beats a full panel to give you very true reading of what is going on.

that's why regionals are a big deal!

but we need all the other shows, or else there would be no tour