Judging/Scoring Tutorial?

[ShortAndFast]

It really does come down to the individual judges writing down numbers, and then the math of how those numbers are combined to get the total score. Generally a group would prefer to be strong in GE as it's weighted more heavily in the total score. Since this was Bluecoats' strongest caption last year and they were competitive in most of the others, the math worked out in their favor.

If you look at the scores a single judge handed out on the recap, the spread tells you something about how comparable s/he thought the groups were in that caption. A loose guide might be:

0.1: groups are very similar, judge had to make a tough call tonight

0.5: there are meaningful differences between the groups

1.0: the groups are not comparable

[DawsonBurnes]

I'm just confused on how judges quantify a performance. Is one note crack a loss of half a point, or something like that? And are they judged like AP tests (points added to the score based on their achievement), or do they start at the top score for the caption and take away points? 

[mingusmonk]

It is a Build-Up system. Start at 0 and achieve your way up.

[barigirl78]

Here are some actual sheets.  Maybe that will help.

http://www.dci.org/ViewArticle.dbml?ATCLID=209937964

[Cappybara]

It might be easier to understand that these scores aren't made in a vacuum. From my understanding, the judges score a corps a certain way and then all subsequent corps are scored in comparison to each of the preceding corps that have already performed (not completely, but that's where you will see the difference between an 88 and 87)

It's why you often don't see a corps get a 100 in a caption if they aren't the last corps on the field and a reason why many people would like to see a randomized order of performances in order to get more variability in scoring (and, of course, to moan and whine about it later)

Edited June 2, 2017 by Cappybara

[Jurassic Lancer]

18 minutes ago, Cappybara said:

It might be easier to understand that these scores aren't made in a vacuum. From my understanding, the judges score a corps a certain way and then all subsequent corps are scored in comparison to each of the preceding corps that have already performed (not completely, but that's where you will see the difference between an 88 and 87)

It's why you often don't see a corps get a 100 in a caption if they aren't the last corps on the field and a reason why many people would like to see a randomized order of performances in order to get more variability in scoring (and, of course, to moan and whine about it later)

And not only in comparison to the other corps on a given night, but in a recent Marching Roundtable podcast, the featured judges (I can't remember who they were) mentioned that to max out a caption score, the performance has to be the best they have ever seen in that caption.

[Lance]

i'm a teacher, and have found that even the best-designed rubrics do very little to eliminate subjectivity.

the rubrics dci uses are very poorly designed.

scoring in dci is a big joke. i've enjoyed more shows every year since i quit caring about scores and placements. i LOVE reading the threads about all of it on here, though. 

[mjoakes]

30 minutes ago, Cappybara said:

It might be easier to understand that these scores aren't made in a vacuum. From my understanding, the judges score a corps a certain way and then all subsequent corps are scored in comparison to each of the preceding corps that have already performed (not completely, but that's where you will see the difference between an 88 and 87)

It's why you often don't see a corps get a 100 in a caption if they aren't the last corps on the field and a reason why many people would like to see a randomized order of performances in order to get more variability in scoring (and, of course, to moan and whine about it later)

So would a number be jotted down in an official way for each corps when the performance is over? That matters, I think, if scoring is anything like evaluating a set of case studies. My early scores are always a bit tougher than my later scores. I even go back and update a few - to improve consistency - after I've finished the lot of them.

[dcsnare93]

The scoring system we have is basically a criteria reference one. There's descriptive words on the sheet that define their achievement in that caption, and currently 5 scoring ranges. So the judge determines based on that criteria where the corps falls in a scoring range. When the judge gets ready to assign a score, they'll review that criteria and their tote sheet. They'll have an idea of where the corps falls, and where their caption fits. From there, the judge will start comparing them with other corps performing in that event. It's both objective and quite subjective.

While us fans love to compare caption scores between corps, the more meaningful comparison is how far a corps is from the next scoring range.

Edited June 2, 2017 by dcsnare93

[ShortAndFast]

11 minutes ago, mjoakes said:

So would a number be jotted down in an official way for each corps when the performance is over? That matters, I think, if scoring is anything like evaluating a set of case studies. My early scores are always a bit tougher than my later scores. I even go back and update a few - to improve consistency - after I've finished the lot of them.

In DCI, this is called "numbers management". If you have 5 groups still to perform, you need to assign a number to the group you just heard that allows the subsequent groups to score higher if they are better. An easy way to see this at work is to look at the 2010 Championship Week percussion recaps, when Phantom won drums from 6th place. 

I have read some reports that these days at big shows, judges can adjust scores at the end of a block of corps, but I don't know for a fact that it's the case. Either way, managing numbers for a show like DCI Prelims with 40-ish groups performing is a daunting task for the judge.