How this works:

The composite number is entered, and then the judge's individual rating is entered.  If a judge rates a band lower than the composite, then the judge would get a -1 for that band.  If a judge rates band HIGHER than the composite, then the judge would get a +1 for that band.

​

Example:      Judge 1         Judge 2        Judge 3           Composite

                            3                    1                    2                          2

​

Judge 1 would have a score of -1.  Judge 2 would be +1, and Judge 3 would score 0. We believe this is the fairest way to evaluate the ratings of judges.  We then add up all differences (-1, 0, 1, etc) to see which side of 0 landed on, and divide that by the number of bands.  That is the percentage shown on the graph. 

​

Why was this done?  We've all heard band directors in the halls saying "Oh, this judge rates 3A harder, or that judge is easier on 2Cs" or something like that.  Is this true?  For the most part, no.

​

For the graphs, a minimum count of 10 bands in the conference was used, as we believe this will give us the best statistical analysis

​

Where you see a difference of less than 10% either way, we believe this to not be statistically significant.  Have we run the math to prove it is?  No.  You can be your own arbiter.  

​

There are several instances where a split panel is used.  That has not been factored into an individual judge's score, as we cannot determine which judge of the two it was.  A few contests list the judges as "x" or "not given".  Those are listed under "x" and "not given".