Illustration - Removing Errors and Bias
In traditional scoring, judges use their own point scales, which can create errors and unfair influence on competition outcomes. NSS fixes this by:
-
Standardizing each judge’s scores so they’re all on the same scale
-
Normalizing ranking differences so no single judge’s wider or narrower scoring range can dominate results
-
Reducing error and bias so results better reflect the consensus ranking of competitors
Instead of relying solely on raw points, NSS focuses on relative performance across judges — which means if everyone agrees who should win, NSS preserves that outcome even if judges score differently.
As you can see below in the “BEFORE NSS” graphic, Judge 3 has effectively controlled the outcome of the contest by using a large point spread (we call this “point spread bias”).
Here’s how NSS fixes the problem. Each judge’s scores are ranked from highest to lowest. A judge’s first-place contestant receives 100 points, second place receives 99, third place receives 98, and so on. These become the judge’s “normalized” scores.
A judge’s choices do not change. The contestant the judge scored highest is still their first-place contestant, second is still second, and so on. NSS simply puts every judge on the same scoring range. That means every judge has equal influence on the final outcome—regardless of whether one judge normally scores within a few points and another uses a much wider range.
Now look at the “AFTER NSS” results. Each judge still has the contestants in exactly the same order they originally chose—but now every judge is using the same scoring range. Judge 3’s unusually wide point spread can no longer overpower the other two judges. Two of the three judges ranked Contestant 1 first, all three ranked Contestant 2 second, and two of the three ranked Contestant 3 third. Each judge now has an equal voice in the final outcome.
Want the Full Explanation?
For a detailed description how the NSS removes errors and bias, with examples and comparisons to traditional scoring, you can download a detailed explanation.
Share
Print 