RCLiveScore uses two different methologies to arrive at normalized and scaled scores based on the number of contestants and judges within a particular class:
Z-Score
Z-score normalization, also known as standardization, transforms a dataset to have a mean of 0 and a standard deviation of 1. This is achieved by subtracting the mean from each data point and then dividing by the standard deviation. The resulting z-scores indicate how many standard deviations a data point is away from the mean.
Think of each judge’s score sheet like a ruler with its own zero mark and spacing.
-
Find the judge’s average (their personal middle).
-
Measure how far each score sits above or below that middle.
-
Divide by their typical spread so big and small rulers become the same length.
After that quick math:
-
0 = exactly that judge’s average.
-
Positive numbers = above their normal.
-
Negative numbers = below.
Because every judge’s ruler now shares the same center (0) and spacing (1 unit = one typical jump), generous and harsh judges cancel out and the playing field levels automatically.
Min-Max + Winsorization
Winsorization is a statistical technique used to mitigate the impact of outliers by replacing extreme values in a dataset with less extreme values from within the same dataset. Instead of removing outliers, winsorization effectively "caps" them by setting them to the value of a specific percentile, like the 5th or 95th percentile. Min‑Max takes the toughest judge’s lowest mark and calls it 0, the most generous judge’s highest mark and calls it 1, then stretches everything else proportionally in between. Rank order stays the same; the whole set just fits neatly on one common ruler. This is similar to a grading curve on academic tests.
Which is Used When?
The system automatically applies the most rigorous normalization method possible given the number of contestants and judges within a particular class. If there are fewer than 3 judges or 6 contestants in a particular competition class, the system will use the Min-Max methodology because Z-Scoring will generate anomalous results. If any class exceeds those thresholds, the system will employ the Z-Score method, which is more accurate a those levels. In all cases, adjusted scores are then scaled according to the scale range selected by the event host for each scoring standard.
Why normalize?
Judges have different “centers” and “spreads.” One judge may hand out mostly 9s, another prefers 6s. Normalization aligns those personal scales so the athlete (or pilot, dancer, etc.) is rewarded for performance—not for drawing the “easy” judge.
How our methods differ from “drop the high & low”
Dropping outliers throws away two opinions entirely. Z‑score and min/max balancing keeps every judge’s vote but dials down the impact of someone who’s far off the group pattern, giving you fairer results while preserving more information.
Take‑away for Event Hosts
-
Fairer finals – the same performance gets the same credit no matter who’s judging.
-
No extra work – enter raw scores, LiveJudge does the rest.