The calculation of the variance by dividing the sum of squared differences by N, rather than N-1, introduces a bias in the estimation.
The ans in the variance function is divided by data.length instead of data.length - 1.
The result of variance, which affects the stddev, is biased. That leads to the calculation of the outliers of on the finalizeValidation function to be biased, potentially leading to not rewarding validators due to biased results of stddev.
Manual Review
Adjust the variance function to calculate the ans by dividing with data.length - 1 according to Bessel's correction to avoid bias in the result. Source
The contest is live. Earn rewards by submitting a finding.
This is your time to appeal against judgements on your submissions.
Appeals are being carefully reviewed by our judges.