The ratings have been adjusted so that they have approximately the same distribution. My theory is that, according to the law of large numbers, the whisky ratings use a normal distribution, provided that the number of ratings is sufficiently large. Therefore, in order to make the ratings comparable, I have applied conversion formulas to the ratings so that they have (approximately) the same average and standard deviation. Also, I have made sure that all the ratings have the same maximum (96 points) since the focus of most comparisons will be on the top ten or so entries. The conversion formulas are:
Thanks to these conversions, I believe that the ratings are comparable, even though taster A might define 80 points as "mediocre" and taster B might define it as "very good". In my opinion the ratings should only be used when comparing two or more whiskies and therefore it is up to the individual to decide how good "80" is. The actual averages and standard deviations can be seen here, but beware! It takes a while to calculate... |