Well, no matter where you cut it off it's gonna leave a mystery of "who is next that's not on the list". I just think it's useless to predict more than 6 months in the future based on 2 weeks of exp gains.
btw could you list standard deviation or variance on their exp gain? (I assume you've got their exp once a day so 14 samples)
btw could you list standard deviation or variance on their exp gain? (I assume you've got their exp once a day so 14 samples)

