Can anybody provide statistics for appearance of triplet in cycle of 37 spins

[sekuritati]

I was going to ask Bayes for some info regarding this but saw he hasnt logged in for almost a month so I am asking if anybody could provide me with statistics of the appearance of a triple repeater in 37 spins.

In other words, in 1 million or 5 million spins, always looked upon in a moving window of 37 spins, how often there are unique numbers + repeaters but not such a number that repeated 3 times.

example:
Unhit - 8 numbers
1 hit - 20 numbers
2 hits - 8 numbers
3 hits - NO TRIPLETS

Thanks in advance