I shouldn't have said "out of a group of 1000", I should have said "out of a group of unknown size" - sorry about that.
What I mean to say is that, if you don't know the sample size, you can't draw any real conclusions.
But can we have a guess at the sample size? I'm going to have a go, and I'm purposely going to err on the side of yielding a lower sample size:
Suppose that, out of all the people who read this forum, only 100 are going to post about something unlikely happening. Further suppose that they have each played for 10 hours, i.e. 1,000 game hours altogether.
Now we have to guess how many combats have occured in those 1,000 hours. Let's say that it works out at 20 combats an hour. (That may be really far too low - sometimes I have 50 different units involved in combat PER TURN.)
Anyway, that gives a sample size of 20,000. If you had 2 occurances of a 1 in 10,000 result, you wouldn't be surprised.
In fact, the sample size would almost certainly be much larger - at least 100,000 individual combats. You WILL see some very unusual results with sample sets that large.
What I mean to say is that, if you don't know the sample size, you can't draw any real conclusions.
But can we have a guess at the sample size? I'm going to have a go, and I'm purposely going to err on the side of yielding a lower sample size:
Suppose that, out of all the people who read this forum, only 100 are going to post about something unlikely happening. Further suppose that they have each played for 10 hours, i.e. 1,000 game hours altogether.
Now we have to guess how many combats have occured in those 1,000 hours. Let's say that it works out at 20 combats an hour. (That may be really far too low - sometimes I have 50 different units involved in combat PER TURN.)
Anyway, that gives a sample size of 20,000. If you had 2 occurances of a 1 in 10,000 result, you wouldn't be surprised.
In fact, the sample size would almost certainly be much larger - at least 100,000 individual combats. You WILL see some very unusual results with sample sets that large.
Comment