If more than 50 percent of games go longer than 16 turns then “13 to 16” can’t be the average number. For a number of turns to be the average, (exactly) half the games should take more turns than it and (exactly) half the games should take fewer. Without seeing the actual data, what you described makes it sound like 17 turns is average, or maybe somewhere between 16-17.
I believe you’ve confused median and mean.
?
The median is the centre-most value if you were to create an ordered list of all values. The mean (colloquially the “average”) is the sum of all values divided by the number of values.
Do you have something meaningful to add to the conversation, or are you just being purposefully obtuse?
You wrote “For a number of turns to be the average, (exactly) half the games should take more turns than it and (exactly) half the games should take fewer” and this is false. That property is true of the median.
For example, if your data set is [0,0,0,0,10], the mean is 10/5 = 2. We see that 80% of the data points are below the mean of 2. Half of your data will always be below the median by definition, but you cannot make any such assumption about the mean.
3 Likes
Fair enough. I was assuming roughly a bell curve to the data, although he stated that it is slightly top-heavy. This is why the saying “there are only three kinds of lies: lies, damn lies, and statistics” rings true.
I wrote 13-16 because the Slaghund data breaks it down by factions. You should check it out.
I guess I could have written, “The turn data is skewed slightly to the right, which implies that within the respective factions more than 50% of the games go longer than 13-16 turns”, but didn’t see the point of using such precise wording. Sorry for the confusion.