Re: A question on "median average"...

posted at 4/27/2012 11:07 AM EDT

In Response to

Re: A question on "median average"...:

Nice to see the shill, who claims not to read my posts, making another attempt to troll a personal attack thread. Yes, small minded folks, there is such thing as an average of medians over subsets.

All of those things you said are mathematical words, with definitions, and can be applied to situations sure, but not to a single string of data without recatagorizing it. It sounds like you're saying that you are using the average of medians of all of our game's subsets (based on that vague undefined rhetoric you've given).

The set of all subsets of a set (say that 5 times fast) of data is known as it's power set. The number of sets in this set, can be found by raising two to the cardinality of the original set. We've played 18 games, so the number of medians you would have to average to do so is 2 to the 18th power, otherwise known as 262,144 medians. I find it unlikely you are doing this, or that it would even have any merit.

A pure statistical median is probably the best way to factor out the weight of outliers while still acknowledging the occurance of a high or low scoring game.

Our organized scores are,

0 0 1 2 2 3 3 3 4 6 6 7 9 10 11 12 12 13

With an even number of games we have to take the middle to numbers and average them, which are 4 and 6, which means our median is 5. The Sox bats aren't the problem, a team scoring more than 5 runs half the time should have at least a .500 record.