Pedantic
05-26-2003, 02:32 PM
SO Private League Week 8 Stats. You might notice the different format...perhaps... :p
I figure I shall explain some of the new categories.
Median is the middle figure or average of the middle figures in a descending list of all of the figures.
Mean is the arithmetic average. You might notice the total and mean are very close or the same as some of the totals. That's because I use average to compute ERA and WHIP and the BA is very nearly the average of the weekly figures, except I take the total hits over the total ABs for a more accurate figure. It works a point or so in favor for many of you. ;)
Prediction is a prediction of figures for the next week, based on a linear regression of the previous weeks' data points. It's not too accurate, but it's fun. :D
Consistency is the total for that category divided by the standard deviation, except for ERA and WHIP. I multiplied the total by the deviation for those categories. I added total into this category so that there would be somewhat of a "bonus" for teams that do well. Standard deviation is how much the total data set deviates from the mean. There aren't units for this category. You can only measure your team's own performance based on the other teams' scores. In the first eight categories, higher is better, but in keeping with the principle of ERA and WHIP, lower is better for those two categories.
Believe it or not, this new format is actually way easier for me. Now it takes 10 minutes instead of two hours. ;) :D Sorry, no more individual player stats. Those were a royal pain in the ass. :) Oh, yes, I'm sure you'd like to see the actual stats. Go here (http://phasersonkill.tripod.com/calculations.htm). Oh, yes, I'd better disable my large sig pic so there's actually some bandwidth left. ;)
I figure I shall explain some of the new categories.
Median is the middle figure or average of the middle figures in a descending list of all of the figures.
Mean is the arithmetic average. You might notice the total and mean are very close or the same as some of the totals. That's because I use average to compute ERA and WHIP and the BA is very nearly the average of the weekly figures, except I take the total hits over the total ABs for a more accurate figure. It works a point or so in favor for many of you. ;)
Prediction is a prediction of figures for the next week, based on a linear regression of the previous weeks' data points. It's not too accurate, but it's fun. :D
Consistency is the total for that category divided by the standard deviation, except for ERA and WHIP. I multiplied the total by the deviation for those categories. I added total into this category so that there would be somewhat of a "bonus" for teams that do well. Standard deviation is how much the total data set deviates from the mean. There aren't units for this category. You can only measure your team's own performance based on the other teams' scores. In the first eight categories, higher is better, but in keeping with the principle of ERA and WHIP, lower is better for those two categories.
Believe it or not, this new format is actually way easier for me. Now it takes 10 minutes instead of two hours. ;) :D Sorry, no more individual player stats. Those were a royal pain in the ass. :) Oh, yes, I'm sure you'd like to see the actual stats. Go here (http://phasersonkill.tripod.com/calculations.htm). Oh, yes, I'd better disable my large sig pic so there's actually some bandwidth left. ;)