Five years of freekick data tallied together. Considering teams all tried the Richmond way, it's pretty indisputable that there is a bias against us. Conscious or not.
View attachment 21281
It is not tough at all to see this data is skewed radically and unnaturally by something.
The second worst off team for free kicks in this period - Hawthorn - is 255 free kicks worse off than the 3rd best off team. Given there are about 135 matches per team on average in this data, it means the 3rd highest beneficiary, Fremantle, is getting about a 1.89 free kicks per game better differential than the 17th highest beneficiary. Less than half a free kick per 1/4 played in differential.
Yet you move one position up and down the ladder and compare the 2nd highest beneficiary, Collingwood, with the 18th highest beneficiary in Richmond the difference is 533 free kicks, more or less 1 whole free kick per 1/4. If you compare Richmond with the top beneficiary, Bulldogs, you are now looking at well over a whole free kick per 1/4 of football difference, about 4.5 free kicks per match.
Let's do another exercise. Let's couple pairs of teams top with bottom, 2nd top with 2nd bottom all the way through to 9th with 10th. And see how even this distribution looks. I have to say I am no mathematician so I have no idea what I am doing here.
So this is set out in the following format: 1. teams(by position in the table in the post above)/ 2. free kick differential between the two teams/3. the difference in free kick differential between each couple and the one before it.
9 & 10/ 32/
8 & 11/ 45/ 13
7 & 12/ 54/ 9
6 & 13/ 95/ 41
5 & 14/ 126/ 31
4 & 15/ 192/ 66
3 & 16/ 229/ 37
2 & 17/ 345/ 116
1 & 18/ 604/ 259
I actually hope there is someone with much better mathematical acumen than me who can do something more meaningful with these numbers. To my untrained eye it looks a very unexpected distribution, where the last 2 couples, and especially the last one, look way way out of place. I mean the difference from one couple to the next averages about 33 until you get to the final 2 couples where it suddenly leaps to 116 and 259. And if the 2 & 17 coupling(Hawks and Pies) was more in line with the other 7 couples above it, and was say 50 or 60 higher than the couple before it, then the 1-18(bulldogs and tigers coupling) would look even more foreign in this sample.
Either:
A) there is very significant bias in the umpiring, or
B) the top and bottom teams and or top 2 and bottom 2 teams are playing way more differently to each other than 14-16 teams are from each other COMBINED, or
C) both of these things are true.