Rankings (again)

[graeme]

Since you posted the graph, I'll try to explain what's wrong.

The system is based on the idea of a "Normal" distribution - for every good run, there's an equivalent bad run. In such a distribution 1310 point would mean something: a run in the top 7% of all runs. Curiously, Colin probably is about that good! In a kind of antiLakeWobegon sort of a way, there's only Scotia in the top 5% (above 7950) !!!!

Anyway, you can see from the graph that brown and green have a wider spread among slow runners than fast. This is what usually happens, the winner is closer to the average time than the last person*

Blue is unusual, in having a bunch of people at the slow end, meaning that the last* person is closer to average than the winner. (Plus old men are overrated, or so I'm told)


* There's a random rule BOF don't tell you about which excludes *very* slow runners - the bottom three on blue here. It's meant to compensate for the fact the whole system is based on dodgy stats. Sometimes it does, but being random, sometimes it makes it worse.