On Mon, 8 Apr 2002, Adam Spiers wrote: > > Philippe 'BooK' Bruhat ([EMAIL PROTECTED]) wrote: > > > And yes, steps are more accurate, and more correct to the true evolution > > of the golfers scores. :) > > OK, OK :-) How's this? > > http://adamspiers.org/upload/tpr02-step.png > > I still prefer this, personally (updated for final results): > > http://adamspiers.org/upload/tpr02.png >
There are two reasons why I prefer the steps. The first is the obvious one: you can see when you overtook someone. But the second is perhaps more subtle: with the linearised version, it looks as if the _slopes_ are significant, but the real insights are represented by large _drops_, not steep slopes, and this is clear on the step chart. (E.g., I was stuck on 73 for 4 days before finally seeing how to reach 63 on Sunday afternoon. The slopes make it look as if I was improving very gradually for half a week, not dramatically at the end!) -- Stephen Turner, Cambridge, UK http://homepage.ntlworld.com/adelie/stephen/ "This is Henman's 8th Wimbledon, and he's only lost 7 matches." BBC, 2/Jul/01
