gillan1969 said:
the data is the data...the problem would have been not having the outliers in...we can all see the outliers and see what difference they make...that's the beauty of raw data...its not been manipulated before you get it...feel free to remove them and make your own calculations...
I agree. The two biggest problems with the approach Swanson has taken are:
1)That you don't get a true power estimate, you only get a comparison in power to the winner of the time trial, which brings us to the second problem, which is
2) That the quality of the winning rider changes with every race. While one time trial might be won by Martin or Cancellara, another might be won by say, Millar or Posthuma, who would both get creamed by the first two if they raced each other. The time loss per kilometer for Froome would go significantly down for Froome without the top guys in the race.
Those two problems still won't hide the fact that Froome suddenly started being competitive in time trials in 2011, even against the very best. That's something he had never, ever done before, at least at pro-conti level and above.