Thursday, January 1, 2009

Even More PowerCrank Analysis

I thought I should expand on the previous data analysis and show a type of histogram of performance since using PowerCranks compared to the previous historical data. The plots below show weekly max values for 263 different weeks dating back to 2003. Some weeks were removed due to travel overseas where I was without my bike or sickness and I did not ride. In other words, only weeks with riding were considered.

In all the plots, the x-axis shows power and the y-axis shows the week number. Weeks 247 and up are when PowerCrank use began in earnest. The data associated with these weeks is shown in red. The other weeks, shown in green, are non-PC use weeks or when I used PCs for an hour or two. In July 2008 my focus was on time trial training for the Texas State Championships, and I focused on regular cranks on my time trial bike.

For durations of 3 to 15 minutes, average power is used. For the 20 and 60 minute durations, I am using normalized power. For some reason CyclingPeaks is not accurately reporting my historical 20 minute mean maximal data.

It should be noted that the data represents weekly bests. In a given week, I may have conducted multiple workouts which would have had similar results in power. So this is not a true histogram of all data. It is, however, a “weekly histogram” from which some conclusions can be made. I should also state that my focus since August has been on L3 to L5 work. Over the years I have shifted my higher intensity training sooner with each passing year to meet the demands of the early racing season. Thus, the data associated with PowerCranks will tend to shift towards the right in each curve, representative of the focus given to the intensity.

What the data suggests, with the exception of the 20 minute data, is that the data associated with PowerCranks has a somewhat similar distribution as the set of historical data. The 20 minute power plot seems to indicate I’ve been more consistent with 20 minute power, and that I’ve had more occurrences of top-level 20 minute power in the last 4-5 months than in past years. However, I’ve yet to beat any of my all-time best power numbers for any of these durations since using PowerCranks. I’ve been more consistent with the power I do have, but I have not seen an expansion of my power-duration envelope.








No comments: