Notes on USA Cycling Power Based Training Clinic

I attended this clinic last week in St. Louis, Missouri. The clinic was led by Hunter Allen and Andrew Coggan, two of the creators of cyclingpeaks, the predominant software package for analyzing power data. Overall, it was really enjoyable. I am an unabashed bike geek, so spending a weekend with those with similar interests was far from a hardship.

The most important thing I heard over the weekend had less to do with power based training and more to do with understanding such technology as a part of the coaching process. There is no doubt that power based training is tremendously helpful as a means to monitor what an athlete is doing. At the same time, power meters are not a part of the coaching relationship. They are simply a tool that can allow coaches to do their jobs better. One can forget that in the quest for more data and answers to data driven questions. What matters most, however, is that athletes train hard and smart. They should then get good results from their efforts. A power meter will help you get those results, but only when it is used in conjunction with a sound training program. What matters as far as athlete performance is concerned is not the means of analyzing the training per se, but the training itself. Power meters offer unprecedented opportunities to study and analyze training but that knowledge should not overlap with one's overall coaching philosophy. No matter how one analyzes training information, it is the training itself that makes athletes faster.

Now on the power training clinic itself, the two presenters did a nice job of covering the basics of power based training. I have seen triathlon race files but I really enjoyed the analysis of different types of power data. Hunter Allen showed us some data from BMX racers he collected at the USA Cycling BMX track. While I am getting to be a bit of an old man, so I can be skeptical of such things as BMX racing, those guys can put out some serious power, particularly at the start of a race. One of the files Hunter found had the highest ever recorded 5 second power per unit of body weight. This guy made 25 watts per kg for 5 seconds. Now, I don't know how much he weighs but that is unreal! I am lucky to get mine in the neighborhood of 12, so I guess I am truly out of the running for BMX glory. Oh well, too old in a number of ways apparently.

The criterium files and road race files were also cool to look at and discuss. Hunter had data from the Tour de France this past year that dramatically highlighted how different it is to ride in the pack versus be in a breakaway. It was a stage from the first week, I believe, and Jens Voigt went into a break and stayed there for hours. He worked hard while in the break. Christian Vandevelde stayed in the pack all day and didn't work much while there. The days of these two guys could not have been more different. By driving the break at work levels that were unsustainable for the long term at times, Voigt put down a 5 plus hour tempo type workout. Vandevelde's effort levels in the pack were not steady, as there were some demands, it was the Tour de France after all, but his 5 plus hour day ended up in the endurance ride type of effort. The data and analysis was a lot of fun. In fact, the data is freely available for cyclingpeaks users at http://www.cyclingpeakssoftware.com/2006tourdata.asp
so next time you're ready to think about your chances at the Tour, check out the data.

As the creators of cyclingpeaks, I learned a lot from the standpoint of how to manipulate data in the program and the justification as to why it does the things that it does. I really am convinced that its primary features, such a the norm power algorhythm and the Training Stress Score values (TSS) are sound. They might not be perfect, and no one said they were, but these are awesome tools. Since I'm all fired up to analyze some training and racing data, I'd better get me and my clients on the bike! Have fun out there.