Thursday, December 4, 2008

Power Numbers from The Dirty Dozen

The Dirty Dozen is a bike ride/race that has been organized by Danny Chew (best known for winning the Race Across America in 1996 and 1999) since 1983. Each year, riders come out on the last Saturday of November to ride up the 13 most grueling climbs in Pittsburgh. At the start of each climb, a whistle is blown to signal the start of the climb. Points are given to the top 5 finishers up each climb and the winner at the end of the day is simply the one with the most points. Anyone who has ever been to Pittsburgh knows that there are some serious climbs there. Pittsburgh is a town where a grid system was essentially forced on top of mountains. The result of this is that there are quite a few streets where I have to scratch my head and ask, "Why the heck did they even build this road?". The steepest hill on the DD is Canton Ave., peaking at a 37% grade, which by some accounts makes it the steepest hill in the world. For most, it is too steep to drive up or down. As if that weren't bad enough, it's also cobbled.

Of the 150 riders that showed up at the start this year, probably only 20 were in contention for points, though I can guarantee that each and every rider went through their own personal struggle and at the end of the day they all had a story to tell. My goal was to do my best up each climb and make it through the day consistently. The winner of the event for the last 4 years has been my friend and ex-teammate Stephen "Steevo" Cummings. Being that Steevo is 25 pounds lighter than I am and in excellent form for the cyclocross season, I knew that to beat him I would probably need an act of God, so my strategy was to not even worry about him and pace myself appropriately up every climb as if I were doing a series of 13 time trials. This strategy worked out reasonably well, as I saw quite a few riders blow up in front of me throughout the day. I never blew up and I was able to score points on 12 of 13 hills (I pulled out of my pedal on Canton Ave.). I chose to ride a 34 x 50 compact crank with a 12 x 27 cassette, which turned out to be a wise selection as I needed every bit of that granny gear.

Here are the basic stats from the ride:
Total Time: 8 hours
Total Riding Time: 4:05
Distance: 54.7 miles
Average Speed: 13.4 mph
Average Power: 150
Normalized Power: 323
Work: 2198 kJ
TSS: 389
Average Cadence: 76 rpm
Average HR: 148 bpm
Max HR: 193 bpm

(If you are unfamiliar with any of these terms, see my blog on power terms). Note: all stats are based on riding time, time off the bike is not included)

The first thing that stands out is the huge difference between average and normalized power, which signifies a very non-steady state effort. In fact, it doesn't get any more non steady state than this. Overall, there was only about 35 minutes of climbing in the span of a 4 hour ride, but that 35 minutes was all at a sub-maximal effort. Here are the stats for the individual climbs:

#1 (Center/Guyasta): 0.78 miles, 4:16, Avg. = 481 watts, Max = 890 watts
#2 (Ravine/Sharps): 0.80 miles, 4:50, Avg. = 445 watts, Max = 677 watts
#3 (Berry Hill): 0.23 miles, 1:06, Avg. = 691 watts, Max = 1086 watts
#4 (High St./Seavy): 0.36 miles, 1:50, Avg. = 589 watts, Max = 839 watts
#5 (Logan): 0.34 miles, 2:25, Avg. = 540 watts, Max = 744 watts
#6 (Rialto): 0.10 miles, 0:40, Avg. = 877 watts, Max = 1207 watts
#7 (Suffolk/Hazelton/Burgess): 0.48 miles, 3:56, Avg = 442 watts, Max = 665 watts
#8 (Sycamore): 0.55 miles, 3:32, Avg. = 457 watts, Max = 745 watts
#9 (Canton)*: 0.05 miles, 0:37, Avg. = 597 watts, Max = 695 watts
#10 (Boustead): 0.35 miles, 2:16, Avg. = 504 watts, Max = 787 watts
#11 (Welsh Way): 0.26 miles, 1:26, Avg. = 543 watts, Max = 840 watts
#12 (Barry/Holt/Eleanor): 0.44 miles, 3:28, Avg. = 461 watts, Max = 923 watts
#13 (Flowers/Tesla): 0.88 miles, 5:07, Avg. = 390 watts, Max = 737 watts

* I pulled out of my pedal the first time up Canton, so these numbers are from the second time, when I went as easy as possible.

The first question that comes to mind for me is, "Which climb was hardest?". Of course, if you ask 10 different people this question, you will get 10 different answers. Some will rank Canton the hardest because it is so steep and cobbled. Some will rank Suffolk as the hardest because is is long and steep (and cobbled at the top). Personally, I would rank Barry/Holt/Eleanor as the hardest because it is long, steep, and #12. In the Tour de France, climbs are categorized by a combination of length, grade, road surface and where they appear in the route (e.g. the only climb of the day at half way through the race or the 4th categorized climb of the day after having ridden 200 km). I think that this is a pretty good measure, except that there is still no precise formula, which leaves room for different interpretations.

Power numbers are also a subjective way of measuring difficulty because they depend on how hard you decide to go (or could go at the time). However, if you assume that I went as hard as I could up each climb, average speed would be closely related to average grade and one could get a pretty good idea of how difficult the climbs were by averaging the speed and distance rankings. This ranking system produces #7 as the most difficult climb with #2, #12 and #13 tied for 2nd place, which I would argue is pretty close to correct (though I might not put #2 up there because the steep part is at the beginning and it ends on a lesser grade).

Still, there are some that might disagree with these rankings because it does not take into account road surfaces, technical aspects of the climbs, variations in grade or tactics. Most riders are better suited to one type of climbing than another, which is why I always cringe when an athlete tells me "I am a climber" or "I am not a climber" as if it's black and white. A 30 second climb is one step above being a sprint. On the other hand, a 5 minute climb is a much more sustained effort. If we were climbing in the Rockies, the Alps or the Pyrenees we would be facing 20-60 minute climbs which are an altogether different beast. But climbing is about more than just who has the highest power to weight ratio over the given amount of time. I know many riders with high LT or even 5 minute power to weight ratios that I regularly drop on climbs. This is because they do not have the ability to make accelerations within their efforts and recover from them.

Remember last year's Tour de France on the Aple D'Huez stage? Sastre's effort excluded, the last climb was a very non steady state effort back in the field. Evans, Sanchez, Valverde, Efimkin and the Schleck brothers were taking turns attacking each other and playing cat and mouse games. If your goal is to go up the hill as fast as possible, this is probably not the best strategy but if you want to separate yourself from the others and be first up the hill, well, that's a different matter. The hope of any of these attacks is that you can push your competition over the edge of what they can sustain and snap the elastic. Although this is arguably the goal of any attack, things are a little different when you are going uphill; when you crack there is nowhere to hide. Even if you are unconcerned with the other riders, tactics are still relevant. A steep or cobbled section of a climb can push you over the edge just as easily as an attack.

Throughout the day, I did have to accelerate a few times at the end of climbs to gain or maintain position, even though I knew I would pay the price later on. I also felt like I was going to throw up and/or hyperventilate at the top of every climb. At the end of the day I felt like I had just taken 5 years off my life... BUT... I was happy in knowing that I accomplished all of my goals. I never blew up, I did the best I could and I finished the day in 2nd place (though I had to defend 2nd place all the way to the last 10 meters against a fast and surging Chris Mayhew).


  1. Awesome! I kept meaning to ask you about your file. This is way more than I expected. Nice job to you and nice analysis.

    Did you have numbers from last year? I assume this year was much higher, given your dramatic rise in the standings.

  2. You should consider racing the Savageman 1/2 ironman out in Deep Creek Lake, MD. The steepest climb in all of triathlon. It is a fun course.