What is Watts per Kilo
First I think a baseline knowledge of what Wattage is would be a good starting point. Watts or wattage is a measurement of Power, or work divided by time. Essentially a cycling power meter uses torque and cadence to measure a rider’s rate of work output. Work is measured in Kilojoules, where a joule is equivalent to moving one Newton over the distance of one meter. A Newton is the measure of force equating to accelerating one Kilogram at the rate of 1 meter per second squared. Most simply watts is how one measures how fast they can complete a unit of work(moving yourself and your bike forward). In practice cyclists use power meters to measure their output on the bike, which is used for training and fitness tracking. Watts is a unconfounding way to track how fit or fast someone is. As cycling also goes uphill, it is very common to model climbing fitness by a power to weight ratio, Wattage per Kilogram of body mass. While watts per kilo is a good model, it is not a perfect predictor of climbing speed. Unless we are cycling up perfect walls, there are other factors we fight against, rolling resistance and air drag mostly. Both of these are disproportionately related to watts rather than mass, meaning a heavier rider who rides at 5 w/kg will climb faster than a lighter rider riding at 5 w/kg. Complex formulas and calculations can be done to equate climbing speed to weight and absolute power. A diagram shows this below.
Credit to @fabio.lanzone.coach on IG for the chart. On the left shows rider weight, the top bar shows the percentage grade of the climb. Each cell shows the associated watts per kilo required for a cyclist of corresponding mass on corresponding climb to keep up with a 60 kilogram rider at 5 watts per kilo. As you can see, riders of heavier weight do less w/kg to climb, particularly on less steep climbs. So what is the best predictor of climbing speed? What ever formulas or calculators this guy uses? Probably. Is w/kg the next best, simpler estimation? I don’t think so. I believe Compound Score is the better metric to use.
What is Compound Score
Compound Score is Watts squared over weight. Essentially a Power to Weight ratio that favors power. This metric is still quite simple and I intend to show that it is a better predictor of athletic ability than pure watts or watts per kilo. First I will show a study that essentially says so. In a study titled “The Compound Score in elite road cycling” done by Peter Leo and colleges, they use maximum 5 minute absolute power(W), relative power(W/kg), and compound score(W^2/kg) to correlate to podiums and wins is one day U23 races. In this study, they found that absolute power and compound score actually had much better predictive success than relative power. Compound score slightly edged out absolute power, with an R-value of .54 compared to .52, which both outweigh relative power at .11. Below are the plots for all three sets of data.
Overall they conclude that cycling is a fight against two resistances, gravity and drag, and that a metric which takes both into account is likely the more ideal model. They also go on to look at what compound score predicts success in European U23 races. They also conclude that further research into power profiles and fatigue resistance would be required to better predict performance. The article is certainly worth a read for anyone interested.
Now anecdotally. Recently I have done two rides with a training partner, where we went up our local rolling climb in different circumstances each time. First about the segment. It is a rolling climb 1.3 mile segment averaging 2.5% with a few pitches above 10% and a few downhills. I think this is a segment that is representative of a lot of American bike racing finishes, punchy but far from a long sustained European style climb. I have two examples in which we rode it, one where I was leading and breaking wind, and one where my friend was. In the instance where I was pacing, we went up the segment in 3:50, with my average power being 360, whilst my friend rode at 384. I weigh 60 kilos and he weighs 78, giving w/kg scores at 6 and 4.9, and compound scores of 2160, and 1890. In the other occasion where he paced, we went 4:00, with averages of 330 and 383, giving 5.5, 4.9, and 1815, and 1890. I suppose my point in either case is that despite confounding variables of drafting and pacing, our compound scores are always the closest metric when we climb at the same speed. As one extra segment, we did a more steady climb averaging 5.5%, with an average speed of 17 mph, where I averaged 366 and he averaged 425, corresponding to 6.1 and 5.4 w/kg and 2232 and 2316 w^2/kg. To me its clear that compound score is the better model to use in real world riding, particularly in a racing environment that lacks sustained hard climbing.
Takeaways
The study I referenced concludes with “To our knowledge, the compound score is able to measure performance characteristics for U23 one day racing success”. I guess in my long explanation about rides I’ve done recently I can justify this to myself that compound score seems to be a more accurate predictor than w/kg. What does this mean? For me this means I should probably gain weight. Compound score favors watts more than weight, therefor it would be advantageous to chase watts with weight gain, particularly for lighter riders trying to ride not in the mountains. Seemly obvious but I feel this metric is one that doesn’t get enough attention, and is a specific way to chase power, along side weight to track progress.