The Problem with Power

Power is hot. It’s the new it. And I’m the first to agree that it’s a great tool. One of the best. For those scratching their heads, I’m referring to bike training with power (or a power meter). It’s a great tool, as it measures the energy output that you’re putting out during your ride. I will be the first to tell you that using a powermeter in conjunction with a heart rate monitor is probably the most ideal training setup on the market today when you’re on your bike. But with all it’s pluses, it still has it’s issues.

Power is an output. Just like speed or pace, it’s a result of the effort, or stress, that you put your body through. And similar to all outputs, it’s absolute. In fact that’s one of the talking points that the uber powermeter crowd touts. A watt is a watt. I would agree. Most outputs are. A pace is a pace, a mph is a mph. BUT, and this is a big but, your body’s ability to produce that watt is variable. And therin lies rub #1.

A lot of the discussion out there is that a power meter is more accurate than a heart rate monitor, so just training via power  alone is a better way to go. I gotta say, that statement doesn’t really make a lot of sense to me. That’s like saying the fuel gage in your car is more accurate than the temperature gage. They measure 2 entirely different things. Powermeters measure the power that your body has generated. Heartrate measures the stress your body is under while you produce that power. So if you leave the heart rate monitor at home for your next ride or race and simply go by the goal wattage that you have calculated for the event, you may be setting yourself up for a tough day. This is because 250 watts in hour 1 in comfortable weather is much different that 250 watts in hour 5 in the heat. It’s these slow changes over time that make it so critical to have an eye on your body while you have the other on the power.

Another problem you run into is the need for constant testing if you want to do it right. Since a watt has no connection with effort of the body, your improvement will be tougher to recognize without constant testing. For instance, lets take our fictional athete Joe. He does a threshold test for power and comes up with a wattage of 100. So he builds a training plan around that wattage and follows it to the T.  After a couple weeks of training though, his body has adapted and improved. So now his effort level has dropped well below what it was intended to maintain the test goal power. Because of lower effort levels his improvement stagnates some. The next test shows that his power has increased to 150.  So somewhere along the way he should have increased the ride wattage to keep up with his improvement. Without testing though, it’d only be a guess.

Now take Larry who trains wattage and heart rate, but does his threshold tests based on heart rate. So Larry finds that at his threshold heart rate of of 150 he is crankin out 100 watts as well. Larry follows his plan just as religiously, except Larry’s wattage increases througout the training cycle as he maintains the goal heart rates. The next time Larry tests he finds that his his threshold heart rate has increased to 152, but his wattage at that effort level is now 165. Larry’s rate of improvement is faster because he’s monitoring both the stress of the work and the output of that stress. Or we can test Larry using wattage like Joe, but correlate a HR to the threshold. Now when Larry improves and is riding 100 watts, he notices that his heart rate is now 130. Time to test again.

Add to this the fact that power doesn’t tell you when you are getting dehydrated, tired or about to bonk. If all you are looking at is a power for guidance, you will miss the fact that your heart rate has climbed ever-so-slowly over the last 2 hours, and now you are starting to run the engine way too hot. It’s the perfect set up for a disaster of a run. Pair the power and the heart monitor together though, and you will see the dehydration coming before you ever have a chance to get into trouble.

The other big problem with training exclusively with watts has less to do with the technology and more with the approach. If you look through the training practices, you’ll see that there are goal and average wattages that are tested for. So Joe buys a powermeter, tests for his functional threshold power (ftp) and finds an average or goal power for any given distance. So far so good, but the next part is the problem. Once given the average power, Joe tries to stay at that average regardless of terrain. So he ends up riding  easier than normal up the hill and then harder than normal on the descent.  On the surface, not a big deal, but add in aerodynamics and wind resistance and the worm starts to turn. The air resistance on the climb is much different than the descent not to mention the gravitational resistance.  So while power is much more important on the climb, Joe eases up, and when aerodynamics are much more important on the descent, Joe pushes harder. Need an example? Where is the Tour de France won and lost? The mountains. An not just the mountains, but the climbs. If you have 2 riders exactly the same size, and both average 250 watts for a ride, they should have the same time, right? Wrong.

Take rider A, who does just as I mentioned earlier. His goal for average watts is 250, so he keeps that wattage across the entire ride. Up hill, downhill, flats, whatever. His effort is balanced and his ride seems strong. Well done.

Now take rider B, who rides the exact same course and averages the exact same wattage, but takes into account all aspects of the ride. On the climbs when he can get the most power bang for his buck, he pushes 350 watts instead of 250. While this is a bit more tiring, he also makes huge ground on rider A. With little air resistance climbing, all that added power translates directly into speed. Then, rider B only puts up 150 watts on the descent, and at times, even coasts. Why? because the air resistance is exponentially stiffer as speed increases, so with a focus on aerodynamics and rest, rider B only loses a small amount of time to rider A and is able to fully recover.

“The Look”

Now the real world example (although admittedly a bit extreme). Watch a mountain stage of the Tour de France. All the moves are made on the uphill because the power output most directly influences the pace. Once you’re dropped, it’s pretty much over. Sure, riders can make up some time on the downhill, but it’s incredibly small by comparison, even if the wattage is considerably greater than your competition. You could even call it return on investment.  So while you kill it on the downhill and get a small return, your competition is banking that effort for the uphill, where they will get a huge return by comparison.

So if you get more bang for your buck generating more power on the uphill, how can you tell when more power is too much power for you to handle? That great big red thing between your boobies. Strap on a heart rate monitor my friend, and find out how much stress your putting on yourself.

Power on Wayne. Power on Garth.

  • Delicious
  • Facebook
  • Digg
  • Reddit
  • Google
  • LinkedIn
  • StumbleUpon
  • Twitter


  • first off I would like to say, good article and I agree with you about power and heart rate for bikeing. a quick thought about the uphill surges with regards to power and taking advantage of those uphills. wouldnt that be the same as what many refer to as burning matches or going to hard? if not maybe you could clarify what kind of surge uphill, based on your expample 350 watts up a hill for most would be burning a match. I am thinking about this approach for oilman and to me it seems like i would be toast after the bike was over. your thoughts?

  • It’s a great question that I can’t answer. I know how I respond on that course but that’s no help to you. I know that when I hit a hill and my HR hits X, then I need to taper off the effort. More importantly though, on the downhill if my HR doesn’t drop to Y, I know that I put in too big an effort. This is where HR becomes so important. If you can learn those types of markers, then odds are you won’t pop on the run, regardless of the course. If however you keep that power steady without learning the ranges of your effort, well you get the idea.

Leave a Reply

Your email address will not be published. Required fields are marked *