FTP Test

Today marks day one of my build up on the bike to the Ironman 70.3 U.S. Pro Championship in St. George, Utah. I haven’t done any training since Muskoka 70.3 on September 8th, so I thought it would be best to start with a test to determine where my Functional Threshold Power currently is. In my last post I said that I would follow Coggan and Allen’s FTP testing protocol, which in a nutshell consists of a 5 minute all-out effort followed by a 20 minute effort holding the highest wattage possible, and then taking 95% of the average wattage of the 20 minute hard effort to be your FTP. I decided I would alter this a bit to keep my training log consistent. The last time I performed an FTP test was December 26th, 2012. In that particular instance I warmed up for 20 minutes and then held the highest wattage possible for 20 minutes. I then took 95% of the average wattage over the 20 minutes to be my FTP. During that test I held 373w. This works out to an approximate FTP of 355w (.95×373=355). I decided to repeat this exact same procedure today.

First off, I should mention that I have been wondering about the accuracy of my Quarq Cinqo Saturn powermeter. At Muskoka 70.3 I held about 335w for the 94k bike course (339w normalized). It was good for the fastest bike split of the day. Here is power vs. time from the race:

Muskoka Power

But, I have done some research on wattages from other professional triathletes (Training Peaks has published many files) and know that this is on the higher side. Thus, I figured it wouldn’t hurt to remain sceptical. It doesn’t vary from day to day, so it is alright to use in training, but the numbers may not be meaningful for comparison. Anyways, I am quite confident that the CompuTrainer Lab Model is very accurate and so today I planned on measuring the results of the test with both the CompuTrainer and the Quarq, and finding out once and for all if the Quarq on my bike is accurate.

Additionally, I should mention that I was a bit overwhelmed by just how much the CompuTrainer has to offer. I stayed up until 3 am last night tinkering around with the 3D Course Builder and other fun stuff…it reminds me of the days when I used to play video games…it is very addictive! Also, I ran a new in-practice 10k pb (with the help of my dad’s awesome pacing and encouragement!!) yesterday of 31:24, so I was expecting this to have at least a slight effect on my performance.

First, I aired both front and back tires to 120 psi. I then did a 20 minute warm-up. During this time I ran the CompuTrainer calibration program to ensure utmost accuracy (afterwards there was 3.10 lbs of press-on force). At about 13 minutes into the warm-up my good friend Gary Hutchinson popped into my head and I realized I wasn’t wearing my heart rate monitor, so I hopped off for a second to put it on. After 20 minutes and a few accelerations I started the 20 minute max effort. I chose to do it on a flat 10 mile strip (pre-made in the CompuTrainer Racer One Software package). I started off a bit ambitious and by 5 minutes was averaging around 415w. It started to catch up with me and dropped to around 411w by 10 minutes. From 13 to 17 minutes was quite painful, but I managed to keep the watts fairly steady. Once 20 minutes had passed I paused the course and lapped my Garmin 405, then saved the CompuTrainer data.

I was very pleased with the results. According to the Garmin 405 I averaged 415w for the 20 minutes. According to the CompuTrainer I averaged 408w for the 20 minutes. My average heart rate over the duration was 158 bpm, with the maximum of 169 bpm occurring right at the end of the effort. As you can see the heart rate steadily crept upwards the entire time (Garmin 405 data):

Heart Rate and Power on Garmin

Here is a graph of the CompuTrainer data for the 20 minute hard effort:

CompuTrainer Data Graph

I was very happy to discover that my Quarq Cinqo Saturn powermeter is only off by around 7w. I can live with this, but still have a bit of scepticism. After cooling down I then calculated what I will take to be my FTP. 408x.95=387.6w. This number is VERY important as I will base all of my subsequent workouts and training intensities on it from now until December (when I will taper and repeat this exact same test again).

Over the next couple of days I will write another blog post about how I will utilize my FTP that I determined today to create a chart containing all of my target wattages for bike workouts over the next 2.5 months. Check back on the weekend for this.

8 thoughts on “FTP Test

  1. World class power numbers Lionel ! Can’t wait to see you race next year. It will be great for Canada !


  2. Great stuff Lionel. The numbers are very impressive. Remember to stay consistent and healthy bro – longevity is key! Numbers are a good guide, but so long as you keep getting PB’s in training and at races that’s really all that counts bro – the results will reflect as they already have. What ever works for you buddy, obviously you are doing something right – just remember to listen to your body well because chronic fatigue syndrome is debilitating. Looking forward to witnessing your successes Lionel. All the best.

  3. Pingback: Training Zones | LSandersTri

  4. I’d like to add that according to my experience, I see some drift between the power readings of my CompuTrainer vs my bike powermeter (Powertap units). I originally believed the computrainer better until I realized the calibration value is sensitive to heat and riding conditions of the previous minutes. Sometimes it’s within 2 watts of my Powertap, Some other times it can be off by 20 watts – and that is using the same warming up protocol from session to session and making sure the PTAPs are calibrated.

    So the last year I’ve been relying on my bike’s powermeter as continuous reference (recording uploaded to Training Peaks) also during the winter. Since it’s also the one you use when training/racing on the road, it makes sense to keep the measurement tool uniform across the board rather than mixing-and-matching. Just thought you might find this info useful in case you do also see a variation from session-to-session between the 2 measurement devices.

  5. P.S. Awesome pacing on your 20′ effort it seems!
    P.S. Weird how the computrainer power readings for the 20′ effort is pretty steady vs the Quarq (Garmin Graph, right?) which shows that cyclic ondulation. What’s up with that?! The latter does not seem right…

    What mode did you use the computrainer on, for the FTP test?

    • Hey Thierry. Thanks for the comments. I am currently doing the same thing. I have been using my bike’s powermeter in conjunction with the CompuTrainer, and will continue to do this at all times. So far the two seem to be quite similar. I have done subsequent rides after the FTP test where the difference was even less than 7w. The CompuTrainer claims to be accurate to within +/- 1.5%, which at 400w output would be 6w. I’m not entirely sure, but I would imagine the Quarq has a similar guarantee. For logging purposes I will likely take the average of the two. For my purposes I am not too concerned with the validity of the measurements, as long as it doesn’t vary much from day to day.

      As for pacing, a friend convinced me that good pacing on the bike is the key to running fast off the bike, so it has been one of the most important elements in my training over the last 1.5 years.

      As for the graph, a lot of the effect is because the Garmin Training Centre automatically smoothed the data. I uploaded the Garmin data to TrainingPeaks and it looks a lot more like the CompuTrainer data. I will upload this graph as a note in a later post.

      For the test I used the “Power Training” portion of the CompuTrainer One software, with the 10 mile pre-loaded flat course.

Comments are closed.