How to “Cheat” by Using a Power Meter in an Ironman, by Joe Friel
Tuesday, October 30, 2012 at 2:27PM Using a power meter in an Ironman triathlon is almost like cheating. While others are trying to gauge intensity through a cloudy veil of emotion, which makes perceived exertion and even heart rate nearly useless, the athlete with a power meter is focused on a number that, if maintained with only slight variations, will produce an optimal bike split.
Note that I said “optimal.” The goal on the bike is not to produce the fastest time possible, but rather the time that will leave the athlete with enough spring in the legs to actually run the marathon. A maximally fast bike time means a 26.2 mile walk. Such a marathon always leads the dejected athlete to believe he or she is simply a poor runner. That’s seldom the case. It is usually a bike-related issue.
Why Power is Important
A power meter prevents this from happening if used correctly. It provides several real-time measures of progress throughout the race. Two that are critical to the Ironman triathlete are Training Stress Score (TSS) and Intensity Factor (IF). Determining what these should be weeks in advance of race day and then training with these numbers allows the athlete to ride at an intensity that will produce an optimal split and a best- possible Ironman run time.
If you already have a power meter you are probably aware of TSS and IF. Few athletes, however, really understand their value in a long, steady-state bike race such as an Ironman. So let’s do a quick refresher using a couple of recent Ironman Hawaii examples and then I’ll show you how to use these numbers to produce a fast Ironman.
TSS and IF came from the mind of Dr. Andy Coggan, a highly regarded exercise physiologist and roadie. He has singlehandedly created a system for effectively measuring and using power data. What is described here comes from his genius.
What are TSS and IF?
You don’t need to understand all of the details of these terms, but for those who may be interested, here are the ways TrainingPeaks and WKO+ software determine each after a ride. During the race or workout the head unit on your handlebars is doing the same calculations and displaying the numbers in real time. Bear with me for a short time while I explain them. We’ll start with the formulae and then I’ll explain.
Training Stress Score is a way of using a number to tell how hard a ride was. It’s based on the interplay of intensity (power) and duration (time). The formula is…
TSS = IF2 x bike time (in hours) x 100
That then brings us to Intensity Factor. Here’s the formula. Again, this is not on the test, but it’s important for where I’m taking you with this discussion—to a faster run off the bike.
IF = Normalized Power (NP) ÷ Functional Threshold Power (FTP)
Don’t worry too much about “Normalized Power.” It’s just a number your power meter and the software compute which is similar to average power but with an emphasis on the metabolically costly surges you made throughout the race. What you must understand is the other part of the formula: Functional Threshold Power (FTP).
FTP is not calculated by the power meter or the software. They accept whatever you tell them your personal FTP is. It’s simply a number that lets you know what the greatest average power is that you can do for an hour. It generally takes a test or race to get this number right. This is the most critical component of using a power meter for a fast race. (You can read more about how to determine your FTP here.)
Comparison: Pete Jacobs vs. Age Grouper
OK, now that you have a better understanding of what the numbers on your handlebars mean, let’s take a look at two athletes—a pro and an age grouper—from this year’s race in Kona to get a sense of what it all means.
Here are the basic power metrics for Pete Jacobs who won the Men’s Pro category and for Christopher Thomas of the Timex Multisport Team who finished fourth in the Men’s 40-44. (We thank both athletes for being gracious enough to share their data; Jacobs' was collected through his Quarq power meter and Thomas' was collected through his Timex devices).
;)
View the complete Kona files of Pete Jacobs and Chris Thomas.
Note that their TSS were almost the same—286 for Chris and 285 for Pete. So why was Pete 29 minutes faster if their TSS were essentially the same? Because his FTP is higher. Pete’s is 370 watts and Chris’ is 296. The higher your FTP, the faster you can ride.
Pete's Quarq power meter showed that he rode at an IF of 0.79. That means his Normalized Power was 292 which is 79% of 370. The men’s pro winner in an Ironman almost always finishes the ride with an IF of about 80%. You can count on it.
Chris’ IF was 0.75 so he rode at an average of 75% of his 296 FTP. That produced a bike split of 5:04:13—very respectable for the heat, humidity, and hills of Kailua Kona.
I’d like to point out, however, that Chris probably could have improved his run time (a not-too-shabby 3:08:39) a little by pacing his bike a bit more evenly. Note that in the “% NP by Half” metric that he rode at a much higher power output for the first 56 miles (108%) than he did for the last 56 miles (92%). This is partly explained by the greater amount of hill climbing in the first half of the course. But note that Pete, on obviously the very same course, parsed it out with a slightly more balanced power distribution (105-95%). The pro race strategy must be significantly different from that of the age groupers—an interesting topic for another time. Chris’ strategy would have resulted in less fatigue by the start of the run had he done something like 103-97%. That would have produced the same NP but would have better reflected the nature of the course while conserving energy.
I’m certainly not trying to be critical of Chris here. Finishing 4th in your age group at the Ironman World Championship is quite an accomplishment and his finish time of 9:25:30 is outstanding. What he definitely got right was the overall Intensity Factor and Training Stress Score. Had he gotten these too high or too low it would have resulted in a far slower finishing time. He could just tweak his pacing a small amount to make the bike split even faster.
Here’s a table I use to help determine what an athlete’s IF and TSS should be for an Ironman race. This was created by Rick Ashburn. If you know how to use it you can nail your Ironman bike and come off with good legs for running.
;)
How to Use the Table
Your bike-finish TSS should fall into either the light gray, dark gray, or yellow sections depending on how you categorize yourself (see the color-code legend to find your race category). Then on the left side of the table find your goal bike time. By looking to the right of the goal time and in your color-coded category you’ll see a range of one to four TSS numbers. Then by glancing up to the top row for both ends of this colored range you’ll see what your IF should be throughout the race. Then you simply ride in that IF range on race day and—voila!—you have your optimal bike time and are ready to actually run the marathon.
So let’s look at an example. My goal is to have a six-hour bike split in my next Ironman given my experience, the nature of the course and anticipated race-day weather conditions. I categorize myself as an “age group athlete with good preparation” (dark grey). By looking to the right from 6:00 I can see the anticipated TSS must be in the range of 269 to 286. Then looking up I can see that the IF range must be 67 to 69%. That means I’m going to ride at 67 to 69% of my FTP. So if my FTP on race is expected to be 250 watts (note that FTP rises as your fitness changes so it must be tested every four weeks or so), then my power range on race day must be 168 to 173 watts. I’ll go about 20% higher on the steepest hills (200 watts), and I’ll coast the steepest downhills (0 watts). By setting my head unit to show both instantaneous power and IF I can monitor and gauge my intensity throughout the bike leg of the race. And, of course, I will do workouts in which I practice doing the same, exact numbers. So on race day there will be nothing new. I just do what I’ve rehearsed dozens of times in training. And have a great race!
For complete Ironman Kona data analyses of many of this year's top athletes, check out the TrainingPeaks Kona page.
Are you new to training and racing with power? Joe Friel's new book The Power Meter Handbook offers triathletes and cyclists a simple user's guide to using a power meter for big performance gains. Learn more here.
Joe Friel, M.S. is the author of ten books on training for endurance athletes including the popular and best-selling Training Bible book series. He holds a masters degree in exercise science, is a USA Triathlon and USA Cycling certified Elite-level coach, and is a founder and past Chairman of the USA Triathlon National Coaching Commission. For more information visit his website at www.joefrielsblog.com. You can also view and purchase Joe's training plans on TrainingPeaks.
Reader Comments (19)
What head units show instantaneous TSS and IF? I have the original yellow wireless Powertap head unit, which only shows power. I had looked at the Garmin Ant units, but didn't want to deal with keeping the batteries charged.
I've been reading your posts for a long time and love all the power information. I have a power meter on my road bike (quarq) and on my mountain bike (power tap). I have a couple questions I will lump together right now. The first is power testing. Is the difference better power meters and bike geometry going to produce a great enough difference in my FTP that I should test on each bike separately? If so, should I test each bike each 4 weeks or alternate? Next, my focus is on endurance mountain bike races, typically in the range of 6-12 hours, sometimes longer. When pacing for these long races, can I completely ignore HR and ride based on power, or should I pay attention to my HR? I ask because I recently did a race and had my power in the correct zones, but my HR was skyrocketing. Sure enough, I bonked 3 hours in. Lastly, I saw a post of yours on a power file analysis of Jordan Rapp's where you talked about how athletes should stay in Z2 except for a few efforts where you can go into Z4, but not for more than 5 minutes. What guidelines (or training tips) would you have for mountain bikers when the course dictates when you have to put in the hard efforts (such as short steep sections where the only way to get up them is with s sprint)?
Chris 1,
All the CycleOps and Garmin units show TSS and IF. Any of them will require keeping the batteries charged. Since you're familiar with CycleOps products already, try the Joule 2 or the new Joule GPS.
-Joe
Chris 2,
Hard to say if there will be a significant difference between power meters and bikes. I've seen some where it is and others were it isn't there's only one way to know for sure and that's to test both set ups.
Sorry to hear of your bonk. But if power was right on and heart rate high then my conclusion would be that your aerobic fitness had a lot of room for improvement. Power is the key metric to pay attention to during a race.
The key to using a power meter during a MTB race is the same as for an Ironman--establish an appropriate Intensity Factor, which was discussed in this article. I can't tell you what that should be for your race.
-Joe
I've seen this table before, and while I may agree with the premise I'm not sold on how it's presented here and would love to hear your thoughts. If I understand this table correctly all of the data is based on IF, which is a ratio. At an extreme it says that if your FTP is 100W, you could finish an IM ride in a little over 5h if I just ride at 70W. I know you state in the post that you need to pick a time based on experience, and clearly if my FTP was 100W, finishing a 112mi ride in 5h is not realistic, but still, for the novice that piece of data can get lost. Especially with over optimistic first timer's. Another reason I don't like this table is that it doesn't highlight enough how important VI is to a good run. You can easily wind up doing a bunch of mini intervals coming out of turns, nutrition/nature breaks, or simply inattention and find at the end of the ride you did many 1min intervals at 120% of FTP, yet still come in with an IF for the ride in the 60-80% range, but good luck running well after that.
What I've been trying to figure out is how can you take the TSS goals ( say 250-260 for a novice), and knowing my FTP calculate an estimated IM time and IF, which I could then adjust based on training and actual race conditions of that day?
Chris,
I can't figure out the chart whatsoever. If in the age group example from above, how could Chris keep the same ~5 hour time but have more left in his legs for the run. He'd have to run at a lower IF, right? So lets put him in the safe zone for 5 hour time and see what happens. Working right from the 5hour mark, his TSS is between 231 and 245 ( makes sense, less stress equates to more left in his legs ). Looking up from the 245 ( higher end of safezone ) his IF would be .70. Chris' FTP is 296 so: 296*.7 = 207. Ok good, less NP than before I get it, more left over energy. But the problem is, if Chris rides with a NP of 207, how can he possibly still make the 5 hour mark? I put all the numbers back into the equation and solved for time and got 5hours so it checks out, but I don't understand how if Chris doesn't change his FTP and rides with a less NP, how can he keep his time the same?
I have been reading a lot of your posts on using Power Meters, but before one can begin to use the technology one first needs a Power Meter. Do you have any good recommendations that won't break the bank, but can make reasonably good use of why one would get and use a Power Meter in the first place? Thank you. Regards, Bruce
I'm having a hard time digesting this and I think I've figured out why. I'm trying to go through the math and there's a gap that I can't cross. I feel like the chart above implies there is a relationship between power and time, correct? Given it's a good way to simplify the process and give an approximate of what your time might be, but I think there's still a big gap which is the rider's unique power/speed profile (his aerodynamics if you will)
TSS = (IF^2)*(time in hours)*100
IF = %age of FTP
If I'm going to do a race on a course I've not ridden, my time would be a bit of a guess, so for me to say that I'd ride 5:10 or 5:30 would really be a bit up in the air until I see the course and the race day conditions. But, I think what I can take away is that my TSS should be (assuming I'm of the dark gray category) in the 270-285 range. Then I can back calculate the IF I need to ride at to reach a particular time, but without knowing my specific power numbers on a specific course, it's nearly impossible to calculate my needed FTP to produce a particular time. Am I misunderstanding something or am I trying to do too much w/ the numbers? I guess maybe I should take the more general knowledge of target IF to be 71 +/-4, but I'm not sure really. Can you help me fill in the missing pieces? Thanks Joe! Always appreciate your breakdowns of the sport for those of us that are always learning.
Matt
Thanks for a really nice, practical application of IF and TSS. I understand it all much better now. Still though, like Matt's comment above, the table doesn't seem to account for variabilty in an rider efficiency and between courses. Two extreme examples would be between two otherwise equivalent riders, one in aero for 100% and the other sitting up, or between a completely flat 112mi and a 112mi that goes straight up, say, 8000 ft. The chart doesn't differentiate across these variabilities and they would seem significant. I can't imagine an IF = 0.7 would yield me the same time at Lanzarote as it might at Florida. .
What would be a recommended target TSS for HIM ? 180ish ? .85IF for 2.5hrs, .8 for 3hrs
ChrisS, Yes, you're right. The table is based, in part, on selecting a goal time which is reasonable. You don't base it strictly on IF. And you're also right on VI. It must be low. Should be 1.05 or less. If the rider stayed in the TSS range from the table that would certainly keep VI low. But there are likely to be hills. Power should increase slightly (about 20%) for the high end on climbs, depending on the grade and the length. For downhills there may well be some coasting, which is a wise move, again depending on grade. Both of these will cause VI to rise slightly but if done as I described here then VI will be below 1.05. This article was not intended to be the end-all of training. There are many, many issues. Another, for example, would be the NP splits by half which was briefly mentioned above. There's a lot to training for IM, isn't there?
Adam, Not sure I fully grasp your dilemma. He had to ride at 75% to do ~5 hours. If he rode at 70% he would not have gone that fast.
Bruce--It comes down to three matters: 1. Cost. They are relatively expensive. 2. Interchangeability between bikes. Some are easier to swap around than others and this comes down to things like the wheels you prefer to race with. 3. Weight. Some weigh more than others. Since I don't know your money situation or how many different bikes you may want to use it on or how concerned you are with weight, I really can't answer the question. I'd suggest doing some research and figuring our which one best fits your needs. Good luck!
Matt--Yes, you need to determine a goal time for the race. This may be based on experience (yours or others), using a tool such as www.analyticcycling.com or pure guess. Once you decide what you can determine the appropriate power range.
Any chance of one of these tables for Half IM and Olympic Distance? Keep up the great work
Cheers Chris
how many TSS points up in an average week Ironman pro in running and biking
thanks. ¡
My question is what is the typical difference in speed in km/h from the FTP to an IF of say .70 to .80. I appreciate it would change depending on what your speed and power at FTP is but can any general statements be made? I don't have a power meter, but my experience is that to get those last few km/h takes a significant amount of power. I guess assuming an athlete could maintain there FTP power over the course of an IM distance what is the time difference (or average speed difference) compared to what is achieved at around a .75 IF?
Also interested to know what my power output would need to be for a given time. For example let's say I want to complete the distance in 5:30 and in the dark gray that correlates to a TSS of 285 at an IF of .72. Is there a way a can calculate what my FTP ought to be to achieve that?
Hi, thanks for the article. I'm preparing for a tough long ride in the mountains. 235km, 4000m climbing. I would expect my time to be anywhere between 11 to 13 hours. You graph doesn't show info for over 6.5 hours. Do you have any recommendations?