Training Metrics

Joe Friel provides an overview of the key training metrics that help coaches quantify training load and plan for performance.

If your athlete’s FTP is wrong, then a lot of the other things you’re going to be making decisions about will be wrong—so let’s get that right.”


In this video, Joe Friel explains some of the key training metrics used in determining load, such as frequency, intensity, duration, normalized power (NP), normalized graded pace (NGP), training stress score (TSS), intensity factor (IF), and functional threshold power (FTP). He underscores the importance of making sure your athlete’s FTP is correct, as it is the backbone from which you create and build your athlete’s training plan.

Friel said: “It’s important to remember that FTP is only helpful if it is accurate. If an athlete is training hard and FTP improves, training zones need to be adjusted to reflect the new FTP or the athlete might not gain as much benefit from the training.” 

He advises retesting FTP every 6 weeks to be sure it’s an accurate reflection of where your athlete is in their training progression. Friel added: “Keep an eye on their training data, as data from a hard group ride or threshold interval training can be a good way to spot changes. While much of the software available will provide an estimation of FTP, avoid reacting to minor fluctuations. After all, it’s an estimate. Be sure to weigh other factors that reflect the athlete’s progression.”

Video Transcript

Let’s talk about the intensity of training. That’s probably the most difficult for the athlete, and for the coach to measure—duration is easy. You can talk about how many minutes or hours the workout lasted. But when it comes to intensity, that’s much more difficult. We use something now that’s called the Functional Threshold (FTP). This is a good way of measuring, for example, power on the bicycle (or pace for running) .

It can be used in many different ways, one being to set up the zones that the athlete is training in when they do their workouts. The types of zones we will talk about are:

  1. Power zones
  2. Heart rate zones
  3. Speed zones (the most important metric in this presentation because almost everything else we mention is based on this).

How to Determine FTP

How do we determine the FTP? Well, there are actually several ways of doing this. The most common way to do this is with a 20-minute test. Basically, the athlete warms up, needs to be a fairly common repeatable warm-up that the athlete does, because this test is going to be repeated many, times over the course of a year. Therefore, we want to keep things as close as we can to being the same every time. The warmup is done by the athlete, and then the athlete commences to do a 20 minute time trial. Basically, as hard as the athlete can go for 20 minutes. I can guarantee the athlete will always start too hard the first time they ever do this, that’s the biggest mistake made. If they do that, they wind up limping to the finish line at the 20-minute line.

What you should do is tell your athlete to start off conservatively, and every five minutes make a decision about how hard they should go for the next five minutes. Ask them, “Was the last five minutes too hard, or too easy?” Make small changes all the way through the test. That keeps the athlete from crashing and burning before they get to the end of the test.

There are other methods of doing this, I’ve used the 30 minute test with the athlete. However, the 20-minute test is the most common way of doing this.

Option 1: The 20-minute test checklist

  • Take the athletes data
  • Find their average power for those 20 minutes
  • Subtract 5%
  • Get a good notion of what their FTP should be

However, is this perfect? No, it’s not. There are going to be some athletes who are very good at 20-minute tests, and for other athletes it’s not going to work nearly as well. Find a test that you think works best for your coaching methodology and for your athlete, and then use that test every time when you’re having your athlete perform the test to determine their FTP.

Option 2: Observe data from a super hard workout

Another possibility, which is kind of interesting, is this chart you see in the video above (02:49), which has to do with power distribution over a period of time or over the course of an entire workout. The athlete does a very, very hard workout. For road cyclists, for example, this could be a group ride, a race-like ride. It could even be a race. It can be a one-hour race the athlete does.

What you see is the way it’s always going to look. It’s always going to have this pyramidal shape to it where there’s a high point in the middle, there’s a slope up on the left side, and the slope down the right side. The FTP, interestingly enough, is always going to be on the right side of that high point. It’s going to be at a point where there is a noticeable drop in the data points. You can see on this chart in the video above (03:36), that there’s a point where the bar graph drops off rather rapidly. The point just before that, for some reason, typically winds up being the FTP for your athlete.

Is it perfect? No. It’s just another way of looking at data.

What you’re always trying to do as a coach trying to get this FTP correct is gathering data about your athlete. It could be WKO or it could be this power distribution chart. It could be a 20-minute test. It could be race data you’re looking at, but you’re always looking at data trying to decide if you have the FTP correct. Because if the FTP is wrong, a lot of things you’re going to be making decisions about later on are also going to be wrong. So let’s get that right

Normalized Power and Normalized Graded Power (NP/NGP)

Let’s talk about another topic related to this whole idea of intensity of training, which is really very difficult to measure. Turn to the chart you see in in the video above (04:33), and  look at the dark line that connects the two dots, that that would represent an athlete going out. For example, riding their bike and riding very steadily over the course of time, some period of time, let’s say they rode one hour very steadily, didn’t change their power at all. Whatever the power may have been remained that way all the way through the states, 150 watts. The athlete rode 150 watts non-stop for one hour. It was perfect. They did it exactly right. That’s very easy to measure, it’s 150 watts. That’s all there is to it.

However, we all know that athletes never do that, you never go out and ride exactly a precise number all the way through the workout. There’s always things that are causing the power to go up and down all the way through, just as you see on the graphic in the video above (05:20). You can see that red line, that’s more like what happens when an athlete is riding in a workout. It’s always going to be something like that, it’s never going to be a straight line. But the averages could be exactly the same. The dark line you see connecting those two dots is the average power for both of those workouts. Yet, it’s not the same workout.

What’s the difference between the two? The big difference is the metabolic cost. The metabolic cost of riding steadily is quite low. The metabolic cost of changing intensity, for example, high intensity to low intensity, up and down that hill, is very high. The cost is very high. That’s because of acceleration. Acceleration requires a lot more energy than maintaining the same intensity all the way through the session. That’s what we’re going to be talking about here is this idea that there’s a normalized power. Now the way this normalized power is determined is quite complicated. I’m not going to go into the whole thing, but there’s a long formula that has to do with taking the average power to the power of four, manipulate it in several ways. Then, by the bottom line, you come out with a number, which is an indicator of how hard the workout was metabolically. What it always does is, it gives a greater emphasis to the accelerations, that’s what the formula does. Therefore, when there’s an acceleration, there’s a greater emphasis on that. Just as when there’s an acceleration metabolically, there’s also greater emphasis on that you burn more calories. So we’ve got the same idea here, and the same concept being used to come up with what is called the normalized pace for the athlete. So normalized pace, and normalized power reflect the metabolic cost of doing the workout.

Intensity Factor (IF)

What we’re looking at in the video above (07:14) is a couple of formulas. The one on the left is showing us how intense the workout was based on a comparison of normalized power to FTP. You should understand what those two mean by this point. If you take the normalized power the athlete had in the workout and divide it by the athlete’s FTP, we then know what is called the intensity factor. In other words, how hard were they working relative to their FTP. It’s going to be a percentage. Let’s go back and give examples of that. Let’s say in the example, that the athlete’s normalized power is 188 watts, and the athlete’s FTP is 250. If we divide 188, by 250, we get an intensity factor of 0.75. What that tells us is the athlete was working at 75% of their intensity factor for that workout.

So that’s an indication of how hard the workout was. But, what does that number mean to us? Well, any time the number is 1.0 or higher, we know the athlete was at their intensity factor, or above it, that would be very hard, because we know what the FTP is, it’s an extremely hard effort for about 20 minutes. Therefore, the athlete is there. If it’s 1.0 or higher, they’re at FTP or higher than FTP, we know that’s going to be very hard. Everything below 1.0 is going to be easier. We’ve got a range of intensity factors that tells us how hard the workout was for the athlete, that’s going to become our way of expressing what the intensity was for the workout, and we’ve based that on FTP and normalized power.

Training Stress Score (TSS)

Let’s take all that now and put it into a formula so we can come up with a way of expressing how hard the workout was based on intensity and duration. We’re going to need to combine these things into one number so we can talk about how hard the workout was. This is something called the Training Stress Score (TSS). You can see in the video above (09:10) how the TSS is calculated. I won’t go into all the details of this. But you can see that everything in red there has to do with intensity: normalized power, intensity factor, and FTP. Those are all intensity-related elements. The things in grey: sec stands for seconds, and that shows how many seconds was the workout. Below that you see 3,600, that’s the number of seconds in an hour. Then we have a standard number, a default number of 100.

Whenever we plug the numbers in there (the number of seconds in the workout), the normalized power and the intensity factor, we multiply that out, and then we divide that by the FTP multiplied by 3600. We take that derivative and multiply that by 100, we get something called TSS. I know that sounds really complex, there’s a lot going on here, a lot of math, the good news is, software will do that for you. You don’t need to think your way through all this.

On the right hand side of the video above (10:16), you can see examples of this using the very same numbers we had on the previous screen, we come up with a TSS of 112.8. That tells us how hard the workout was for the athlete with one number. Now we know what that difficulty for that workout was that athlete.

How Metrics are Highly Individualized

Now what does that number mean for that athlete is 112.8, high or low? It depends on the athlete. We’ll come back to that topic much later on, but for right now, just realize there are some athletes who can train with a very high TSS. For example, pro road cyclists are typically turning around on average 150 TSS per day, that’s extremely high. A novice athlete may be training on a TSS which is closer to 50 TSS per day, that may be very hard for that athlete because of their level of expertise, their talent. These numbers are intended to give us a starting point with a number for the entire workout considering both the duration and the intensity. But, realize it’s specific to that athlete, we can’t compare this athlete’s TSS, with another athlete’s TSS, and draw conclusions about their potential or abilities.

Estimating TSS

Let’s talk about how you can do this in a very quick manner and almost wind up with exactly the same number without going through all that stuff I talked about earlier. This is a very easy way of doing it. If you take the intensity factor for the athlete that they did that day and square it, and then multiply that by the number of hours for the workout. If it was two hours, you multiply by two or if it’s two and a half hours you multiply by 2.5. Then, multiply that by 100, you also get the TSS. Interestingly enough, which is almost exactly the same number we got by using the full formula on the previous slide 112.5 and 112.8 are essentially the same number it’s not significant. You could do this in your head.

I’ve talked to the athletes before, and I’ll tell them, I want them to come up with a TSS, something particular for that day and need the average intensity factor and here’s how you can think about it yourself, and they can do it themselves. We can do this very very closely in your head while you’re riding if you get to the point where you’re pretty good at math, you can work with almost any number you want. You can estimate what that number squared would be, without getting involved and getting down to precisely what the number is. We can come up with like six times six is 36. Seven times seven is 49. Eight times eight is 64. We can start coming up with ballpark numbers, which give us an idea of how the athlete is doing per hour, then multiply the hours and you get the whole thing figured out.

How TSS Guides Daily Training

I mentioned earlier in this talk that there were three metrics that go into determining how hard the athlete is training. So far, I’ve only talked to you about two: intensity and duration. But I also mentioned earlier, there was a third thing called frequency. These three things are the big things that make up what the athlete’s training load is all about. I’ll get into a little bit more in detail next presentation. But, realize here now that we’ve got with the athletes, I’ve been talking about here, he’s got 112.8, which was the athlete’s workout on that given day. What if that was a hard workout for that particular athlete, we’ve decided for this particular athlete, it’s a hard workout. Frequency would tell us we don’t want to do that every day, you don’t want your athlete going out and doing 112.8 every day. We need to have some easy workouts in there. Maybe instead of having 112.8 tomorrow, the athlete is now going to do 64 or something like that. A much smaller number, a much lower number, so they can recover from the hard workout.

Now we’re starting to build a plan where we’re talking about frequency, where we have hard days and easy days. We need to make up our minds about how we’re going to organize those days but now we’ve got a way of measuring what is hard and what is easy for this particular athlete. That’s the key point here. What is hard for this athlete? What is easy for this athlete? Therefore, I want to design a workout which includes frequency or day to day training, manipulating what the TSS is going to be on any particular day.