Power and data without the religion.

The religion problem.

Before we get into any of this, the durability stuff, the training system, any of it, we need to talk about power and data. Because if you don't understand what your numbers actually mean, nothing else I teach you is going to make any sense.

Here's the issue. Most cyclists have this weird relationship with data. Either they ignore it completely. I've seen a professional use their power meter as a simple speedometer, so they're just riding by feel, essentially, which is fine until you want to get faster. Or they treat it like scripture. They are so dialled in on the number, convinced that every watt means something, that every ride needs detailed analysis, because something really important must be hiding in that really fine detail.

Neither of these is useful. Data is a tool. It's not a religion. And the moment you start worshipping your FTP number or obsessing over your TSS, you've stopped using data and started being used by it.

This chapter is about what actually matters. What the key numbers are, what they tell you, and, just as importantly, where they stop being useful. Because that's the part that doesn't really cross over into the language communicated to athletes most of the time.

Critical Power and W prime.

Let's start with the most important concept in all of this. If you only understand one thing about power, make it this.

I want to explain why I've chosen this as the starting point, because it's a deliberate choice. Most coaches and most platforms start with FTP, Functional Threshold Power. It's a 20-minute test, you take 95% of it, and that's your number. And it works, to a point. But FTP is a single number from a single test. Critical Power, on the other hand, comes from the relationship between multiple efforts. It's built from your entire curve, not just one snapshot. I moved to CP years ago because it gives me, and you, more information and more reliability. And that's not a controversial opinion in the science anymore. Jones and colleagues established in 2019 that CP is the gold standard for identifying what's called Maximum Metabolic Steady State, the boundary above which your body can no longer maintain homeostasis. Karsten and colleagues confirmed in 2021 that CP and FTP agree closely at the group level but diverge substantially in individual riders, too much agreement loss for them to be used interchangeably. But in the cycling world, FTP still runs the show because it's simpler to test. I'd rather give you the better tool and teach you how to use it.

The other reason I use CP is because of W prime, and we'll get to that in a moment. But what it means is we can get a much bigger picture of you as a cyclist by measuring what's under Critical Power and then your capacity over Critical Power, because of how these relate to each other and to specific parts of training.

Let's look at this. If we made a graph with power on the left-hand side and time along the bottom, and I asked you to do a bunch of different maximal efforts, a 3-minute effort, a 4-minute effort, a 12-minute effort, a 20-minute effort, and we plotted your best power for each of those durations, what you'd get is a curve. If it was done correctly at max effort, you'd get a curve. High on the left, short efforts, lots of power, dropping down as the efforts get longer. That's your power-duration curve, and it's basically your fingerprint as a rider.

Now here's the important bit. That curve doesn't just keep dropping forever. At some point it levels out. It flattens. And the power where it flattens, that's your Critical Power. CP.

When we talk about what CP actually is, it's the highest power you can sustain in a metabolic steady state. Below that number, your oxygen uptake and lactate can stabilise, you can reach a steady state and keep going. Above it, they can't. They rise progressively until you're done. It's not the same as FTP, though they are related. FTP is an estimate from a single test, and it's shown to be around 5% lower than Critical Power. Morgan showed in 2019 that the average difference between FTP and CP is only about 1% at the group level, basically indistinguishable. But at the individual level the discrepancies can run to 12 to 13%, sometimes more. So your FTP might be close to your CP, or it might be 30 to 40 watts off. You wouldn't know unless you tested properly. CP is derived from the relationship between multiple efforts. It's more robust, both in its test and scientifically. And it's worth noting that CP isn't a razor-sharp line. Poole and colleagues, 2016, describe it as a practical fatigue threshold with predictable day-to-day variability. It's a zone, not a border. But the way you get to CP, through a curve, not a single test, tells you a lot more.

The battery: W prime.

Now, like I touched on, everything above that line, above CP, that's where it gets interesting. Because the work you can do above CP is finite. Think of it like a tank, or even better, think of it like a battery. In the science, it's called W prime.

Think of it like this. You've got a battery. When you ride above CP, you're draining it. When you ride below CP, you're recharging it. Now, that is very simplified, and we don't even have it nailed down exactly how much it's going to recharge. But what we can use this for is to understand the size of that battery, which is expressed in how many kilojoules you can spend above CP before you're done. That, in essence, is your W prime.

Typical W prime values: if you're an endurance-focused man, it might be between 9 and 15 kilojoules. If you're more punchy, this attacker style of rider, think Tom Pidcock or even Mathieu van der Poel, 15 to 18. Women typically 6 to 10 kilojoules. And these aren't fixed. They respond to training.

This is why it matters practically. If you attack a climb and your W prime is small, you've got less margin. If your W prime is big but your CP is low, you can surge but you can't sustain. The interplay between these two defines what kind of rider you are.

And this also explains why you blow up. If you go above CP and you drain your battery, you drain your W prime faster than you can recover it, at some point the battery hits zero. And when it does, you are cooked. You are done. And it's not a mystery. It is simply maths.

Why fresh numbers stop explaining.

So here's where most people stop. They know their CP, maybe their FTP. They know their W prime. They've got zones set up. And they think they understand their data.

But here's what they're missing. Everything I just described, CP, W prime, your power-duration curve, all of that is measured when you're fresh. Fresh legs, fresh head, good night's sleep, well fuelled. These are test day numbers.

And how often are you fresh when it actually matters, meaning at the end of a ride, at the end of a race? Almost never. You're four hours into a gran fondo. You're at the back end of a training week. You're three weeks into a block and your legs feel like they're full of sand. And in those moments, your numbers, the ones from the test, they don't describe what you can actually do. They describe what you could do if you weren't tired. And that's a really large gap.

Your fresh numbers describe your capacity. What you can actually reach when you're fatigued, that's your access. And the difference between those two things is what we're going to spend a lot of time on later in this playbook, because it's the part that almost everyone ignores.

For now, just hold onto this: your data is most useful when you understand where it stops being true.

What to actually look at.

Now let's think about what we're actually looking at, the stuff that matters day to day when you're riding a bike.

Power, obviously. Heart rate. RPE, rate of perceived exertion. Traditionally it was a 6 to 20 scale; now, generally, the RPE that's used is a 1 to 10 scale. Those three things together are the important ones. Not one of them alone. All three, because they tell you different things, and the mismatches are where the real information lives. We'll go deep into this in the five golden rules chapter.

What your head unit should show you. Keep it simple. Current power, 3-second average, not instant. Instant is nearly impossible to keep steady while you're riding. Heart rate. Time. Cadence if you want it, and it is useful to understand your natural cadence over time, so that when you're manipulating it, you've got a baseline. That's it. You don't need 17 data fields. You only really need four.

What to look at after the ride. There are a bunch of general questions you'd ask if you're doing any type of analysis. Did your power hold where you wanted it? Did it meet the expectation or the intent of the ride, did you stick to the power that was prescribed? What did your heart rate do relative to the power?

And this is where it gets interesting, because power itself is what it is. Like in the gym, a kilogram is a kilogram, a pound is a pound. That doesn't change. What changes is your response to that power. A lot of information is in that response. Understanding what your heart rate is doing, how quickly it goes up, how quickly it comes down, whether it gets up and plateaus or whether it keeps rising across a complete interval or a complete day. If it starts nosediving after a couple of hours. There's a lot of different things you can see. You can start to see where you fade, where fuelling may impact you, where heat may impact you. And we all know heart rate isn't perfect, but it has its place in this process.

What to ignore, and why. There's a concept I want to introduce here, because it applies to a lot of metrics: when a metric becomes the goal, it's no longer effective. This goes back even to the 20-minute test for power. If you just want to be good at testing for 20 minutes, there are many ways you can do that, but you can overestimate or underestimate your FTP by losing the original intent of the test and trying to optimise that test instead. That's where a metric becomes useless. And competitive athletes always ask me how they can get the best number on any type of test. I always have to pull them back and say: the intent is not to have the best number for the sake of the best number. The test is trying to be a snapshot, not just picking up one score.

So knowing what to ignore and how to ignore it is really important. TSS as a daily scorecard, not useful, fundamentally because of the flaws in the system, the inability to separate intensities within the TSS number itself. And here's one that catches a lot of people out. Sanders and colleagues showed in 2017 that power-based and heart-rate-based zone calculations give you completely different intensity distributions for the same ride. HR underestimates time spent at high intensity because heart rate lags power. By the time HR catches up to a hard effort, the effort has often already shifted. So if you're using HR zones to track your training distribution, you're getting a different picture than power zones would give you. Strava estimated power, don't worry about it. AI-generated insights that don't know anything about you, we could poke fun at Strava here, it's going to get better, but right now it literally means nothing. And this is the big one: anyone telling you that a single number defines your fitness. There is no single number I know of that clearly defines fitness. Not CTL. Not FTP. There is literally no number.

Without the religion.

I want to say one more thing about data before we move on.

There's this idea in cycling that's been growing for the past ten years, that more data is always better. More metrics, more devices, more analysis. The tech companies love this, because every new metric is a new product.

Even this morning I spoke to an athlete who was getting too caught up in the numbers. I had to separate them from the numbers, specifically their HRV and resting heart rate, because it was influencing them so badly that they were questioning their own instinct. They felt good. They were riding well. 100% ready for a race in a couple of days. And they were almost pulling out of the race because the numbers didn't match how they felt. Now we haven't got to the race yet, so I can't tell you what happens. But as far as the numbers getting in the way of what you feel, I simply had to ask them: "Not considering the numbers, how do you feel?" And that's the thing. The numbers can be good. But you need separation from them, for starters. It's why it's good to have another person looking over you. And they can absolutely get in the way.

This is a small lesson from what I've learned coaching hundreds of riders. The ones who get the most out of their data aren't the ones with the most of it. They're the ones who know which numbers to pay attention to, which to ignore, and when the numbers are lying to them, when they're not helping in the way that they should. All of this information should give you what you need to make decisions on. And if it's not giving you that, good or bad, then it's not working right.

Data without context falls into this category. It's just background noise. Power without understanding fatigue is only half the picture. And any number measured fresh is only telling you what's possible. It's not telling you what's available at specific points along your training and racing and events that you need it.

As we go through this playbook, I'm going to keep coming back to this. Use the data. I love the data, I'm into it. But respect it, don't worship it. Because the moment you let a number override what your body is telling you, you've got the relationship backwards.

Right. Let's talk about what your curve actually looks like.