Why You Don’t Need ERG Mode for a Ramp Test
By Ric Stern
So, you’re getting ready for winter training. The nights have drawn in, the smart trainer is set up, and before you start a block you do what everyone does now: a ramp test, to set the numbers your whole winter will be built on. And if you’ve read the forums, you’ll have been told there’s only one correct way to do it. ERG mode on. Stay seated. Don’t move until you stop.
I developed the method of estimating FTP from a ramp test back in the 1990s — before FTP existed, before Zwift, and long before ERG mode. And I can tell you: you don’t need ERG mode to do one.
It started in 1995. I was at university doing sports science, and the lab had a Kingcycle test rig. You’ve probably never heard of one, and that’s fine. Damn, it was loud. But it was smooth, and it hooked up to a computer. You typed in the ramp — say 30 watts per minute — and away you went: an ever-increasing effort you controlled with your own pedaling, keeping your power inside a range on the screen. Let up and the bar dropped. Push too hard and it rose above the range.
I rode that rig a lot. I was racing road and criteriums and time trialing at the time — 10-mile and 25-mile time trials, like most UK riders did then — and I wanted to know whether you could predict 25-mile time trial power from a ramp test. FTP didn’t come along for another five years or so. There was no ERG mode doing the work for you. The athlete controlled the effort, the way you do on the road, where the resistance rises as you ride harder.
None of this means a ramp test is a free-for-all. You still have to follow the ramp. If the target is 300 watts, you need to produce 300 watts — not 250, not 350. ERG mode just takes away the need to control the power yourself. If you really want to use it, there’s no rule against it. But there is a problem with it, and it shows up exactly where it matters most: the final minute.
ERG holds you to the target power by adjusting the resistance for you. As you tire towards the end of a test and your cadence starts to fade, the trainer increases the resistance to keep the power up — which makes it harder still to turn the pedals, so your cadence fades further. Riders call it the spiral of death. The change in how the resistance responds alters how you’re pedaling, and the acute increase in loading can bring a rider to a premature halt — sometimes over a tiny, momentary falter that would cost you nothing on the road.
A ramp test lives or dies on its final minute. That last minute is the number everything else is calculated from. Riding the ramp yourself — what Zwift users call world mode or free riding — you can grind, stand up, squeeze out the last few seconds. In ERG, a brief dip in cadence can end the test on the spot, and your result comes out lower than the fitness you’ve actually got.
I see it in my own numbers. My MAP — maximal aerobic power, effectively your best one-minute power at the top of the test — is about 30 to 50 watts lower in ERG mode than when I ride the test myself. I see the same pattern in other riders too, though not everyone. Set your winter zones off the lower number and you’ll spend months training at the wrong intensities, which is one of the quieter ways winter training stops working.
And while we’re clearing things up: there’s no rule that says you have to stay seated either. I have no idea where it came from, because it was never a condition. By 1997 I was running ramp tests on a motorized treadmill — the gradient and speed rising together until riders were producing over 500 watts at under 159 pounds (72 kg) of bodyweight, out of the saddle in the dying seconds, squeezing out the last bit of power. (I once tested a junior on that rig who went on to win medals and a Grand Tour. He was out of the saddle too.) What does matter is consistency: whichever way you test, test the same way each time, so your numbers are comparable.
One more thing the software gets wrong. Some built-in ramp tests end the moment you falter. In the lab, when an athlete started to fade, we didn’t stop the test — we shouted. Strong verbal encouragement, back on the power, finish the minute. A momentary dip isn’t failure, and micro-breaks don’t invalidate a test. Software that kills it at the first wobble is throwing away your best effort and calling the remains a result.
Done properly, a ramp test gives you MAP, and from MAP you can estimate FTP — with a caveat I’ll keep making: FTP isn’t one fixed number. For most riders it lands at around 72 to 77 percent of MAP, and how long you can actually hold it varies from about 35 to 75 minutes depending on the athlete. Which is why the test matters, and why a mode that understates your final minute matters too.
If you’ve done a ramp test, my free calculator will work out your training levels from it, check whether your FTP looks realistic, and more besides. And if you want those numbers turned into a winter that actually progresses, that’s what MAP: The Build is built on — progressive training structured around your current fitness, your goals, and the time you have available. It costs £25 (about $34) every four weeks.
So no, you don’t need ERG mode for a ramp test. You never did. The test is older than the mode by decades of lab practice — all it asks is that you follow the ramp, and finish the minute.
Have a training question for Ric?
Ric Stern has spent decades answering riders’ training questions and has coached cyclists since 1997, from club racers to Paralympic medallists and Tour de France professionals. Cardiovascular fitness, strength training, VO2 max, race tactics, nutrition, or whatever is keeping you from a stronger season: ask away. We publish questions using your first name and last initial only.
Coach Ric Stern is the founder of CycleCoach. He holds a first-class degree in Sport Science, has published peer-reviewed research in exercise physiology, and has coached riders from track sprint through to ultra-endurance and the Tour de France.
At 57, he’s still racing himself and is fitter now than he was in his 20s.
