- ⚡302 W xPower, held start to finish
- 🔥~1,000 kJ of work every hour
- ⏱️for 9 hours, 15 minutes
- 📉with only a 5.5% fade in the last hour
That's how Mads Wurtz Schmidt won the Unbound Gravel 200. When he crossed the line first, the headlines went to the result. We went to the power data, because the watts tell a story the result can't.
That 302 W is just 76% of his FTP. But FTP isn't the story here: his durability is. What won the race wasn't how much power he can make fresh; it's how little of it he gave back across 207 miles and 9,378 kJ of work.
Durability: the second axis of FTP
FTP is a fresh-legs number, the power you can hold for about an hour when rested. It tells you nothing about whether that power is still there after 6,000 kJ of work. That second question, how much of your power survives deep into a race, is what we call the second axis of FTP. It's the axis that wins gravel.
Two riders, two very different lines. Mads (orange) barely sags: his modeled threshold drifts from a fresh 5.75 w/kg down to 5.44 at race depth — 94.5% retained after 133 kJ/kg of work. Cobe (blue), eight years younger, actually starts higher at 5.88 w/kg, but his line dives to 4.87 — just 82.8% retained. Same race, same brutal dose of work, and the rider who gave back the fewest watts won while the one who gave back the most finished 16th. That gap between the lines is Durability.
The real story is hour 9
Compare his first half to his second half and you'd see a fade of essentially zero, but that number lies, because his easy opening hours drag the first-half average down. The honest fatigue signal shows up in a single hour:
He held 307 W through hour 8, then faded just 5.5% to 290 W in hour 9. After 118 kJ/kg of work, that's the only real crack in the whole ride. But he had a safe lead, so he could afford to let off the gas a little bit.
| Hour | Power | W/kg | % FTP | Zone | kJ/kg | Cum kJ/kg | mph |
|---|---|---|---|---|---|---|---|
| H1 | 293 W | 4.25 | 74% | Endurance | 12.4 | 12.4 | 21.9 |
| H2 | 275 W | 3.98 | 69% | Endurance | 13.3 | 25.7 | 23.1 |
| H3 | 322 W | 4.67 | 81% | Tempo | 15.1 | 40.8 | 23.1 |
| H4 | 310 W | 4.49 | 78% | Tempo | 15.4 | 56.3 | 26.3 |
| H5 | 306 W | 4.43 | 77% | Tempo | 15.4 | 71.7 | 22.4 |
| H6 | 305 W | 4.42 | 77% | Tempo | 15.6 | 87.2 | 21.6 |
| H7 | 307 W | 4.45 | 77% | Tempo | 14.9 | 102.1 | 25.6 |
| H8 | 307 W | 4.44 | 77% | Tempo | 15.7 | 117.8 | 20.4 |
| H9 | 290 W | 4.2 | 73% | Endurance | 15.0 | 132.7 | 20.3 |
| H10 | 283 W | 4.09 | 71% | Endurance | 3.2 | 135.9 | 22.2 |
The work: 136 kJ/kg
Total mechanical work was 9,378 kJ, or 136 kJ/kg once you normalize for his roughly 69 kg. That's the single most comparable durability number there is: it needs no maximal effort, and every rider gets one. 50 kJ/kg is a good endurance ride; 100 is a hard century; 136 is elite ultra-endurance.
His rate of work was about 1,017 kJ per hour, and once the race settled he held it remarkably flat (roughly 15 kJ/kg every hour) for six straight hours. That steadiness is the durability.
A quick word on how steady
One number captures it: his fatigue-weighted power, xPower, a CoachCat metric, came out at 302 W with a Variability Index of just 1.07. In plain terms, he didn't surge and recover, surge and recover. He held a near-constant effort for nine hours and turned a gravel race into a time trial. Steady is durable.
What this means for you
The takeaway isn't that his FTP is 5.8 w/kg; it's that durability matters as much, if not more so, than FTP. Two riders with the same fresh threshold can finish a gravel race twenty minutes apart, and the difference is entirely the second axis. The good news is durability is trainable: long rides, fueling, and threshold work deep into accumulated fatigue all move it.
A full video walkthrough of this analysis is coming soon to the FasCat YouTube channel. Subscribe so you don't miss it.