When @yinuo-chen signed up in February, he told the coach two things about himself, in one sentence: he weighs 70 kilos, and he wants 4 watts per kilo. Then he went and rode, most days, for six months.
For five of those months, nothing happened. His estimated FTP sat between 226 and 237 from the first of March to the second week of August, a flat line through twenty-four weekly snapshots. Riders quit over less. In the middle of August it went 232, then 259.
The number that finally moved
The jump is the part people will want, so here it is properly. His five-minute best went from 282 to 316 watts. His twenty-minute best went from 231 to 244. His one-minute best went from 398 to 450. Those are recorded efforts, not estimates: the power curve genuinely moved at three different durations inside two weeks.
The estimate that sits on top of that curve moved with it, 232 to 259, and it is worth being precise about what that means. An estimated FTP can only see what you show it. Part of that 27 watts is fitness he built over the spring. Part of it is that he finally rode hard enough, for long enough, for the model to have something to look at. Both halves are real, and neither is a reason to discount the other.
At 70 kilos, he has gone from about 3.3 watts per kilo to about 3.7. His own target is 4.
The day it showed up
On a Sunday in late August he rode 松山大海坨, four hours and twelve minutes in the mountains northwest of Beijing, 292 TSS, more than two thousand kilojoules of work. It was not a steady tourist ride. Threaded through it were repeated Z5 to Z7 efforts, short and sharp, on climbs, deep into a long day.
That was the third such session in a week. Two rides at 静湖 came before it, both at 164 TSS: one just over two hours at an intensity factor of 0.88, the other with fifteen separate hard efforts inside it. Neither went to plan in the usual sense. The first was not on the plan at all, and the second ran about 40 percent longer than the session he had been given, at more than twice its intended load. The week those three sessions landed is the week his power curve changed shape.
What the score noticed
| Week | Total | Fitness Base |
|---|---|---|
| May 31 | 68 | 64 |
| Jun 28 | 46 | 44 |
| Jul 26 | 59 | 68 |
| Aug 16 | 60 | 77 |
| Aug 23 | 74 | 77 |
| Sep 6 | 79 | 77 |
The June trough is the honest part of this story. He was riding through it, most days of most weeks, on a Polarized plan, and his fitness (CTL) was climbing the whole time, from 47 in early March to 67 by late August. It just was not turning into anything the numbers could see yet. Fitness Base is the column that eventually agreed with him: 44 in late June, 77 in the latest snapshot.
There is another quiet number in his file. On a set of two-minute efforts in late August, he held 222 and 223 watts through four repeats with his heart rate eight beats lower than the same work four weeks earlier. That is not the kind of thing anyone posts about. It is the kind of thing that shows up later as watts.
Twenty-three watts
His plan runs in auto mode and has been adjusting his weeks for him since the spring; every adjustment with a recorded outcome has landed well. The fortnight since that August week has been daily endurance and threshold work, and the self-set target date he wrote down in February, early September, has now arrived: the FTP test his plan has been counting down to falls this week.
At the latest snapshot his estimate reads 257, so twenty-three watts separate him from 4 W/kg, and his twenty-minute best has moved again in the meantime, from 244 to 255. He spent five months looking like a rider who was not improving, and he was wrong about that, and so was the estimate. Watch what his curve does after this week's test, once a fortnight of threshold work has given the model something else to look at. That is the snapshot I will be waiting for.
You can follow @yinuo-chen at intervalcoach.app/p/yinuo-chen, and see your own power curve and eFTP history on your Analytics page. If you would like to be featured in a future spotlight, reply to any IntervalCoach email and say so. The riders in this series are real, the numbers are theirs, and the work is the point.