Jay Vine cut his weekly training from 24 hours to 20 and got faster. Two Vuelta stage wins, an Australian time-trial title, the Tour Down Under. He sat in front of me and said the quiet part out loud — most pros don't ride 30 hours a week. He doesn't. He rides 20.
That's not a "less is more" platitude. That's a specific, measurable change in how a professional builds a week. And the reasoning behind it is the bit most amateurs miss when they read the headline and think "great, I'll cut my training too."
Here's what he actually changed. And what to take from it if your week sits at ten or twelve hours, not twenty.
What changed when he moved to UAE
When Jay was at Alpecin, his FTP was being estimated through critical-power testing — a protocol that uses short maximal efforts to model your sustainable power. The number it gave him was, in his words, "Contador eating a bad steak sort of ridiculous." So high it was unreachable. So unreachable that he couldn't complete the prescribed sessions built off it. Every week was a quiet failure stacked on top of a number that didn't reflect what his body could actually do.
UAE switched him to 20-minute race efforts to find his FTP. Not lab tests. Not a ramp protocol the day after a hard week. Real efforts inside real races, where the wheel doesn't get let go and the data is as honest as it gets. The number came down. The sessions became completable. The training started to bank.
The hours dropped at the same time. From 24 a week to 20. Not because UAE believe less is better in a generic sense — because the terrain in Andorra, where Jay lives, doesn't allow soft-pedalling. A "five-hour ride at 200 watts" doesn't exist there. Every climb is a 260-watt climb whether you want it to be or not. So the four hours he dropped were the four hours that were doing nothing useful — they were just adding fatigue under the disguise of "endurance."
What replaced them was specific time-trial work. Position drills. The kind of focused intervals you do when you've already accepted that the bottom of your aerobic pyramid is built and the only thing left is sharpening the parts that decide races.
That's the part to internalise. Less wasn't the lever. Specificity was the lever. Less just made room for the specificity.
Why the same logic matters at amateur volumes
If you're at ten or twelve hours a week, the first reaction reading this is "great, I'll cut to eight, problem solved." That's the wrong read. Most amateurs already train too little, not too much. The lever isn't "ride less." The lever is "stop wasting the hours you have."
What does wasting hours actually look like at amateur volume? Three patterns show up over and over with the riders who apply to the Not Done Yet community.
The grey-zone week. Four or five rides at the same moderate intensity. Nothing genuinely easy, nothing genuinely hard. The rider feels like they're training, the legs feel a bit cooked all the time, the FTP doesn't move. The fix is the same one Prof. Stephen Seiler has been writing about for twenty years — go easy when you go easy, go hard when you go hard, eliminate the comfort zone in the middle. We covered this in detail in the polarised training breakdown.
The volume-without-direction week. The rider rides because they have time, not because the rides are doing a job. Saturday's three-hour group ride is hard for ninety minutes and easy for ninety. Tuesday's intervals are at "whatever I feel up to." The hours go in but the body has no clear signal about what it's adapting to. This is what UAE removed from Jay's week. The fix is the same — every session needs a stated purpose before you push off.
The mismatched-specificity week. The rider trains for a hilly sportive but never does long sustained climbs. Or they target a flat 100-mile gran fondo and do all their work as 90-second VO2 efforts. The engine being built doesn't match the demand of the event. The fix is brutally simple — ask what the event demands and audit whether your week reflects it.
Jay's quote, when he was talking about the difference between trying to "win a race" and being given a specific job: "You know what six and a half watts per kilo feels like for 20 minutes and when it's done it's done. You finish, you've done your job." That clarity — knowing what the session is asking of you, knowing when you've delivered it — is the thing most amateur weeks lack. Not hours. Clarity.
The FTP testing point most riders get wrong
This is the part of the conversation that should change how you handle your numbers.
Vine has stopped trusting lab-style FTP estimates because the protocol gave him a number he couldn't ride to. Race-based 20-minute efforts give him a number that holds up in training. The lesson generalises hard.
Every step of an FTP test protocol is a place for the number to drift. The trainer wasn't calibrated. The power-meter zero-offset hasn't been done in six rides. The room was warmer last time. You did the test on the back of a heavy week. You did the warm-up differently. You did the 20-minute effort on a course that drops down at the back end. Two or three of those together and you've moved 3-5%, which is more than enough to make a real improvement look like a plateau, or to mask a real decline.
If you've been "stuck" at the same FTP for nine months, the first thing to investigate isn't your training. It's the testing protocol. Have you used the same method, the same equipment, the same physical state, the same warm-up, the same time of day? Most amateurs haven't. Most "FTP plateaus" are partly measurement noise.
The fix Jay is using — extract the number from a real maximal effort inside a race or a hard training session, not from a synthetic test — is the closest amateur equivalent of how the most experienced coaches work. Joe Friel has been making this argument for years: the best test is the most race-like effort you can sustain. We laid out the full diagnostic in Why Your FTP Is Stuck — five causes, and measurement noise is the cause that most often gets dismissed.
If you've been training seriously for more than a year and the numbers have stalled, the Plateau Diagnostic is the four-minute version of that audit. Specific question, specific recommendation. Free.
What "specific" actually means at amateur level
Specificity isn't a buzzword. It's a structural decision about what to put in the week and what to leave out.
For Jay it means: time-trial position rehearsal, sustained efforts at race wattage, sauna work in the background to handle the heat in Spain, gym maintenance, recovery management around the demands of stage racing. He has identified the components his races demand and the week is built around those components.
For an amateur targeting a hilly hundred-mile sportive, the specific list looks different. Sustained climbing efforts at 90-95% of FTP. A weekly long ride that gradually approaches the duration and elevation profile of the event. Pacing practice on rolling terrain so you stop blowing up on the climbs because you went 20 watts over what's sustainable. Strength work in the gym during base. In-ride fuelling rehearsal at 90-100g of carbs an hour, because the engine you're building has to be one that can actually use that fuel. (Our Fuelling Calculator sets the baseline if you've never measured it.)
For a time-crunched cat 3 racer trying to break out of cat 3, the specific list is different again. Race-pace surges. Repeated VO2 efforts with short recoveries. Sprint endurance work. The middle of the week protected for full recovery so the weekend race-pace work lands. We laid the full version out in the time-crunched racer training breakdown.
The point in all three cases is the same. The week is reverse-engineered from the demand of the event. Volume is what's left over after the specific work and the recovery have been built in. Not the starting point.
The piece amateurs underestimate: how the team plans nutrition
The bit of the conversation that should make every age-group cyclist sit up: Vine's daily fuelling is co-ordinated between the team nutritionist and his coach. They map out a target weight by a target date. They build the calorie and macro plan to hit that weight without crashing his hormone levels — without, as Jay put it, leaving him "with no testosterone by the end of it." They don't try to drop a kilo and a half in four days before the Tour de France.
That last sentence is the one. Pro teams don't crash-cut. They hit weight by sequencing food and training across months. Most amateurs do the opposite — ride more, eat less, hit the deficit, watch the FTP drop, watch the hormones drop, watch the morale collapse, give up two months in.
The Roadman position on this hasn't changed. Body composition matters. Crash-cutting is the fastest way to wreck a season. Hit the weight by fuelling the work, not by under-eating around it. We laid the full case out in the body-composition breakdown and in Anthony's own 9kg-eating-more story.
The point isn't that you need a team nutritionist. It's that the planning horizon needs to be measured in months, not days, and the food has to support the training rather than fight it.
How to apply the Jay Vine lesson without copying it
Three honest takeaways from the conversation.
One — audit the hours you have before you ask for more. If your week is ten hours and the FTP isn't moving, the problem isn't the eleventh hour you can't fit in. It's the structure of the ten you've got. Pull the last four weeks of training files. What percentage was below 70% of max heart rate? What percentage was above lactate threshold? Where did the rest sit? If the middle bucket is bigger than 25%, you have the grey-zone problem and adding hours just stacks more grey.
Two — test FTP the way Jay does. Treat the most race-like effort you've done in the last month as the most honest data point. If you've got a 20-minute climb you ride hard once a fortnight, your power on that climb is a better FTP signal than the synthetic test you did on the trainer six weeks ago in different conditions. Standardise the protocol when you do test. Measurement noise alone makes a 3-5% mess of most plateaued riders' files.
Three — write down the demand of your target event before writing the plan. Duration, elevation, intensity profile, fuelling demand, mental demand. The plan should be reverse-engineered from the demand. Not "what's the standard 12-week build?" — what does this event actually require, and how does the week reflect that?
Jay's gain wasn't from doing less. It was from doing the right things, in the right proportions, around a number that was honest. The four hours he dropped weren't the lever. They were what fell off when the lever got pulled properly.
Most amateurs don't need fewer hours. They need the same hours, structured around what their body and their event actually demand. That's the lesson under the headline. The headline is just bait.
A note on the AI-coaching question
Vine is sceptical of AI replacing human coaches and I think he's right for the reasons he gave. Performance isn't only data — the bit a good coach reads is the texture of how the athlete reports a session. Whether the rider is hiding fatigue. Whether the family stress at home is stacking up. Whether the confidence wobble after a bad race is a real problem or a one-off. No model trained on power files reads that yet.
Where AI helps right now is the joining-up of ecosystems Jay flagged — Whoop and Oura sleep data, training-load data, in-ride fuelling data, all sitting in their own silos and never talking to each other. The coach who pulls those streams together for an athlete week to week is doing real work. The model that one day stitches them automatically will be a useful assistant. Probably not the prescriber. Probably not for a long time.
If you've read this far and the line that hit was "the four hours he dropped were the four hours doing nothing useful," that's the audit to do this week. Find your four. They're in there.