Watts in a Flat White: Does the Café Stop Make You Faster or Slower?

Cyclists will argue about anything you can measure in watts or grams. Tyres, inner tubes, socks, the angle of your elbows. We’ve stood in café queues listening to grown adults debate whether a waxed chain is worth the faff for a couple of watts, and we’ve been those grown adults more often than we’d like to admit.

On a ride a couple of weekends ago, somewhere on the drag back up from the café, one of us asked the obvious question. If we’re counting every watt, why has nobody ever counted the coffee? That flat white went in at the halfway stop. Is it making us faster up this hill, or slower? Is it a performance upgrade, or two hundred grams of hot milk we’re now hauling up onto the moor?

We didn’t settle it on the bike. We didn’t have the breath. So we went home and did the sums properly. We’re two brothers with a calculator and a café habit, not sports scientists, so take this in the spirit it’s meant. But the numbers are real, and they turned out more interesting than we expected.

The watts lost: carrying it up the hill

Start with the bit the weight weenies care about. A flat white is roughly 200g of liquid, and once it’s in you, you’re carrying it.

For the sums we used a fairly ordinary rider: 75kg, on an 8kg bike, holding a steady 250 watts up a 4km climb at 7 per cent. That takes a shade under 18 minutes. Add a flat white and holding the same speed costs about 0.6 of a watt. Over the whole climb, that’s around two seconds.

Add a slice of cake (call it 150g, we’re not weighing it at the counter, we’ve got some dignity) and that’s roughly another 0.4 of a watt.

So the full café stop, coffee and cake, costs you about one watt on a proper climb, and next to nothing on the flat, where weight barely matters. If you’re the sort of rider who’d halve your toothbrush to save a gram, even you can relax about this one.

The cake isn’t dead weight anyway. A decent slice is something like 400 calories, and there’s a handy rule of thumb in cycling that every kilojoule on your power meter costs you roughly one calorie, because your body is only about a quarter efficient and the units happen to cancel out. So that slice is about 400 kilojoules of riding, which is around half an hour at 200 watts. It’s the only upgrade we know of that carries its own weight and then some.

The watts gained: what caffeine actually does

This is where it gets good. Caffeine is one of the most studied performance aids in sport, and the research is remarkably consistent.

A 2018 review that pooled 46 studies found caffeine improved average power in endurance time trials by about 3 per cent, and finishing times by a bit over 2 per cent. For our 250 watt rider, 3 per cent is 7.5 watts. Put that next to the cake: you’re giving away one watt and getting seven and a half back.

And it isn’t just caffeine pills. A 2013 study on trained cyclists gave them either caffeine, actual coffee, decaf or a placebo an hour before a 45-minute time trial. Caffeine and coffee both made them about 5 per cent faster. Decaf did nothing at all, which anyone who’s been handed one by mistake could have told them.

Part of how it works is in your head, in a good way. A 2005 analysis of 21 studies found caffeine made the same effort feel about 5.6 per cent easier. It doesn’t so much give you new watts as make the ones you’ve already got hurt less, which, if you’ve ever been on the rivet up a Devon lane, is basically the same thing.

It’s also entirely legal. Caffeine used to be on the banned list, with a limit on how much could show up in a test, but it was taken off in 2004 and is now only monitored. It might be the only performance aid in sport that comes with a slice of lemon drizzle.

The café stop is in exactly the right place

Here’s the bit that made us laugh. Most of the research gives caffeine an hour before the start. But a 2016 study tried it mid-ride. Cyclists rode for 80 minutes, then had caffeine, then finished with a time trial at the two-hour mark. Just 100mg made them roughly 4 per cent faster. 200mg made them about 7 per cent faster.

Ride for an hour and a bit, have a coffee, ride home. That isn’t a sports science protocol. That’s a Sunday.

In fairness to the scientists, they gave it in a sports drink on the bike, not as a flat white followed by forty minutes of standing around talking about tyres. But we’re choosing to believe the café ride was optimised all along.

The watts lost that nobody puts on a spreadsheet

If the weight is a rounding error, what actually costs you at the café? This lot comes from years of experience rather than any study:

• The stop itself. Forty minutes at zero watts. Your elapsed time takes a battering, and it’s the one number the mate who didn’t stop will bring up.
• Café legs. The first ten minutes back on the bike, when your legs feel like they’ve been borrowed from someone much older. Nobody’s studied it. Nobody needs to.
• Getting cold. Sitting outside in damp kit in October turns a good ride into a shivery one faster than any headwind.
• The second coffee. More isn’t better forever. Past a point you get the shakes and a dodgy stomach, not more watts, and you’ll be eyeing up quiet laybys within the hour.
• Tonight’s sleep. Caffeine hangs around for hours. A late-afternoon double can cost you a decent night, and that’s tomorrow’s watts gone.

One thing that isn’t on the list is dehydration. The old line about coffee drying you out doesn’t really hold up. A 2014 study found moderate coffee kept regular drinkers about as hydrated as the same amount of water.

The balance sheet

At the café stop	Watts
Caffeine, taking the research at face value	+7.5
Carrying the flat white up the climb	−0.6
Carrying the cake up the climb	−0.4
Net, roughly	+6.5

On our 4km climb, that works out at roughly 25 seconds quicker to the top with the coffee in you than without.

Now think about how cyclists normally pay for watts. Ceramic bearings, aero socks, a skinsuit for the club run: people happily spend hundreds of pounds chasing a watt or two. A café flat white is about three and a half quid. That’s roughly 50p a watt, which makes it the best-value upgrade in cycling by a country mile.

The honest small print

We said we’d always be straight with you, so here’s what the table leaves out.

• One flat white is a fairly small dose. Most studies use 3 to 6mg of caffeine per kilo of body weight, which for a 75kg rider is 225 to 450mg. Café coffee varies wildly, and a study of Glasgow coffee shops found single espressos ranging from 51 to 322mg, so you genuinely don’t know what you’re getting. For a lot of riders one cup won’t reach the doses in the studies, which means the 7.5 watts is on the generous side. The real gain is probably smaller, but it’s still comfortably bigger than what the weight costs you.
• Everyone responds differently. Most people in the studies got faster. A few didn’t, and a couple of studies even saw riders go slower.
• These are trained cyclists in a lab, not two brothers on a Sunday ride arguing about whose turn it is to pay.
• Caffeine isn’t for everyone. If it doesn’t agree with you, or you’ve been told to cut back, none of this is worth it. Have the cake.

So, faster or slower?

Faster. Comfortably. On the numbers, a decent coffee at the halfway stop is one of the few things in cycling that gives you more watts than it costs, and it’s the only one that comes with a chair.

Your elapsed time on Strava might disagree. But Strava doesn’t know how good the coffee was.

Honestly, the watts were never the point. The stop is where the ride gets talked about, the route gets argued over, and someone says “same again next week?”. We wrote about where that habit comes from recently, and it turns out riders have been doing it for nearly 150 years without a single spreadsheet. It’s just nice to know that, for once, the numbers are on our side.

If you want to test the theory on a real ride, come out with us. We’ll stop at a good café, we’ll have a coffee, and we’ll all pretend it’s for performance reasons.

— John & Rich