How Many Watts Does Drafting Save in Cycling?

Photo by Paolo Candelo on Unsplash

I'm a lazy cyclist. Worse than that, I'm an impostor. On fast group rides, I'm more often hanging off the back of the group than doing turns on the front. Last time, at an unreasonably high heart rate, I found myself wondering. "How many watts am I actually saving by drafting?" Ask around a club run and you'll hear that drafting saves you "about 30%", or "a third", or "40% if you're really tucked in". All of those are quoting a drag number and calling it a power number. They aren't the same thing, and the gap between them is big enough to change how you ride.

I built a cycling drafting power calculator on top of the peloton aerodynamics literature, so rather than guess I ran every position, every group size and the full power range through it. Every number below is the tool's own output.

Drafting cuts your drag by up to 38%. It cuts your power by 18-32%.

The short answer#

On a flat road at typical group-riding speeds, sitting in a single-file paceline saves 20-30% of your total watts compared with the rider on the front. The second wheel saves the least, around 20%. The best seat, roughly position 5 or 6 in a decent-sized group, saves about 30%. The absolute ceiling in the model is 32%.

The reason it isn't the 38% you see quoted for drag: drafting only helps you with the air. Rolling resistance doesn't care whether there's a wheel in front of you, and it's 10-15% of your power on a typical road bike. So the drag saving gets diluted before it reaches your legs.

Watts by position#

Here's a full eight-rider paceline with 300 W on the front. That leader wattage solves to a group speed of 39.5 km/h.

Three things worth pulling out of that table.

Second wheel is the worst place to hide. It's still a 63 W saving, but you're paying 237 W where the rider three places back pays 208 W. If you've just finished a pull and swung off, dropping further down the line is worth about 30 W of recovery.

The savings peak in the middle instead of the back. Positions 5 and 6 are the best seats. That's the well-documented middle-of-the-peloton effect: you're shielded by the riders in front and getting a small assist from the low-pressure region of the rider behind you.

The last rider gives some back. Position 8 needs 223 W, more than positions 4 through 7, because there's nobody behind them to supply that trailing assist. The model deducts two percentage points from the last rider in any group of five or more.

Why speed changes everything#

Aerodynamic power scales with the cube of speed, while rolling resistance scales linearly, so the faster you go, the more of your power is aero, and the more a given drag reduction is worth.

Here's the same five-rider paceline, across the calculator's full range:

Note what does and doesn't move. The percentage saved barely budges: 19% to 22% on second wheel across a 17 km/h spread. The watts saved nearly quadruple, from 29 W to 111 W.

That's the practical takeaway. Drafting is worth having on a slow social ride, but it's worth enormously more in a fast chain gang, because you're saving a bigger slice of a much bigger number. It's also why getting dropped from a fast group is so much more punishing than losing a slow one: at 47 km/h you've just lost 152 W of assistance.

The best seat depends on how many of you there are#

The peak-savings position moves as the group grows, because the last-rider penalty keeps landing on whoever is on the back.

Below five riders, the back of the line is the best seat and there's no penalty worth worrying about. From six up, the sweet spot settles at position 5 and sitting on the very back starts costing you.

What the rider on the front actually gets#

The leader isn't quite as exposed as a genuinely solo rider. When someone sits on your wheel, the low-pressure region between the two of you gives you a small push, worth about 3% of your drag.

It's a real effect but a small one. At 300 W with four riders behind, the group moves at 39.5 km/h. A true solo rider would need 308 W to hold that same speed. So leading a paceline is worth about 8 W compared with riding alone.

Eight watts is not why you take a turn on the front. But it does mean the leader's number in every table above is very slightly flattered relative to true solo riding, which is worth knowing if you're comparing group data against a solo effort.

Where the model stops#

The figures come from Blocken et al. (2018), a CFD and wind-tunnel study of a 121-rider peloton, sanity-checked against on-road paceline field data. The power model assumes a flat road, steady speed, still air, roughly a metre of wheel-to-wheel gap, and a representative 80 kg rider with CdA ≈ 0.32 m² and Crr ≈ 0.005.

Four places it stops being useful:

Climbs. On a sustained gradient most of your power goes to gravity, not air, so drafting saves much less. van Druenen and Blocken (2021) measured about 7% saved at 6 m/s on a 7.5% gradient, against 25-30% on the flat at the same speed. Don't use these numbers on a climb.

Crosswinds. Once the wind is off to one side, groups form echelons and the aerodynamics change completely. A rider sitting directly behind the wheel in front in a strong crosswind gets far less shelter than the tables suggest.

Big gaps. These savings assume you're close. Open up to three metres or more and you lose over half the benefit.

Team time trials. The single-file numbers are the foundation, but dedicated TTT formations do better. A four-rider diamond can drop the protected rider's drag to around 38% of solo, well beyond anything in a single line.

If you need gradient, wind and your own CdA as actual inputs, that's a different tool: my bike split calculator takes a GPX file and models a whole course, and I've written about how it works and how well it holds up against real rides.

Try it on your own group#

Plug in your own numbers and see where you sit: the cycling drafting power calculator takes the leader's watts and the group size and gives you every position in the line. You can share a specific setup straight from the URL, so it's easy to settle an argument about who's actually working hardest.

Two things I'd use it for. Before a chain gang, work out what you can hold on the front without blowing up before the rotation comes round. And after a group ride, compare your average power against what the tool says your usual position should have cost, which tells you fairly bluntly whether you were sitting in or doing work.

If you're curious what a given wattage means in a particular gear, the gear calculator covers the cadence side of the same question.