Take the 2-minute tour ×
Bicycles Stack Exchange is a question and answer site for people who build and repair bicycles, people who train cycling, or commute on bicycles. It's 100% free, no registration required.

Watching the Tour de France on TV I notice that the riders like to ride close behind other riders when going up a steep hill.

I understand that rider can benefit from a slipstream on the flat section, but on a climb the speed is so slow, the slipstream has minimal effect.

When climbing, the TV commentators make a big deal of the importance of riders 'getting on to the wheel' of the riders ahead as if it will be much easier if the rider could attach himself to the group ahead.

What is the benefit of being attached to a group when climbing?

share|improve this question
Spite, maybe? I've noticed that hill climbs are a bit nicer for me when I'm not leading, but I don't race. Interesting question. –  WTHarper Jul 15 '12 at 15:53
Keep in mind that they're still often going up that hill at 20mph -- far faster than mortal humans would do in the same circumstances. Wind resistance is still a major factor. –  Daniel R Hicks Jul 15 '12 at 18:42
Uh, stupid question, but could it just be because they go slower uphill? If you have a bunch of racers together and they have to slow down, they'll naturally group closer together: the time between each racer will be the same, but the distances will be smaller simply because they're moving slower. You see the same thing in, for example, Nascar races when they turn corners. –  BlueRaja - Danny Pflughoeft Jul 16 '12 at 4:07

8 Answers 8

up vote 55 down vote accepted

I have raced Cat 1/2 for a number of years (elite road cycling) and I can unequivocally say the many here have no clue what they are talking about. On a climb "getting on the wheel" is rarely about slipstreaming (unless there is a severe headwind), instead its about positioning, pacing and psychology.

Position and Pacing

Rides who attack or lead out the climb are strong and will likely be climbing near their aerobic threshold. If you don't get on immediately, a gap will form. If you want to catch up (close the gap) you will have to work hard likely putting in an anaerobic effort. This is not good as it will reduce your capacity to respond to future attacks (you only want to go anaerobic or near anaerobic when you absolutely have to). Then best analogy is you are like book of matches, you get so many strikes so use them wisely. If you immediate latch onto the wheel it is your best shot at keeping your pace aerobic, therefore conserving a match.

Conversely, lets suppose you say "%^#* it I'll let a gap form." You continue climbing eventually matching the pace so that the gap doesn't get too big. When you reach the top, you will have put in virtually the same effort, but they will have started on the downhill before you reach the top. The gap will now get even wider and all of a sudden you are screwed. You are at speed and now you have an even bigger gap to close plus you now have the added disadvantage that slipstreaming has suddenly become important again. Facing this you have two real choices, cross your fingers and hope the peloton slows (which it may - the peloton is a fickle beast), OR bust your ass and close that damn gap. The rub is if you do close the gap you put in twice the effort you would have if you had simply got on the damn wheel when the climbing started.

Road racing is like chess. You need to plan at least five moves ahead, but unlike chess there are virtually no restriction to the type of moves you can make and there are 150 simultaneous players.


Some here seem to believe there is no psychology involved with getting on someone's wheels on a climb. WRONG.

If someone is climbing hard (at a pace you find hard to match) the best thing you can do is sit on their wheel as close as possible (we are talking 2 inches gap). You just stare at that frackin' wheel and you DO NOT let that gap widen. You tell yourself that you don't care how much it hurts, you will not let the gap widen. You then keep repeating this to yourself and before you know it the climb is over.

In all seriousness, when you are right on your aerobic threshold you have to nail pacing. Riding really close lets you judge when you are slacking off. It is much easier to work to pick up an inch or two that you lost when your concentration waned, than it is to be farther away from the rider and realize he/she has put 5-10 feet on you. When everyone is at a similar fitness level you need to fight for every bloody inch you can get. Focusing on something like the wheel in front helps you not focus on the pain you are currently enduring.

Aside - road racing questions are fun, someone should ask about fun strategies such as guttering the peloton or rolling attacks.

share|improve this answer
Ah, you're right: I downplayed the psychological aspect but I'll amend my answer. +1 to you. –  R. Chung Jul 16 '12 at 7:10
Good answer. +1 –  Sam Meldrum Jul 16 '12 at 8:05
You must know something that Phil Ligget doesn't know. In his book - Tour de France for Dummies, he specifically mentions the importance of the guys who pull, especially on mountain stages, because of the benefits of drafting. –  Randy Minder Jul 16 '12 at 15:34
"I can unequivocally say the many here have no clue what they are talking about" Ok, but have you thought about the numbers? A savings of even a few watts, even from a tiny draft, can make a big difference in an elite race. –  David James Jul 16 '12 at 22:05
"someone should ask about fun strategies such as guttering the peloton or rolling attacks" - you know it's legitimate to ask a question and answer it yourself? –  armb Aug 19 '13 at 17:55

The physics of the answer are actually pretty well known, and it doesn't require any psychological explanation (there are psychological reasons but the physical reasons suffice; the psychological reasons are in addition). You're right that the speeds are slower while climbing so the absolute benefit of drafting is less. However, gaps are always a problem whether climbing, on the flat, or descending and as long as you are riding in the atmosphere and not in a vacuum there is some benefit to drafting. When climbing (as opposed to riding on the flat or descending) small differences in power make a consequential difference in speed. That's because the power needed to overcome aerodynamic drag varies approximately with the cube of speed while the power needed to overcome gravity drag varies only with the speed itself. That means, ceteris paribus, on a steep hill a 5% difference in power means almost a 5% difference in speed while on the flat a 5% difference in power only means ~1.7% difference in speed. So, although the aerodynamic benefit of drafting on a hill is small, it's the only game in town. That is why a team leader will follow a domestique even if the leader is more powerful -- not because leaders are attempting to attack their own domestiques, or because domestiques necessarily ride more steadily than the leader but because the domestique offers some (small) advantage to the leader. How much advantage? At the speeds you see in the WorldTour even up steep climbs, a pro rider can still receive a "savings" in power of between 5 to 10 watts by following someone else (either his own domestique or a rider from another team). This may not sound like much but if you are right on the cusp, it can be the difference between hanging on and getting dropped -- and, as the answer above shows, getting dropped can have disastrous consequences.

To a lesser extent, but by the same physics, a strong tailwind can shatter the peloton in the same way that a steep hill can, while a strong headwind tends to keep the peloton compact. You probably understand that when there is a strong headwind, drafting becomes more beneficial. Effectively, a headwind penalizes the rider in front a lot and therefore benefits the riders behind a lot. The opposite is also true: when there is a tailwind, the benefit of drafting decreases so small differences in power between riders become more apparent. The tailwind makes everyone go faster, but it exacerbates power differences between riders, so strong smart riders will never attack in a headwind but will attack on a hill since that's where their advantage is greatest.

Conversely, the combination of a tailwind and hill can be deadly for team time trials. For example, in the 2005 Tour de France, the stage 4 team time trial between Tours and Blois in the Loire Valley took place on a day with a strong tailwind on a course that would normally be considered relatively flat. Speeds were high because of the tailwind but at a slight hill about two-thirds of the way along the course riders were getting popped off. The tactics of team time trials demand that the best teams moderate their effort on hills or with tailwinds in order to prevent losing too many of their riders.

Crosswinds can also shatter the peloton because they can reduce the drafter's benefit relative to those in front.

share|improve this answer
I hate tailwinds on hills because your cooling disappears. –  Daniel R Hicks Jul 16 '12 at 0:54

I've looked around a bit for empirical data on this, but can't find much. My opinion and guesses on this:

  1. If there is any headwind, like in normal drafting, the front rider takes the brunt of it.
  2. The psychological "I'm still in the pack" effect. I can push myself to stay with a pack's pace, but once I drop behind it, it's much harder to keep going at the same pace and the gap quickly widens.
  3. Following the leader takes less mental power. Mental energy is very significant if often overlooked. Imagine being tired and having to drive home on a mix of winding roads, lots of turns, and some rural highways. It's much easier if you following someone going to the same place then if you have to make all the decisions. Likewise going up a hard grueling hill. You can just follow the line of the leader.

Happy Riding.

share|improve this answer
Also, perhaps bing close to a leading opponent makes easier to catch it if he sprints, or overtake it by surprise if he gets distracted for a moment. –  heltonbiker Jul 15 '12 at 18:40
In my experience racing, following a leader up a hill, who is compatible and/or friendly to your goals, is easier. But that goes out the window if the rider has a different style (e.g. prefers to go steady when you like to accelerate or vice versa) or is on a different team. –  David James Jul 16 '12 at 22:11

In my limited experience it's not really about slipstreaming, it's about the physical distance between riders. As heltonbiker suggested above, on a hill the leading ridser is above the rest as well as ahead, and that extra height gives them an advantage if they attack. And getting past them means not just the usual power to speed up, power to break the wind, plus manoeuvring to keep the overtaken rider off their tail. As well, they have to climb the hill. But if the lead rider attacks, they get the advantage that every metre they are ahead is also a fraction of a metre above the chasing riders.

There's also the psychology of it, where on a hill everyone expects riders to be working hard, attacking steep sections and occasionally getting off the saddle to put power in. So it's harder to see the break coming, you can't just go "sudden increase in power, they're making a break".

Math break... I'm curious about the power involved. I know that for the speed records at Battle Mountain riders are getting as much power from the slope as from their legs, so I wonder how the hills on the tour compare. Assume 10m/s ground speed (36km/hr) on a 1 in 5 slope, which means 2m/s upward velocity. For a 100kg combination (heavy, but plausible) that's 200W going into the climb, of the 300-400W that the lead rider would normally put out. So on the steepest part of the climb half the riders power is going into the hill rather than into the air.

share|improve this answer
The maximum slope on the test section at Battle Mountain is -.0067 (i.e., 2/3rds of 1 percent). When the Varna Diablo set the record at slightly above 37 m/s, Sam Whittingham was putting out slightly above 500 watts. The "benefit" from the slope would have been svm*g where s= -0.0067, v=37, m=90, and g=9.8, or about 200 watts (i.e., if the course had been dead flat an additional 200 watts would have been needed). –  R. Chung Jul 16 '12 at 6:18
According to Sam his average would have been about 250W and the Diablo weighed significantly more than you've allowed (Sam alone weighed about 90kg IIRC). The slope during his run-up is also 1.5° not .5°. Suffice to say that there was never any prospect of running the course in the other direction. That's why human powered records are held on the flat at low altitude. –  Kohi Jul 16 '12 at 8:36
The IHPVA rules say that the maximum slope over the run-up cannot exceed 2/3rd of one percent (see section 3.3.1 here). I believe Sam weighed closer to 70kg, and he has posted an SRM data file which shows his power during one of the record attempts. You can see his speed and power here. –  R. Chung Jul 16 '12 at 11:49

The other answers I've seen so far offer some general insights. Here, I want to share some sources that go deeper. They discuss equations which help better understand the quantitative relationships between power, speed, air resistance, drafting, and so on.

To get warmed up, check out this interesting but accessible article by Marilyn Trout: Relationship Between Drafting and Climbing.

Trout quotes Cycling Uphill and Downhill by David Swain at length. For example: "At very slow speeds (on the order of 16 km/hr or less) air resistance is negligible, and drafting becomes nearly meaningless."

But why is the "magic" number 16 km/hr? Let me say: there is no single magic threshold. To figure out that threshold, you have to define, first, what percentage is "small enough" for your question at hand. For example, if you were to ask, at what point does air resistance stop contributing more than 0.5% of your total power output, an equation can give you an answer.

At the elite levels, I would make a rough guess that as little as 1 Watt over the course of a long climb could make the difference between victory and second place. My point is this: don't assume drafting is negligible until you have done your homework on what "negligible" means.

And how do you do your homework? Asking here for thoughts on hill-climbing and drafting is one place to start. But if you want to get a better understanding, ask for references and then read scientific papers. Inside you will find studies and equations.

Trout mentions this equation from Swain, who cites its source as Equation of motion of a cyclist by P. E. di Prampero, G. Cortili, P. Mognoni, and F. Saibene.

W = (kr M s) + (ka A s v^2) + (g i M s)


  • W is power
  • kr is the rolling resistance coefficient
  • M is the combined mass of cyclist and bicycle
  • s is the bicycle speed on the road
  • ka is the air resistance coefficient
  • A is the combined frontal area of cyclist and bicycle
  • v is the bicycle speed through the air (i.e. road speed plus head wind speed)
  • g is the gravitational acceleration constant
  • i is the road incline (grade; however, this is only an approximation, as the sine of the road angle to the horizontal should technically be used)

If you want to dig into the aerodynamic benefits of drafting, I would recommend checking out The understanding and development of cycling aerodynamics by Lukes, Chin, and Haake. In particular, check out the section on drafting on page 67.

Drafting behind a single rider with a 0.2 to 0.5 m gap was found to reduce oxygen consumption by 18 ± 11% at 32 km/h and 27 ± 8% at 37 km/h and 40 km/h.

Drafting behind one, two and four riders resulted in the same oxygen consumption reduction at 40 km/h (27 ± 7%).

share|improve this answer
I think you're missing a g in the first term and a rho/2 in the second. –  R. Chung Jul 16 '12 at 22:12
The obvious should be noted: When a mortal human is going uphill at 10mph (16kph) he's maybe working a 5% slope (pulling a number out of the air). The TDF rider, on the other hand, is probably doing a 10% slope or better at that same speed. And I'd guess that the energy expenditure for climbing a 10% slope is probably 3x that for a 5% slope, ignoring wind resistance. That means that the relative advantage of drafting (% saved over total energy expenditure) is less for the TDF rider than for the mortal rider AT A GIVEN RATE OF SPEED. But at a given slope it's the opposite. –  Daniel R Hicks Jul 16 '12 at 22:16
Stage 10 of this year's Tour from Macon to Belgarde sur Valserine went over the 17km long 7.1% Col du Grand Colombier. Chris Anker Sorensen, a climber for Saxo Bank, went up this HC climb at 330 watts and 20.9 km/h. –  R. Chung Jul 16 '12 at 22:34
@DanielRHicks I think you are pulling too many numbers out of the air. A rider, at his threshold, by definition, is going to be putting out (more or less) the same wattage no matter what the grade is. –  David James Jul 16 '12 at 22:37
@DavidJames-- Right. But, depending on your speed, the energy is split differently between climbing and overcoming wind resistance. The stronger rider will be using a larger fraction of his energy to overcome wind resistance, because he's going faster. –  Daniel R Hicks Jul 17 '12 at 0:42

Just to add some info on the physics side of the equation (and there is definitely psychology at play as well, as motivation plays a part in how hard you can ride).

There are plenty of references regarding how much the air resistance is reduced by following close behind a rider, plus anyone with a power meter will likely have seen similar data for themselves.

The amount the air drag is reduced does vary of course depending on a range of factors, but a typical reduction of ~30% when following close behind the rider in front is about right. It can of course be more or less than that. This reduction applies on climbs, flats and descents, but of course only applies to the proportion of power demand used to overcome air resistance.

So looking at the relative energy demand from the various resistance forces, I did this sample chart to show how that changes with gradient, for a rider + bike of 75kg and a steady power output of 300W and no wind (using the equations as described in the paper by Martin et al, Validation of a mathematical model for road cycling power):

enter image description here

Obviously the exact values for any individual will vary depending on their mass, aerodynamics, rolling resistance factors and so on. This is just to explain the principles involved - IOW the overall shape and trends shown will be the same for everyone.

So what we can see, e.g., is for such a rider on a 1% slope, a little over 60% of their energy output is used for overcoming air drag (300W x 61% = 183W), while on a 6% slope, that proportion of energy output drops to only ~10% (30W) as much more of their power is used to overcome the force of gravity.

Now the wattage "saving" by drafting closely behind another rider at those speeds, would be approximately 30% of the power used to overcome air resistance.

On the 1% gradient, that's ~30% x 183W = 50-60W, while on the 6% slope, that saving drops to 3-4W.

Now of course the fitter or more powerful the riders are, the faster they travel up hills at any given gradient, and so those relative wattage "savings" at each gradient from drafting behind another rider will increase.

As an example, if a rider with same weight etc was doing 400W on the 6% slope, the speed increases and the proportion of power used to overcome air resistance also goes up to 15-16% of total demand, in this case 62W, and ~30% of that = 15-20W.

A 15-20W saving when you are at your limit is substantial.

On an 8% grade this more powerful rider can attain a ~10W saving by drafting, and on a 10% slope they can still achieve a 5-7W saving.

Even 5-10W can be the difference between hanging on or cracking.

share|improve this answer
I would provide the alternate view that that 5-10W represents about 1.25-2.5% of the total output at 400 W. While you will take anything you can get, it is unlikely that this will be the difference between hanging and cracking. For example if the peloton surged on the climb and need 1000 W to hang on to the surge, you are now down to 0.5-1% savings. Likely other factors will determine if you stay in or are spit out like a watermelon seed. –  Rider_X Aug 26 '14 at 18:39
Such 5-10W changes are hardly noticeable when you are riding within yourself, however when you are on your limit, then even small changes make a big difference to your ability to go harder and recover. Recovery from any supra threshold surge, once the pace drops back as it eventually will, requires you to ride under threshold, even if only marginally, in order to recover the accumulated O2 debt. If you are riding at threshold after a surge then you are unable to recover the O2 debt and are ultimately forced to slow, or will be unable to manage another surge. –  Alex Simmons Aug 26 '14 at 22:55

In my opinion its purely psychological, unless the grade is less than 6% and theres a headwind. Sitting in the wheel and dragging yourself to the ragged edge is something us taller guys are used to, knowing we'll crush the small guys in the TT. Saying that you can sometimes drop back a few metres, just to let the Heart Rate drop that 2/3 BPM that it takes to recover, jump back on the wheel and even attack, (attacking when you are both can sometimes throw you over the edge but conversely the other guy/guys might have been o the limit and don't have it to follow, now you both feel the same but you have a gap, otherwise fair play to catching mate!!) although that takes a huge aerobic engine to recover in 20 seconds from the limit, then to think of attacking its a tactic that feels great looking back on it after. Usually, sticking to the wheel is the best, and if your climbing worse because your a KG or 2 heavier, remember you can attack on the Descent!

Number 1 rule lads, get those gels in 7/8km before the climb & on the descents, the rest will take care of itself!

share|improve this answer

Time gap at crest = time gap when second rider crests, even though distance between riders will have increased because first rider began descent before second rider crested. Therefore, no time advantage gained. 'Elite' racer's answer incorrect on this point, basic physics. Holding the wheel avoids the gap and allows effective overtaking draft on descent, slingshot.

share|improve this answer
Welcome to Bicycles. The internet doesn't have space limitations, so please write complete English sentences. This will make your posts easier for others to understand your points. Your answer should also stand on it's own - to comment on another post comment below it when you have enough rep. –  andy256 Apr 30 at 23:15

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.