You’ve almost certainly heard about Everesting by now. To ‘everest’ is to climb the height of the world’s highest peak with repeats of the same hill, riding up and down, over and over again. It’s a challenge that was born roughly a decade ago and has since been completed by tens of thousands of hardy riders around the world.
We know a bit about Everesting here at Escape. The founder of the challenge, Andy van Bergen, is Escape’s membership manager. Your present author wrote an entire book about the creation and growth of Everesting, and for more than three years now, Escape’s own Ronan Mc Laughlin has held the world record for the fastest-ever Everesting with a ridiculously quick 6 hours 40 minutes and 54 seconds (most riders take around 20 hours).
That record Everesting, set by Ronan on the Mamore Gap climb in Ireland, forms the centrepiece of a new research paper about the physics of Everesting. By breaking down Ronan’s incredible effort, the paper’s author – East Carolina University physicist Martin Bier – tries to analyse the impact of wind, gradient and more on how long an Everesting might take.
Ronan’s Everesting
To start with, let’s paint a picture of Ronan’s record-breaking Everesting. The Mamore Gap segment that Ronan used climbs 117 vertical metres over the space of 810 metres – a brutal average gradient of 14.2%. The bike Ronan used for the effort was highly tuned for its exact purpose – weighing just 5.5 kg, the bike had a cassette of just seven cogs, an aero fairing, and a bunch of other modifications to make it both as light and as aerodynamic as possible.
Ronan had set the Everesting world record on the same climb in 2020 (7 hours 4 minutes and 41 seconds), breaking a record set just a few weeks earlier by multiple-time Grand Tour winner Alberto Contador. When he came back to Mamore Gap in March 2021 for another attempt, Ronan went 24 minutes faster than his previous best, taking his record back from American semi-pro Sean Gardner.
On the day of his new record – which, again, still stands to this day – Ronan had a tailwind of around 20 km/h (12 mph) while climbing. At first blush it seems like that tailwind must have helped him to slash the record. But did it really?
Making the model
To answer questions about Ronan’s Everesting – like whether the tailwind helped – Bier employes a common technique in the applied sciences: he uses mathematical equations to build a theoretical model to represent Ronan’s ride.
“The equations are a kind of logical structure,” Bier tells Escape via email. “Next we have the input data. That is here the slope, the rider’s mass, the rider’s power output, the length of the road, the coefficient of friction, etc. You put these input data into the equations/model and they lead, after some computation, to predictions for the other data.”
If you’ve got a maths or physics background, or you just want to go super-deep on the subject, there’s plenty to digest in Bier’s paper. But we don’t need to understand every single formula to make sense of what’s going on.
Bier starts by laying out the main forces a rider needs to overcome when climbing a hill by bike:
- Wind resistance, which increases exponentially as riding speed increases.
- The gravitational force, which increases with the mass of the rider and the bike, and with the gradient of the hill they’re riding up.
- The rolling resistance of the tyres on tarmac.
Bier then focuses on the uphill component of an Everesting; to help construct and validate his mathematical model. To do so, he makes a few assumptions, including about the pace that Ronan climbed at.
Bier assumes that as a semi-professional bike rider, Ronan’s functional threshold power (FTP) – the most he could sustain for an hour – is around 5 W/kg – an assumption based on oft-referenced power output tables from exercise physiologist Andy Coggan. At Ronan’s then weight of 67 kg, that works out to an FTP of 335 W. Bier makes the assumption that this was Ronan’s “estimated power generation on the climbs”.
You might be wondering why Bier didn’t just access the Strava file from Ronan’s record Everesting, and input real power data into his model. Bier admits to Escape that he “should have done that”, but adds that “power meters, as you may know, are generally only about (+ or -) 2% accurate. I just made an estimate that led to a good fit of the ride.”
And Bier’s assumption is relatively close: we can see from Ronan’s Strava file that he averaged 327 watts per climb throughout the day, vs the 335 W that Bier assumed. And Bier’s assumption about Ronan’s climbing power gets him pretty close when calculating Ronan’s climbing speed too (from a rather simple formula that factors in power output, weight, the impact of gravity, and the gradient of the climb). Bier arrives at a pace of 12 km/h (7.2 mph) – not far off Ronan’s real average climbing speed of 11.2 km/h (6.96 mph).
The second assumption that Bier makes is that while climbing such a steep grade, the effect of wind resistance is basically negligible. He runs the numbers too: in order to overcome wind resistance at that pace of 12 km/h, Ronan would need to produce 6.5 W, or just 2% of the estimated 335 W he produced per climb.
With an estimated climbing speed of 12 km/h worked out, Bier is then able to calculate an estimated time per ascent: 4 minutes 10 seconds. A quick glance at Ronan’s Strava file shows that this result is also close to the mark.
Bier then looks at the downhill portion of the Everesting. He points out that while descending a constant slope, a rider’s speed will approach a terminal velocity. This is the point where the gravitational pull on the rider, and the aerodynamic resistance slowing them down, are balanced out. The rider is no longer able to accelerate any further and their speed will remain constant.
Knowing the mass of Ronan and his bike, the gradient of the hill, and by making an assumption about Ronan’s coefficient of friction, Bier can use another formula to estimate a terminal velocity of 86 km/h. From there he works out how long it would take to descend in total, given it takes time to reach that terminal velocity on the descent. In this specific case, he calculates 10 seconds to get up to speed, meaning that, in windless conditions, one theoretical descent takes around 46 seconds.
With the effects of gravity (up and downhill) factored in, Bier then turns his attention to rolling resistance. In an effort like Ronan’s Everesting, where high gradients and a fast descent are providing most of the resistance (gravity and wind), rolling resistance is only a minor impediment. By assuming a coefficient of rolling resistance of 0.006 for Ronan’s tyres Bier calculates a 4% decrease in climbing speed and 2% decrease in terminal descending velocity as a result of rolling resistance.
A quick note about the mathematical model that Bier is building: it’s never going to be 100% accurate. The angle of the slope being climbed and descended, for example, isn’t a consistent 14.2% gradient throughout, which will have an impact. Similarly, the extent to which the rider is able to stay aero is “likely to vary in the course of a lap as the rider takes on an aerodynamic tuck during the descent and does not do so on the ascent.” As a result, the static value that Bier uses to represent Ronan’s aero-ness (a factor of air density, drag coefficient, and frontal area) isn’t static at all.
Another factor along the same lines: when calculating the amount of time each lap takes, it’s not just a case of climb time (calculated at a constant speed) plus descending time. There’s also time spent turning around at the top and bottom.
Bier’s model suggests that those two turnarounds cancel each other out. The descent doesn’t start from a standstill, making the descent a little quicker in reality than in the model. On the flipside, the descent also takes a bit longer than calculated because of the need to slow down and turn at the bottom.
For what it’s worth, Ronan did have a considered approach to turnarounds during his record ride. “I’d specifically chosen the front chainring and rear sprockets to not just get up the climb but also to also allow me to accelerate at the top turn to get to terminal velocity as quickly as possible,” he explains. “I think [Bier] assumed I was freewheeling and hadn’t considered that initial few seconds of acceleration …”
The wind
In order for Bier’s model to most accurately replicate Ronan’s ride, it needs to factor in the 20 km/h tailwind that Ronan apparently got on the day. In the paper Bier updates his equations to account for the tailwind up the climb (meaning less force required to climb), and for the headwind on the descent (reducing Ronan’s terminal velocity).
With adjustments made, Bier crunches the numbers to calculate estimated climb and descent times. He ends up with numbers even closer than before to what Ronan actually averaged on the day: a climb time of 4 minus 19 seconds, a descent time of 56 seconds, and a total time of 5 minutes 15 seconds per lap. In reality, Ronan averaged 5 minutes 16 seconds per lap (6 hours 40 divided by 76 laps), putting Bier’s model within a second of real life, per lap.
Which brings us to the big question: what impact does wind speed have on total lap time during an Everesting? To answer this question, let’s take a look at a graph from the paper, which shows the impact of wind speed (horizontal axis) on climbing speed (vertical axis), using data from Ronan’s Everesting.
Based on this graph you’d assume that the stronger the tailwind, the faster the Everesting, right? Not quite, because you need to factor in the descent.
Here’s a graph showing the total lap time (one climb plus one descent) as wind speed increases (again using the parameters from Ronan’s Everesting):
As Bier notes, this is quite a surprising finding. As the wind speed increases (horizontal axis), the lap time (vertical axis) decreases, increases, drops again, then climbs again. Note, though, that we’re only talking about variations of a couple of seconds here (more on this in a moment).
So why isn’t this a straight line pointing towards the bottom right of the graph? Why don’t laps just get faster as the tailwind gets stronger? This third and final graph has the answer.
As we’ve already seen, climb times (middle line) decrease as the tailwind on the climb gets stronger. At the same time, a stronger tailwind on the climb means a stronger headwind on the descent which, as shown by the bottom line here, means those descents take longer. The faster climbing and slower descending push and pull against each in varying amounts, affecting the overall lap time (top line).
But here’s the biggest takeaway of that third and final graph: even though the lap time does vary by a few seconds here and there (see the second graph), that’s hardly significant in the context of the lap time (notice how that top line looks almost flat). In fact, for wind speeds of less than 54 km/h (34 mph), “total lap varies by only 2%, which means that the decrease in climbing time is almost cancelled out by the increase in descending time.”
So for the parameters of Ronan’s Everesting, the tailwind seemingly had minimal impact. It certainly doesn’t seem like it was the biggest factor in his 24-minute improvement between his first and second Everesting records.
It sparks an interesting question though: what if the wind was from the side? While Bier’s model assumes Ronan’s tailwind up Mamore Gap was directly behind him, it’s unlikely that was exactly true. Based on Bier’s calculations, it doesn’t matter too much though.
When you break down a cross-tailwind into its tailwind and crosswind components, then feed those back into the model, well, you get much the same answer as if there was no wind at all. Any benefit from a cross-tailwind on the climb is likely to be cancelled out by a reduction in speed during the descent.
What about headwinds (and cross-headwinds) while climbing? Is a reduction in climbing speed worth it for a relative increase in descending speed?
When asked about this, Bier points back to those graphs referenced above. Here’s that third graph again as a reminder:
“Nothing dramatic happens when [wind speed] changes from positive to negative,” Bier explains. “The curves can be readily extended to the left side of the graph. They are not going to make sudden jumps there. A 1.0 m/s (3.6 km/h | 2.2 mph) headwind on the climb will likely again lead to a lap time of just over 300 seconds.” And just as a cross-tailwind up the climb is cancelled out by a cross-headwind on the descent, so the reverse seems to be true.
Of course, all of this assumes a constant wind speed throughout, and that you’re riding on a perfectly straight road like Ronan was. Ultimately though, when summing up the impact of wind on Ronan’s record Everesting, Bier comes to the following conclusion: “changing the Everesting rules to set limits on allowed wind speeds is not warranted by the physics.” The wind just doesn’t make that much of a difference overall, regardless of the direction it’s coming from.
What about gradient?
Few issues have generated more discussion in the Everesting community than the subject of gradient. What sort of gradient should a first-time Everester be aiming for? How much harder are steeper gradients? And most relevant here – and for those trying to Everest as fast as they can – how does gradient affect total time?
To answer this last question, Bier works through a little thought experiment. What happens if you create a theoretical climb with the same vertical gain as Mamore Gap but that is slightly steeper and shorter? By his calculation, an increase in gradient of around 1% (from a 14% gradient, to 14.14%, say) leads to a reduction of around 0.3% in lap time.
So if you want to Everest faster, the steeper the better, right? Yes, but only to a point.
For starters, most riders can’t Everest on climbs as steep as Mamore Gap let alone steeper – pushing against such high gradients for an extended period creates unmanageable fatigue. And besides: “When riding up hills that are steeper than about 15%,” Bier writes, “the mere force that has to be continuously applied to keep rolling and the lack of good balance at low speed become factors that can no longer be neglected. It appears that the top everesters have converged on an optimal steepness between 10% and 15%.”
Given that, if you’re looking to Everest as fast as you can, it’s probably a case of doing the steepest climb you can possibly manage while keeping a steady effort throughout.
In search of optimal
While beating Ronan’s Everesting record is beyond the capabilities of all but a handful of riders on the planet (we’re still waiting for Tadej Pogačar to set an unbreakable new mark), there are a few takeaways from this paper for any rider looking to ride an Everesting as fast as they can.
First up, a statement from Bier about Ronan’s choice of climb, but one that also applies more generally: “The ascent lasts about five times as long as the descent. A 1% improvement on the ascent time will therefore be much more effective than a 1% improvement on the descent time.”
Another consideration for those trying to optimise: it takes time to get to terminal velocity on a descent, and the shorter the descent, the greater percentage of your descent time you spend accelerating to top speed. In the case of Mamore Gap, Bier estimates that Ronan spent 12 seconds per descent getting up to speed. “With a hill that is twice as long as Mamore Gap and an ensuing 38 instead of 76 downhill accelerations, more than seven minutes could in principle be gained.” That time saving is probably reduced by the accelerations Ronan made at the start of each descent, but the general principle still stands.
There’s another trade-off here. Ronan’s laps involved roughly four minutes of climbing and one minute of rest while descending. If his chosen climb was twice as long, as Bier suggests, it might be that Ronan couldn’t sustain the same sort of consistent effort necessary to set a record time.
So where does all of this leave us? Well, if you thought Ronan Mc Laughlin’s Everesting record was due in large part to a sizeable tailwind, you can probably discard that theory. Likewise if you’re thinking that waiting for a day with a stonking tailwind will help you set a faster time for your next Everesting.
Instead, a fast Everesting seems to require a handful of rather obvious ingredients. First: a climb with a very high gradient that’s still manageable at high power many hours into the effort. Second: A straight road so there’s no braking for corners on the descent. And finally, and most importantly: incredible fitness and strength, combined with a light bike and light rider.
“What the control analysis ultimately tells us is that the most intuitive ways toward faster Everesting times, i.e., reducing weight and increasing power, are indeed the most effective ways,” Bier writes in his conclusion. “There are no clever tricks to get around the necessary diet and exercise.”
Did we do a good job with this story?