How Are Climbs Categorised?

The Tour de France has completed three stages in the Vosges, visits the Jura range tomorrow before the Alps and then the Pyrennes. The race rates climbs with five labels, from 4th category for the easiest all the way to 1st category and then HC for hors catégorie, or “beyond categorisation”. A frequently asked question is how are these categories determined.

The categories serve two purposes:

  • To illustrate the relative difficulty of the climb so riders, the media and spectators get a quick take on the severity of the ascension
  • To set the number of the points awarded for the mountains competition

Contrary to what some think,there’s no algorithm to compute gradient and length into a category, instead it is arbitrary. There is a formula used where the square of the vertical gain is multiplied by the length to give an indication of difficulty but this is for comparison rather than classification. A small climb in the north of France is always going to be fourth category, a 2,500m mountain pass is always HC but you can fill in the gaps in between using common sense. Often in a mountain stage there can be roads in-between the main climbs that rise more than many fourth category climbs of the type selected for the flatter stages in the north of France but they’re not categorised so as to avoid cluttering the route.

Similarly a climb’s category can sometimes vary. A first category climb one year could be a second category in another. Of course the road has not changed, just its importance to the race. ASO might want to adjust the points on offer to tweak the incentives. Or it could help to change the story of the stage, a climb scaled early in the stage could be demoted down for the day so that we don’t think the hardest climb appears early in the stage.

The Name
As touched on before, the Tour often makes its own names. We saw the “Côte d’Oughtibridge” in England which was really Jawbone Hill to the locals. But this wasn’t because ASO was exploring unknown lands, many climbs in France have different names and ASO has a tendency to them as it wishes.

Take tomorrow’s stage which lists three côtes or hills in a row with the Côte de Désertin as the third, it’s only a fourth category climb but is locally known as the Col de la Croix de la Serra, or col meaning a mountain pass. Perhaps the col label is scary and ASO prefer three smaller climbs rather than a gradual mountain pass?

The race will often place the King of the Mountain (KoM) point where it wishes, often it’s obvious – for example on the Col du Tourmalet the “summit” is obvious as the road rises to the pass and drops immediately. But this isn’t always the case, when Stage 8 of this year’s Tour passed the KoM point for the Col de Grosse Pierre, the actual pass and the official black sign that’s there all year was a couple of kilometres further along the route. Why? Well because the race wanted to mark the top of the steep section with the KoM points and the pass itself was further along a rolling road, a less obvious position to award points to the climbers.

The Col
We often talk about the summit of a mountain pass but it’s oxymoronic. A col or mountain pass is not the peak of a mountain but usually marks the lowest route across a mountain range, an easier way of passing from one side to the other. Summit sounds better.

Turn it up to HC
Talking of linguistic inflation, humans have discovered distant planets and classified hundreds of thousands of species of animals but there are some climbs that are apparently beyond categorisation for the organiser of the world’s biggest bike race. Rather than having climbs rated first, second, third, fourth and fifth category, we have the hors catégorie label, as in “beyond classification”. It doesn’t mean the organisers have yet to get round to applying a label, it is hype – Spinal Tap anyone? – in order to make these big climbs sound even more impressive. But these roads are indeed special places and whether you’re a tourist or Vincenzo Nibali, they deserve a unique respect.

King of the Mountains
The best climber was first recognised in 1933, prizes were given from 1934, and the jersey was introduced in 1975 with the polka dots reflecting the sponsor, Poulain, a chocolate maker. But the phrase “King of the Mountains” comes from 1905 when René Pottier climbed the Grand Ballon. He was riding for a team with Peugeot bikes and the company’s factory was down the road – it still is – and workers lined the climb. The spectacle was so great that newspaper L’Auto proclaimed Pottier as “le roi de la montagne“, or the king of the mountain and the label’s stuck ever since.


  • High point of 2014 Tour: Col d’Izoard, 2,360m above sea level. The Souvenir Henri Desgrange prize is awarded the first rider across.
  • Highest col ever: Col d’Iseran, on the route eight times but scratched in 1996 due to bad weather. Last crossed in 2007.
  • Highest finish of 2014 Tour: Risoul, 1855m. This year’s race is a lot of climbing but does not venture often beyond 2,000m.
  • Highest finish ever: Col du Galibier for 2011 at 2,645m
  • Col used most often: Col du Tourmalet, first climbed in 1910 and the Tour has ridden up it 79 times including twice in 1972 on the same stage and twice in 2010.

There are no rules. The categories are applied according to common sense, the relative importance of the climb. The organisers have changed a climb’s status from year to year to suit the race.

43 thoughts on “How Are Climbs Categorised?”

      • First law of modern cycling broadcasting: “According to Carlton Kirby” introduces a phrase which should be taken with very liberal amounts of salt.

        • are you sure? because yesterday, according to carlton kirby, wide set eyes means you’re a “thinking man”. how can this not be true??

          honestly, some days he’s sillier than phil or paul.

    • Of course. Think about it, a long climb with a gentle gradient of 5% could probably be driven in 3rd gear so the Lautaret for example is a third category climb; while the Koppenberg would be 1st category because you need first. On a wet day it’d be HC because the 2CV probably wouldn’t get up 😉

  1. Quality as always! One possible typo: “The Tour de France has completed three stages in the Alps, visits the Jura range tomorrow before the Alps and then the Pyrennes” My French geography isn’t great, but shouldn’t that be “three stages in the Vosges, visits the Jura range…”

    • have they actually gone up to the Cime though?? From memory they usually go over at the Col de la Bonnette which is at 2715 (Iseran pips that at 2770) – the Cime is the little loop they added onto the mountain to make it the “highest road in Europe”

      • They have indeed gone up to the Cime, on stage 16 of the 2008 Tour. John Lee Augustyn led over the top of the Cime but crashed on the descent lower down. The nomenclature is a bit confusing, INRNG. Some maps name it the Col de la Bonette-Restefond, but the road sign, which as Craig says, indicates the Col de la Bonette, at 2715m is at the point where the two sides of the loop that goes up to the Cime de la Bonette meet. According to the IGN geoportal maps, the Col de Restefond is lower down, and off on a dirt road.

  2. A fascinating issue, indeed. One mathemathical way to decide the category might be the vertical gain (difference in altitude from the bottom to the top of the climb), but of course both points can be put wherever the organisers want. Besides, some climbs have downhill sections, what should we do with them? Not one “big” climb but two “smaller” ones?
    And another different subject is how many points to how many riders (scoring system), which can also be shaped to favour some kinds of riders above others. The current system favours GC riders, while the one existing some years ago favoured breakaway cyclists.

    • Good point(s). One of the great things about the tour and pro cycling in general is how ridiculously arbitrary (at times downright contrived) some of the details are.

  3. The first line: ‘The Tour de France has completed three stages in the Alps.’ Should be the Vosges.

    I know the Alps tend to dominate the Tour….

  4. So, is there one ASO official with “buck stops here” responsibility for classification? Purely out of curiosity, have there been examples when the classification of a climb has changed from one year to another?

    Great work as usual. thanks

  5. Is Col de la Bonette not the highest col ever? I know the stage never finished there but it came over it just shy of the “Cima”…

  6. The 2860m height for the Cime de la Bonette is the mountain peak, which is a short scramble from the road (definitely a scramble in cleats!). The actual high point on the road – which is a little loop away from the pass – is 2802m, still higher than the Iseran, but not a true pass. It remains the highest ever point in the Tour de France, and from memory was climbed in 1962, 1993 and 2007. I’ve always suspected the loop road was engineered simply to get it over 2800m.

    The col de Restefond is 2715m.


  7. I’ve always thought there was a whole lot of estimate and hand waving with these ratings. Good to know I was right. An algorithm does have some important short comings. Factors that are important include road surface and condition, number and quality of switchbacks (i.e. off-camber), exposure, road width, historical significance (i.e. alpe d’huez), maybe even climbs on previous stages. The art part of categorizing climbs can be reduced if these and other factors are included.

    • If hitting each cobble dissipates energy equivalent to a couple of vertical centimetres and one meets one hundred thousand cobbles in 15 km of pavé, it adds up to 2000 metres – a HC-class climb, perhaps?

  8. The Col de la Madeleine is regularly rated 1st or HC depending from which side it’s climbed. There’s a downhill section on one side for a few kms reducing the perceived severity of the climb …

  9. There are definitely some rules about the categorization of the climbs.

    This plot shows the correlation between the length of the climb and the average gradient. These are the two main factors for the categorization. The curves are based on the squared gradient criterion (L*G² where D is the length of the climb in kilometers and G the average gradient in %).

    However, other factors exist, like:
    – Altitude (high altitude = less oxygen = harder to climb)
    – Position of the climb in the stage (some mountain top finishes are often given HC if they are on the 1C/HC limit – just take the Pla d’Adet for example while the col d’Agnes is 1C with roughly the same stats, 10 km @ 8.2 %)
    – Maximum gradients (Chevrères is 1C thanks to it part with 18% gradient)
    – Prestige (many people consider that the Galibier south-west shouldn’t be HC – and you can see it on the graph in the middle of the 1C zone, but it’s the Galibier nonetheless)

Comments are closed.