OK that's an easy one. New Road, Draycott or, as it's known to some folk, Draycott Steep. Reasonable name. But what of the rest?
Well, there's Blagdon, that's hard. And Ebbor Gorge, that's pretty tough. Not forgetting Wrington Hill, short but really steep. And Westfield Road. The Wrangle's got a pretty fierce reputation... The debate rages. But now, perhaps, we can put an end to the debate. All thanks to Codifava's index. I discovered this last week. CI is a method of calculating the difficulty of a climb, a calculation...
Codifava's index = (average gradient squared * length/10) + (4 x average gradient)
Quite how anyone (Codifava presumably) came up with this I don't know. It seems, at first, like a reasonable idea, investigate the average gradient and the length of a climb and it should tell you how hard it is. Except that there's a problem. Some climbs aren't that bad in the main but they have a really steep section that makes them tough. At which point enter the sectioned Codifava index (SCI) which looks at the various sections of a climb, gives each of them a CI and then adds them together to make up the SCI. I think it actually works. I did the numbers for 30 local climbs, mostly from the blog but including a couple of new ones to see how they came out. I reckon it's a believable rank order - see below.
Well, there's Blagdon, that's hard. And Ebbor Gorge, that's pretty tough. Not forgetting Wrington Hill, short but really steep. And Westfield Road. The Wrangle's got a pretty fierce reputation... The debate rages. But now, perhaps, we can put an end to the debate. All thanks to Codifava's index. I discovered this last week. CI is a method of calculating the difficulty of a climb, a calculation...
Codifava's index = (average gradient squared * length/10) + (4 x average gradient)
Quite how anyone (Codifava presumably) came up with this I don't know. It seems, at first, like a reasonable idea, investigate the average gradient and the length of a climb and it should tell you how hard it is. Except that there's a problem. Some climbs aren't that bad in the main but they have a really steep section that makes them tough. At which point enter the sectioned Codifava index (SCI) which looks at the various sections of a climb, gives each of them a CI and then adds them together to make up the SCI. I think it actually works. I did the numbers for 30 local climbs, mostly from the blog but including a couple of new ones to see how they came out. I reckon it's a believable rank order - see below.
| Climb | Sectioned CI |
| New Road, Draycott | 3.16 |
| Wrington Hill | 2.67 |
| Westfield Road, Rodney Stoke | 2.08 |
| Cheston Combe | 2.07 |
| Limeburn Hill | 1.97 |
| Ebbor Lane and Deerleap (Ebbor Gorge) | 1.86 |
| Shipham Road | 1.86 |
| Bleadon Hill from Elborough | 1.84 |
| Blagdon via Two Trees | 1.77 |
| Highfield Lane, Compton Martin (The Wrangle) | 1.69 |
| Old Bristol Road, Wells | 1.67 |
| East Harptree; High Street & Middle Street | 1.60 |
| Cheddar Gorge | 1.51 |
| Burrington Coombe | 1.39 |
| Cleeve Hill (Goblin Combe) | 1.31 |
| Bristol Road Hill, Wells | 1.17 |
| Harptree Hill | 1.16 |
| A38 (Cowslip Green) to Bristol Airport | 1.14 |
| Dundry Lane | 1.12 |
| Winscombe to Trott's Corner | 1.06 |
| Litton Combe and Torhole Bottom | 1.02 |
| Wrington to Bristol Airport | 0.93 |
| Litton Combe | 0.93 |
| Bath Road Wells | 0.89 |
| Winscombe Hill | 0.81 |
| Brockley Combe to Bristol Airport | 0.79 |
| Langford X-roads to Shipham | 0.71 |
| Bleadon Hill from Bleadon | 0.69 |
| Lye Hole to Row of Ashes Farm | 0.63 |
| Shute Shelve | 0.47 |
The system would be wrong if Draycott didn't come out on top of the pile. It does. So far, so good. But what of the rest? Wrington in second seems a bit of a surprise. It's pretty short. But then it's steep.
There's many a local cyclist says they can't get up it. The nub of this is the short steep section away from the cross-roads. Look closely at the section; it goes up 30m in 120m horizontal, that's 1 in 4! If you look at the whole of the 'real' section of this climb there's 100m of up in 1km. That's 1 in 10 (10%) as an average. Perhaps it could be the second hardest climb hereabouts.
Westfield Road in third. Yes, I found this tough. 190m of up in 2.15km.
With that average gradient of 9% you'd expect this to be well up the list. The toughest section of this may be only 13%, that's the steep rise approaching the 2km point, but that's a tough average.
It seems to work. There are some things that seem, initially, surprising - Winscombe Hill harder than Brockley Combe... yes, I think so. Brockley is mostly a 'feel good' climb, Winscombe Hill asks questions, especially in that section just past the church. Blagdon only ninth! I did expect it to be higher but then I'm not arguing that anything the SCI places above it is easier. We have some tough hills round here. The Wrangle tenth? Of this I'm not so sure, but that's where the index has a weakness; there's nothing in here about road surface or road width. If The Wrangle were on road of the quality of Cheddar Gorge it would be just another hardish climb. The fact that it isn't, that it's narrow and 10 years from being a mountain bike track, makes it that much harder. The same applies to Dundry, SCI places it 19th but it has to harder than that. Or does it? What would you take out from above it as being easier. To my mind there's only one, Cheddar Gorge. I'm not convinced it's that hard simply because the average gradient is really rather low - but it does have those two very tough corners and I guess they've been forgotten by the time you get up - you've been on easy territory for a while by then.
On the whole, I think I like this Sectioned Codifava Index. Don't get me wrong, I'm not about to start using it to analyse climbs before I go and do them. On the other hand, I reckon it gives a pretty good tool for working out if something was really hard or whether I was just off form. On Saturday I discovered what Mike and I christened 'The Barrowberg' just south of Castle Cary. We thought it tough. The SCI for it is only 0.57 so it would hardly make it onto the list. There's not enough of it score higher but with an average gradient of 9% for 550m we noticed it. If it was twice as long it would be really tough and, oh yes, about the same numbers as Wrington Hill. Length matters! Let's face it, with a climb of only 50m it doesn't get anywhere near being admitted to MRBClimbs. Just off form then I guess. Or should I be thinking about how to include maximum gradient in the calculation. I may have some maths to do...
DP
