Formula for how hard a ride is?

[Deleted member 121159]

Just had another thought. What do you use to plot your rides? (presuming you do that). When I plan a route in Strava it gives me an estimated time for it. Komoot does similar - and in Komoot you can set your athletic level (a scale of one to 5) - the time a route takes at a particular athletic level (which is synonymous to intensity/effort for me) is an indicator of how hard it it is. I am pretty sure that the calculations behind that are based on something similar to http://bikecalculator.com/
Though I have to admit I am not sure what you put in for "grade" - whether it is the over all gradient, or the average gradient for the uphill bits. The former figure tends to get used for outdoor rides, the latter for an online platform I have used. I prefer the latter because it gives you an idea of how steep the steep bits are.

E.g. The Applecross loop is 91 km with 1744 m elevation (according to RWGPS) = 1.9% elevation overall - so pretty lumpy. But it contains an ascent of Bealach na Ba, the climb in Britain with the most elevation in. There are websites that can tell you how much a route is uphill. I think you get my drift...
EDIT - I think I was trying to say what @Dogtrousers was alluding to - I use his rule of thumb

True, I could use the estimated time as an approximate measure of difficulty. I use Komoot and it consistently underestimates elevation so it tends not to be accurate. But still it's a measure of difficulty.

 

[Deleted member 121159]

My personal rule of thumb for overall ascent as a % of distance (easy to workout in metric distances, very difficult in imperial) is:

0-1% flat-undululating
1% undulating
1.5% getting hilly
2% definitely hilly
>2% aaargh
>2.5% nope. Do something else instead.

Combined with distance that gives two useful dimensions.

I suppose you could come up with a third measure of "punchiness", say median positive gradient or something like that but it would be very difficult to calculate and I'm not sure how valuable it would be.

You could combine these dimensions into a single figure with something like @ColinJ 's formula but I think they are more informative separately.

You mean average gradient for the whole ride, correct? Perhaps that multiplied by distance might be a good measure?

 

[PaulSB]

My personal rule of thumb for overall ascent as a % of distance (easy to workout in metric distances, very difficult in imperial) is:

0-1% flat-undululating
1% undulating
1.5% getting hilly
2% definitely hilly
>2% aaargh
>2.5% nope. Do something else instead.

Combined with distance that gives two useful dimensions.

I suppose you could come up with a third measure of "punchiness", say median positive gradient or something like that but it would be very difficult to calculate and I'm not sure how valuable it would be.

You could combine these dimensions into a single figure with something like @ColinJ 's formula but I think they are more informative separately.

Could you explain further please as I truly don't understand. How is 2.5% too hard to tackle? It's almost flat?

 

[Alex321]

Could you explain further please as I truly don't understand. How is 2.5% too hard to tackle? It's almost flat?

He means an average of 2.5% over the whole ride. I'm assuming that for these purposes, going downhill counts as flat, as otherwise all loops would have an average of zero.

But even that seems far too simplistic TBH. You can have an average of 2.5% where you actually spend 50% of the ride at 5%, the other 50% going back down. Or you could have one where you spend 10% of the ride at 25% and the rest flat or downhill. The second is likely to feel much harder than the first.

Looking at the overall % is only really useful if almost all your hills are of a similar gradient.

Overall, it is something which is very hard to really quantify

 

[Deleted member 121159]

Could you explain further please as I truly don't understand. How is 2.5% too hard to tackle? It's almost flat?

I upload my rides on RidewithGPS and it calculates this for me. E.g. my 75 mile ride with 5500 ft of elevation was apparently 1.4% on average.

 

[UphillSlowly]

True, I could use the estimated time as an approximate measure of difficulty. I use Komoot and it consistently underestimates elevation so it tends not to be accurate. But still it's a measure of difficulty.

The elevation under estimation is frustrating.

 

[Deleted member 121159]

The elevation under estimation is frustrating.

Yes, very frustrating! However I can't deal with RidewithGPS's interface when plotting a route. I use it to review rides only.

 

[Dogtrousers]

Could you explain further please as I truly don't understand. How is 2.5% too hard to tackle? It's almost flat?

I mean total ascent divided by total distance.

2.5% is 2,500m over 100km. (8,200' in 62 miles or 1,300' in 10 miles) That's definitely not flat.

 

[UphillSlowly]

I am sure that someone who is very clever and good at writing code could work out a formula based on:
a((%elevation hill 1 x distance x intensity) + (% elevation hill 2 x distance x intensity)...+ (%elevation hill n x distance x intensity)) + b(overall distance x overall elevation x overall intensity) + c(wind factor) + d(weather factor) + e(fatigue factor)
I guess the first two parts are what Strava and Komoot are doing
What I use for training/planning is simply (overall intensity x time) which gives me training load

 

[Dogtrousers]

But even that seems far too simplistic TBH. You can have an average of 2.5% where you actually spend 50% of the ride at 5%, the other 50% going back down. Or you could have one where you spend 10% of the ride at 25% and the rest flat or downhill. The second is likely to feel much harder than the first.

You're overcomplicating matters.

The benefit of this approach is you can look at (metric) route stats and in a moment know where it lies on the scale. Quick and simple.

As I said above you could probably come up with a statistical measure of "punchiness". It would be really hard to calculate and would probably not be worth the effort.

 

[UphillSlowly]

You're overcomplicating matters.

As I said above you could probably come up with a statistical measure of "punchiness". It would be really hard to calculate and would probably not be worth the effort.

Yes. You are probably right. Just curious to how the plotting apps come up with the predicted time. They may just use your formula 🙂

 

[PaulSB]

I mean total ascent divided by total distance.

2.5% is 2,500m over 100km. (8,200' in 62 miles) That's definitely not flat.

Yep, not flat! 🤣 Thank you

EDITED: I started to make a post realised it was nonsense and thought I'd deleted the text. Somehow the first post wasn't deleted and ended up with the above.

Apologies.

 

[Dogtrousers]

Yes. You are probably right. Just curious to how the plotting apps come up with the predicted time. They may just use your formula 🙂

Although I'm generally a fan of RideWithGPS, I think they use a random number generator for their estimated time. It's utter nonsense.

 

[DogmaStu]

A hard training session is one thing, a 'hard' route is another.

How hard a training ride is isn't necessarily based upon how hard the climbs are. You can do 10 sets of 40/20's on a flat road and be totally wrecked.
You can join a fast Club ride on a flat course and get dropped by the Group and work very hard to catch up - and in so doing, tire yourself out more than the previous weekend's slower but lumpy ride.

A 'hard' route could involve a lot of lumpy roads with some nasty steep sections that are demanding. Or be mountainous. Still, a leisurely pace with appropriate 'easy' gearing on a 'hard' route can yield a lower TSS than a hard training ride on the flat.

When someone tells me the ride is going to be hard, I don't worry about the climbs per se, I want to know the pace - that is what dictates how hard the ride will be.

 

[Jameshow]

Ah I see. That's interesting as it tends to backup my rule of thumb.....I think. We work on 1000 feet climbed for every 10 miles ridden is a hard ride. The so called Golden Ratio.

1 mile =


Yep, not flat! 🤣 Thank you

Makes sense my ride in Sunday fleet moss x2 3000ft + 1000ft of ups and downs on the valley would be a good kicking and it was.