Naismith's rule? - but for cyclists?

[PpPete]

Hillwalkers amongst you may be familiar with "Naismith's rule" which is a handy guide for working out time to complete a given walking route:

In it's original version:
20 minutes per mile + 30 minutes per 1000 ft ascent.

Or in metric units:
12 minutes per km + 10 minutes per 100 m

Personally I find that "generous" so long as the path is reasonable, whereas over Pennine peat groughs it's tough going, but with modifications its a useful planning tool.

However.... does anyone know of anything similar for cyclists? Or have a 'pet' method of their own?

 

[on the road]

When I'm planning a ride, I estimate the time based on my typical average speed and then add 20 minutes for stoppages like traffic lights, other junctions and anything else I might come across.

 

[jimboalee]

Jimbo's rule.

Work out the time based on unhindered, flat travel.

Add on 1% for every 'stop junction' and upward contour ( 10m ) more than down contours.

A 'stop junction' is anywhere where you may have to stop to give way to other traffic.

 

[PpPete]

Jimbo's rule.

Work out the time based on unhindered, flat travel.

Add on 1% for every 'stop junction' and upward contour ( 10m ) more than down contours.

A 'stop junction' is anywhere where you may have to stop to give way to other traffic.

So if you have an equal number of up and down contours (which you must have if route is circular) then you make no allowance for climbing?

Better legs than me then !
I'm sure I slow down more on the ups than I speed up on the downs.

 

[Hont]

I've never thought about this, but it would be interesting to look into (if you're a stats obsessive like me that is). The problem is there are a few extra variables, as you point out with walking the terrain itself can make a difference and the same is true with cycling and road surface. Wind is also a factor as is bike type, rider weight and fitness. So it'll be difficult to get a general rule that fits everyone.

 

[jimboalee]

So if you have an equal number of up and down contours (which you must have if route is circular) then you make no allowance for climbing?

Better legs than me then !
I'm sure I slow down more on the ups than I speed up on the downs.

Well spotted, and if one would like to be precise.....

I've found I won't need to stop at every junction.
I've found my psychological effort dismisses the climbing theory.

I've found it's almost impossible to predict the time of a bicycle ride.

 

[gaz]

I've found it's almost impossible to predict the time of a bicycle ride.

This! well said.

 

[kewb]

avg speed , distance = time approximation which is what naismiths rule is ,
i check my distance to the turning point then my average speed , factor in stuff like lights ,
and i can estimate to within 15 /20 minutes how long return journeys will be (accurate enough for me ),
on a new route i just estimate based on a familiar route which usually works out .

 

[PpPete]

Well spotted, and if one would like to be precise.....

I've found I won't need to stop at every junction.
I've found my psychological effort dismisses the climbing theory.

I've found it's almost impossible to predict the time of a bicycle ride.

ALthough junctions and road surfaces will have a non-negligible impact I suppose I was expecting a "read-across" from walking and the ascent to be the major factor apart from the distance.

 

[Oldlegs]

My digital map (Tracklogs) has a Naismith road cycling calculator built in.

The user setup requires speeds for flat, 1:10 ascending and descending and something called the Trantor correction. Cannot see the algorithm but it's doubtless more complicated than the walking one.

After a bit of fiddling with the variables it gives pretty accurate results.

 

[piedwagtail91]

My digital map (Tracklogs) has a Naismith road cycling calculator built in.

The user setup requires speeds for flat, 1:10 ascending and descending and something called the Trantor correction. Cannot see the algorithm but it's doubtless more complicated than the walking one.

After a bit of fiddling with the variables it gives pretty accurate results.

i'll second the tracklogs one, it's pretty accurate once set up.

 

[jimboalee]

Garmin's 'ETA' function has something built in.

Country speed predictions are always faster than city speed predictions, which indicates to me junctions are compensated for.

I haven't assessed whether Garmin compensates for hills if Topo GB is installed.
NO CHANCE if City Navigator is installed.....


I have my own calcs sheet which compensates for temperature, wind direction and wind strength.
It gets the 'Trip duration' to within 2 minutes in an hour's trip; 10 minutes on a 100km Brevet Pop, and 20 minutes on a Rando 200km.

As I said, NO accuracy when predicting bike rides.

 

[hubbike]

Naismith rule is that 1 unit of vertical travel is equivilent to 7.92 units of horizontal. (use miles, kms, feet, inches, bananas as you wish.)

Analysis of cycling data suggests a similar rule (1:8.2) for cycling on mountainous roads and tracks.

So going up 2000m is equivalent to adding an extra 16km. no matter how steep.

However, going down speeds you up much more on a bike than on foot and this is unaccounted for in Naismith.

any help?

 

[snorri]

I use the rule 'if it takes one hour in a car, it'll take one day on the bike'

 

[HLaB]

I use the rule 'if it takes one hour in a car, it'll take one day on the bike'

That rule wouldn't work here Going to Edinburgh takes half an hour in the car, add parking etc and it can take longer; I can cycle it in under an hour. Fully laid down with pannier it still only takes an 1h and 1/2.