average speed

2»

Comments

  • ai_1
    ai_1 Posts: 3,060
    tehtehteh wrote:
    I thought (perhaps wrongly) that whenever a hill slows me on the way up, I will compensate for that on the way down, and as the start and finish point of my ride is the same place then there's equal amounts of climbing and descent and it should balance out....
    As others have said, it doesn't work that way.
    It'll probably be obvious why if you run through a quick example.

    So say your average speed on a flat windless 10km long route would be 30km/h and being flat and windless you'd be at or about 30km/h consistently throughout the ride.
    Now lets assume but instead you ride a route which goes up hill at 10% for 5km and then downhill at 10% for 5km. Let's say you can climb a 10% gradient at 12km/h for 5km and that you have nice smooth straight roads on the descent that allow you really move and you average 90km/h on the descent.

    On the flat route: 10km at 30km/h = 20mins
    On hill route - ascending: 5km at 12km/h = 25mins (already slower than a flat 10km and you haven't even started decending)
    On hill route - descending: 5km at 90km/h = 3mins 20sec
    Hill route total: 28mins20sec

    So as you can see in this example, no matter how fast you go down the hill you can't make up the time lost ascending. Obviously this is an extreme example where the whole ride is up and down a 10% hill and you may disagree with the speeds I've chosen but if you re-do this in more detail for your own routes you'll find the same thing. You'll always lose more time ascending than you can gain descending.

    Average speed for a climb and descent is not the average of the climbing and descending speeds.
    e.g. in the example above (12+90)/2 = 51km/h but that is NOT your average speed for the ride because average speed is time dependent and you spent much longer climbing than descending. So, the speed ascending has a correspondingly weighted impact on the overall average speed. 12*25mins+90*3.33mins/28.33 mins = 21.2km/h
    Easier to just do it this way though: 10*(60/28.33) = 21.2km/h
  • Mikey23
    Mikey23 Posts: 5,306
    Thanks... An excellent explanation!
  • norvernrob
    norvernrob Posts: 1,447
    Some of my rides:

    20 mile training run, 1,000ft of elevation - 19.5mph
    63 mile Peak District ride, 7,300ft of elevation - 13mph
    103 mile Lincolnshire Sportive, 3,500ft of elevation - 17.4mph.

    And all those would be different if I did them again depending on the weather conditions. Basically someone else's average speed means nothing, even your own means nothing unless you're doing the same route with similar conditions.
  • Given constant power the key reason hilly courses have lower speed is exponential drag increase.

    You would think that the speed decrease caused by going uphill at 5% slope is the same as the speed benefit of going downhill at 5%. But it isn't because the drag increases exponentially so the downhill guy loses some of his benefit.

    The other issue is a human one. Most cyclists put down less power when descending than when climbing and on the flat, so average power is reduced on hilly courses.
  • ai_1
    ai_1 Posts: 3,060
    Dippydog3 wrote:
    Given constant power the key reason hilly courses have lower speed is exponential drag increase.

    You would think that the speed decrease caused by going uphill at 5% slope is the same as the speed benefit of going downhill at 5%. But it isn't because the drag increases exponentially so the downhill guy loses some of his benefit.

    The other issue is a human one. Most cyclists put down less power when descending than when climbing and on the flat, so average power is reduced on hilly courses.
    I disagree

    The first has an impact but it is minor compared to the main reason explained in my last post. Were this not the case then windy days would not mean slower rides (but they do). As with the case of hilly rides, your average speed is impacted negatively because you spend significantly more time at low speed than at high speed. If you did an out and back ride on a windy day so that you had a headwind outbound and tailwind return (no sidewinds) then this would be somewhat comparable with the hill ride example earlier. If you ride at a constant airspeed (so speed increase is irrelevant) then your outbound groundspeed will be lower. It'll still take longer to do the ride and you'll have a lower average speed because you spent longer going slower. This is mathematically demonstrable as above.

    I also believe your second point is questionable but that's just my personnal opinion. I typically expend much more effort on most hills than on the flat and I also tend to push harder on descents because I enjoy them and the satisfaction of a fast decent makes it easier to push harder. My average HR will always be much higher on a mountain ride than a flat one unless I'm intentionally taking it incredibly easy. This is primarily because of the climbs but my HR does not typically plummet on the descents either (also bear in mind, as explained previously, you spend less time descending so that phase of the ride not only has less impact on average speed but also on any other averages like average HR, average power, average cadence, etc).

    Put simply: If you have hills, you have more work to be done. If you have constant power, that work will take longer.
  • Ai_1 wrote:
    tehtehteh wrote:
    I thought (perhaps wrongly) that whenever a hill slows me on the way up, I will compensate for that on the way down, and as the start and finish point of my ride is the same place then there's equal amounts of climbing and descent and it should balance out....
    As others have said, it doesn't work that way.
    It'll probably be obvious why if you run through a quick example.

    So say your average speed on a flat windless 10km long route would be 30km/h and being flat and windless you'd be at or about 30km/h consistently throughout the ride.
    Now lets assume but instead you ride a route which goes up hill at 10% for 5km and then downhill at 10% for 5km. Let's say you can climb a 10% gradient at 12km/h for 5km and that you have nice smooth straight roads on the descent that allow you really move and you average 90km/h on the descent.

    On the flat route: 10km at 30km/h = 20mins
    On hill route - ascending: 5km at 12km/h = 25mins (already slower than a flat 10km and you haven't even started decending)
    On hill route - descending: 5km at 90km/h = 3mins 20sec
    Hill route total: 28mins20sec

    So as you can see in this example, no matter how fast you go down the hill you can't make up the time lost ascending. Obviously this is an extreme example where the whole ride is up and down a 10% hill and you may disagree with the speeds I've chosen but if you re-do this in more detail for your own routes you'll find the same thing. You'll always lose more time ascending than you can gain descending.

    Average speed for a climb and descent is not the average of the climbing and descending speeds.
    e.g. in the example above (12+90)/2 = 51km/h but that is NOT your average speed for the ride because average speed is time dependent and you spent much longer climbing than descending. So, the speed ascending has a correspondingly weighted impact on the overall average speed. 12*25mins+90*3.33mins/28.33 mins = 21.2km/h
    Easier to just do it this way though: 10*(60/28.33) = 21.2km/h

    Surely a more coherent example would be;

    5km climb at 20km/h = 15 minutes climbing.
    5km descent at 60km/h = 5 minutes descending.

    = the same.

    Although you would have to push hard on the climb, you would be mostly freewheeling on the descent**, so v.low HR.

    I would imagine that for me, my average HR over a ride like that would be about 145-150 (160ish up, and 110-120 down).

    To do 30km/h for 10km on the flat (no wind) I would estimate an HR of 140.

    So I've worked harder on the hilly course overall, but to my mind, that is simply down to the exponentially higher drag encountered on the descent.*

    A power meter would make this experiment much more precise and useful.



    *the extra drag on the descent is far higher than that saved on the climb
    **descent assumed to be straight, i.e. no braking
  • ai_1
    ai_1 Posts: 3,060
    Ai_1 wrote:
    tehtehteh wrote:
    I thought (perhaps wrongly) that whenever a hill slows me on the way up, I will compensate for that on the way down, and as the start and finish point of my ride is the same place then there's equal amounts of climbing and descent and it should balance out....
    As others have said, it doesn't work that way.
    It'll probably be obvious why if you run through a quick example.

    So say your average speed on a flat windless 10km long route would be 30km/h and being flat and windless you'd be at or about 30km/h consistently throughout the ride.
    Now lets assume but instead you ride a route which goes up hill at 10% for 5km and then downhill at 10% for 5km. Let's say you can climb a 10% gradient at 12km/h for 5km and that you have nice smooth straight roads on the descent that allow you really move and you average 90km/h on the descent.

    On the flat route: 10km at 30km/h = 20mins
    On hill route - ascending: 5km at 12km/h = 25mins (already slower than a flat 10km and you haven't even started decending)
    On hill route - descending: 5km at 90km/h = 3mins 20sec
    Hill route total: 28mins20sec

    So as you can see in this example, no matter how fast you go down the hill you can't make up the time lost ascending. Obviously this is an extreme example where the whole ride is up and down a 10% hill and you may disagree with the speeds I've chosen but if you re-do this in more detail for your own routes you'll find the same thing. You'll always lose more time ascending than you can gain descending.

    Average speed for a climb and descent is not the average of the climbing and descending speeds.
    e.g. in the example above (12+90)/2 = 51km/h but that is NOT your average speed for the ride because average speed is time dependent and you spent much longer climbing than descending. So, the speed ascending has a correspondingly weighted impact on the overall average speed. 12*25mins+90*3.33mins/28.33 mins = 21.2km/h
    Easier to just do it this way though: 10*(60/28.33) = 21.2km/h

    Surely a more coherent example would be;

    5km climb at 20km/h = 15 minutes climbing.
    5km descent at 60km/h = 5 minutes descending.

    = the same.

    Although you would have to push hard on the climb, you would be mostly freewheeling on the descent**, so v.low HR.

    I would imagine that for me, my average HR over a ride like that would be about 145-150 (160ish up, and 110-120 down).

    To do 30km/h for 10km on the flat (no wind) I would estimate an HR of 140.

    So I've worked harder on the hilly course overall, but to my mind, that is simply down to the exponentially higher drag encountered on the descent.*

    A power meter would make this experiment much more precise and useful.



    *the extra drag on the descent is far higher than that saved on the climb
    **descent assumed to be straight, i.e. no braking
    I don't think those figures are realistic. Certainly not for me anyway, and I suspect anyone who could climb at 20 on the same slope they would descend at 60 would go a hell of a lot faster than 30 on the flat, thus remaining slower on the hilly ride.
    The drag issue is a factor but it's not the big one.
    I think some of the confusion on this comes from talking in mph or km/h. If you think in terms of time per km or even pace yourself that way on a few rides I think it would become clear that the extra time lost on climbs far exceeds what can be clawed back on descents.
    10km/h = 6min/km
    15km/h = 4min/km
    20km/h = 3min/km
    24km/h = 2min30s/km
    30km/h = 2min/km
    36km/h = 1min40s/km
    40km/h = 1min30s/km
    50km/h = 1min12s/km
    60km/h = 1min/km
    etc...

    It's a very simple mental excercise. Be realistic or it's a waste of time. Pick a gradient and distance. Decide for yourself what speed you would realistically climb it. Then decide what speed you would decend it. Decide how fast you would cover that distance on the flat.
    If the climb pace is less than half the flat pace, then the decent is irrelevant as you've already exceeded your flat ride time before you even start descending. At some gradient and distance this will be the case for everyone. So clearly drag is not the primary issue here.
  • Ai_1 wrote:
    I don't think those figures are realistic. Certainly not for me anyway, and I suspect anyone who could climb at 20 on the same slope they would descend at 60 would go a hell of a lot faster than 30 on the flat, thus remaining slower on the hilly ride.

    They're realistic for quite a few people on the forum, but I agree, not everybody.

    In my example, I reduced my effort on the flat to be at 30km/h. For the same effort as the hilly 10k, the flat would be more like 35km/h.

    I find that i can ride at 30km/h on a ride without much problem, and hills don't change that very much (eg. a 60km ride would typically have 600m of climbing).

    I think what slows people down in real life is that descents nearly always have bends, so you have to brake at some point.

    So I still reckon that what stuffs the average for a hilly ride is a) exponentially higher drag and b) braking on the descents.

    If it's something else, what is the physics behind that? Why is time being lost elsewhere?
  • davep1
    davep1 Posts: 836
    Loving this stuff, great debate and information! It rings true for me because I did the Legs of Steel last Saturday, and missed the 3 hr 10 gold standard by 3 mins 45. I am planning on re-riding the route, and trying to work out where to put the effort in to make the time up.

    I had done a lot of climbing (for me anyway!) the two previous weeks, 1400 and 1500 m respectively. I thought that alone would mean I could hit my target, but I blew the start (led a bunch of fit lads up Ranmore, near the top they powered off into the distance and I rode pretty much the whole route solo after that) and never really felt like I had much left in the tank. I was thinking if I take it a bit easier on the climbs and really focus on the flats and descents, that would be a better way to approach it. From what Ai_1 is saying not sure any more!
  • DaveP1 wrote:
    Loving this stuff, great debate and information! It rings true for me because I did the Legs of Steel last Saturday, and missed the 3 hr 10 gold standard by 3 mins 45. I am planning on re-riding the route, and trying to work out where to put the effort in to make the time up.

    I had done a lot of climbing (for me anyway!) the two previous weeks, 1400 and 1500 m respectively. I thought that alone would mean I could hit my target, but I blew the start (led a bunch of fit lads up Ranmore, near the top they powered off into the distance and I rode pretty much the whole route solo after that) and never really felt like I had much left in the tank. I was thinking if I take it a bit easier on the climbs and really focus on the flats and descents, that would be a better way to approach it. From what Ai_1 is saying not sure any more!


    That's a good effort, but I would say that you'd be better off working harder on the climbs than the flats and descents.

    Because the hardest thing in cycling is to force your bike fast through the air. You need loads more power to go from 30 to 35km/h on the flat than you do to go from 25 to 30.

    Go hard on the climbs, keep a steady pace on the flats and take it easy on the descents (still pedal, but easily), and practise your descending skills (cornering, braking) as you can gain a lot of time there too.
  • If you did an out and back ride on a windy day so that you had a headwind outbound and tailwind return (no sidewinds) then this would be somewhat comparable with the hill ride example earlier. If you ride at a constant airspeed (so speed increase is irrelevant) then your outbound groundspeed will be lower. It'll still take longer to do the ride and you'll have a lower average speed because you spent longer going slower. This is mathematically demonstrable as above.
    You wouldn't ride at a constant airspeed though.

    10mph into a 10mph headwind is roughly half the power of 20mph in still air, and 30mph with a 10mph tailwind is roughly 1.5 times more. This is because aero dominates, and the force to hold the speed is related to the airspeed, but the power is related to the groundspeed.

    Using http://bikecalculator.com/, which seems plausible, then riding 30k with 250W on a still day takes 47.72min, riding 15k out into a 10kmh wind takes 28.38min, and back with a tailwind, 20.33min, total about 48.7min. So slower but not drastically. Make the wind 20kmh and the return trip takes just under 52 minutes.

    However make the out up a 5% gradient in still air and it takes 52.6min just to get to the turn.

    Anyway average speed in general gives an indication of fitness, but riding to optimise average speed isn't a good way to improve it. As has been discussed up thread.

    Paul
  • CiB
    CiB Posts: 6,098
    tehtehteh wrote:
    I thought (perhaps wrongly) that whenever a hill slows me on the way up, I will compensate for that on the way down, and as the start and finish point of my ride is the same place then there's equal amounts of climbing and descent and it should balance out.
    I know it's been answered but for clarity..

    You have a hill, 5 miles up, 5 miles down the other side; you climb it at 10mph all the way up. 10mph = 6 mins per mile = 30 minutes to get to the top. Going down you roll down at 30mph = 2 mins per mile = 10 minutes to get back down. You've covered 10 miles in 40 mins, which is a 15mph avg speed over the ride.

    You've averaged your 10mph up & 30mph down averages to get an avg speed of 20mph instead of the actual 15mph, and ignored the time factor.

    That said, on not too hilly routes with a few undulations an average of less than 17 for less than a couple of hours riding isn't trying hard enough.
  • DaveP1 wrote:
    Loving this stuff, great debate and information! It rings true for me because I did the Legs of Steel last Saturday, and missed the 3 hr 10 gold standard by 3 mins 45. I am planning on re-riding the route, and trying to work out where to put the effort in to make the time up.

    I had done a lot of climbing (for me anyway!) the two previous weeks, 1400 and 1500 m respectively. I thought that alone would mean I could hit my target, but I blew the start (led a bunch of fit lads up Ranmore, near the top they powered off into the distance and I rode pretty much the whole route solo after that) and never really felt like I had much left in the tank. I was thinking if I take it a bit easier on the climbs and really focus on the flats and descents, that would be a better way to approach it. From what Ai_1 is saying not sure any more!


    That's a good effort, but I would say that you'd be better off working harder on the climbs than the flats and descents.

    Because the hardest thing in cycling is to force your bike fast through the air. You need loads more power to go from 30 to 35km/h on the flat than you do to go from 25 to 30.

    Go hard on the climbs, keep a steady pace on the flats and take it easy on the descents (still pedal, but easily), and practise your descending skills (cornering, braking) as you can gain a lot of time there too.

    I think that's great advice. I've always enjoyed the climbing, but struggle on the flats and descents, but after working hard all summer on my cornering, braking and general fear of going over 35mph, I've gained loads of time on rides...

    The fact of the matter on climbs is that wind resistance is negligible (unless it's windy, obviously!) as the average speed is so low...

    good debate - great advice for us novices :-)
    Wilier Zero.7 Chorus
  • ai_1
    ai_1 Posts: 3,060
    .....So I still reckon that what stuffs the average for a hilly ride is a) exponentially higher drag and b) braking on the descents.

    If it's something else, what is the physics behind that? Why is time being lost elsewhere?
    a) and b) matter but as I've tried to explain, perhaps not very well, they're minor compared to c)

    c) is Maths rather than Physics as per my previous comments.

    If I could pose one question and suggest one more example:
    Question:Is there any hill you think no-one can climb at least 50% as fast as they cruise on the flat? This can even be a hypothetical hill, the argument is still sound.
    Example: How fast could Contador, Froome or Quintana go up a 5km 30% hill? I'm not sure if such a climb exists but let's pretend it did. Even if they could do that at say 20km/h which I'm pretty sure they couldn't, it would take 15 minutes? If they do the decent at the speed of light the average is still 40km/h for the 10km up and down.
    I think you'll agree they can all manage 10km on the flat at 40km/h+? So even for superb climbers granted the hypothetical superpower of speed of light decents (so aero drag or braking clearly isn't a factor) - they can't go as fast over a hill as on the flat.
  • ai_1
    ai_1 Posts: 3,060
    edited September 2014
    DaveP1 wrote:
    Loving this stuff, great debate and information! It rings true for me because I did the Legs of Steel last Saturday, and missed the 3 hr 10 gold standard by 3 mins 45. I am planning on re-riding the route, and trying to work out where to put the effort in to make the time up.

    I had done a lot of climbing (for me anyway!) the two previous weeks, 1400 and 1500 m respectively. I thought that alone would mean I could hit my target, but I blew the start (led a bunch of fit lads up Ranmore, near the top they powered off into the distance and I rode pretty much the whole route solo after that) and never really felt like I had much left in the tank. I was thinking if I take it a bit easier on the climbs and really focus on the flats and descents, that would be a better way to approach it. From what Ai_1 is saying not sure any more!


    That's a good effort, but I would say that you'd be better off working harder on the climbs than the flats and descents.

    Because the hardest thing in cycling is to force your bike fast through the air. You need loads more power to go from 30 to 35km/h on the flat than you do to go from 25 to 30.

    Go hard on the climbs, keep a steady pace on the flats and take it easy on the descents (still pedal, but easily), and practise your descending skills (cornering, braking) as you can gain a lot of time there too.
    I agree it's worth putting the bulk of your effort into the climbs and any upwind sections.
    For the same reasons I've gone into earlier, the time spent at a lower speed, rather than the distance, is what's most important. So getting over a hill or completing an upwind leg quickly is very beneficial. Your average speed for that leg may be improved at the expense of the descent or downwind leg but so long as you don't blow up it will be worth it. With any significant gradient or windspeed you won't lose as much on the easier terrain as you would by taking it easier on the tough stuff.
    Unfortunately it can be hard to gauge what effort is appropriate and I'm no expert on that. I find it much easier to judge what continuous effort I can sustain than judging the maximum effort from which I can recover on the faster sections.

    You do, as bernithebiker says, you get more bang for your buck due to reduced aerodynamic losses when you're adding speed from a lower baseline. However if you're climbing a steep hill that extra speed also costs you more than just the aero loses. You're also gaining altitude faster and therefore adding 5km/h on a gradient is harder than adding the same speed from a similar baseline on the flat.
  • ai_1
    ai_1 Posts: 3,060
    paul2718 wrote:
    If you did an out and back ride on a windy day so that you had a headwind outbound and tailwind return (no sidewinds) then this would be somewhat comparable with the hill ride example earlier. If you ride at a constant airspeed (so speed increase is irrelevant) then your outbound groundspeed will be lower. It'll still take longer to do the ride and you'll have a lower average speed because you spent longer going slower. This is mathematically demonstrable as above.
    You wouldn't ride at a constant airspeed though.

    10mph into a 10mph headwind is roughly half the power of 20mph in still air, and 30mph with a 10mph tailwind is roughly 1.5 times more. This is because aero dominates, and the force to hold the speed is related to the airspeed, but the power is related to the groundspeed.
    Either you're a bit confused about force and power here, or I'm misunderstanding you!
    I agree aerodynamic drag is the main limiting factor when cycling at speed. However I don't know what you mean about force being due to aero and power related to groundspeed. Power is power. It's your rate of application of energy. Regardless whether aero drag, rolling resistance or brake rub is responsible for resisting movement more power will provide more speed and vice versa.
    In my example of constant airspeed I was not suggesting that effort was also constant. As in the hill discussion I was referring to, it would be expected you'd ride harder into the wind so would have a higher windspeed on the upwind section and higher rolling resistance on the downwind leg would reduce airspeed for this leg. Nevertheless, in what was supposed to be a simplified example, even if effort was constant you'd expect variation in airspeed would be due to differences in rolling resistance. Higher groundspeed = higher rolling resistance = lower airspeed for a given effort. Of course things aren't that simple so if we were looking at this as a realistic case study wind gradient would come into play too so that the body of air you're moving through isn't homogeneous and windspeed is zero at ground level and increases higher off the ground. i.e. your feet don't feel as much of a headwind as your head does.
    ...Anyway average speed in general gives an indication of fitness, but riding to optimise average speed isn't a good way to improve it. As has been discussed up thread.

    Paul
    Well put. I often forget that myself when I'm not doing structured training.
  • norvernrob
    norvernrob Posts: 1,447
    It is all very interesting. I'll throw another spanner into the works, most of the descents I did on Sunday were single track, often gravelly and twisty. There was no way I could go anywhere near full gas or even let the bike go down those. On one particularly steep and nasty section averaging 14% my average speed was less than 10mph, even the strava segment is called 'very steep gravelly descent'.

    On sighted, well surfaced descents I'm pretty fearless and will hammer it, but it's impossible on those types of roads unless you have a death wish!
  • bobmcstuff
    bobmcstuff Posts: 11,196
    NorvernRob wrote:
    On sighted, well surfaced descents I'm pretty fearless and will hammer it, but it's impossible on those types of roads unless you have a death wish!

    Indeed - if you look at some descents on Strava it's pretty clear that the only way to get near the top is to have absolutely no fear of dying. Personally I like to slow down coming up to tight blind corners, but based on some people's speeds they must be hammering them at full pelt.
  • ai_1
    ai_1 Posts: 3,060
    NorvernRob wrote:
    It is all very interesting. I'll throw another spanner into the works, most of the descents I did on Sunday were single track, often gravelly and twisty. There was no way I could go anywhere near full gas or even let the bike go down those. On one particularly steep and nasty section averaging 14% my average speed was less than 10mph, even the strava segment is called 'very steep gravelly descent'.

    On sighted, well surfaced descents I'm pretty fearless and will hammer it, but it's impossible on those types of roads unless you have a death wish!
    Yeah, I ride a lot in the Wicklow hills (near Dublin Ireland). There's plenty spots where I'll regularly exceed 70km/h and some where I'll get over 85km/h if I'm feeling brave but on one particular road with a 18% gradient in places I'm doing well to get over 15km/h in bad weather or 25km/h in good conditions due to the bendy, often leaf strewn and slippy state of the road. I hate really steep descents on twisty, blind cornered roads - very frustrating. I love anything in the 3%-12% range on good road surfaces with good visibility of the route ahead. Although I do sometimes get to the bottom and think "I shouldn't have gone that fast...would have been nasty if anything went wrong"
  • Ai_1 wrote:
    .....So I still reckon that what stuffs the average for a hilly ride is a) exponentially higher drag and b) braking on the descents.

    If it's something else, what is the physics behind that? Why is time being lost elsewhere?
    a) and b) matter but as I've tried to explain, perhaps not very well, they're minor compared to c)

    c) is Maths rather than Physics as per my previous comments.

    OK, so what is the Maths? I'm genuinely interested. Because IN THEORY, the laws of conservation of energy suggest that the potential energy that you have gained climbed will be given back to you when you descend.

    So what is c)? Where is the energy / time going?
  • ai_1
    ai_1 Posts: 3,060
    Ai_1 wrote:
    .....So I still reckon that what stuffs the average for a hilly ride is a) exponentially higher drag and b) braking on the descents.

    If it's something else, what is the physics behind that? Why is time being lost elsewhere?
    a) and b) matter but as I've tried to explain, perhaps not very well, they're minor compared to c)

    c) is Maths rather than Physics as per my previous comments.

    OK, so what is the Maths? I'm genuinely interested. Because IN THEORY, the laws of conservation of energy suggest that the potential energy that you have gained climbed will be given back to you when you descend.

    So what is c)? Where is the energy / time going?
    I've already gone through the maths in previous posts with various examples.
    In brief, travelling at two different speeds during a trip, rather than travelling at a constant rate, means the slower speed has a disproportionately large effect on average speed compared to the distance over which you are at that speed because the time spent at that speed is disproportionately longer. So, if you want to average 30km/h and you drop to 25km/h for 1km then doing the next one at 35km/h will not be enough. You'll be getting on for 7 seconds behind schedule. However if you drop to 25km/h for 1 minute and then rise to 35km/h for 1 minute, you'll be back on track. I think people are confusing the impact of time and distance on avergae speed.

    While I agree that potential energy stored by getting to the top of the hill will indeed be utilised on the descent I'm unclear why you would expect that fact to deliver an equivalent average speed if it weren't for the increased drag loses due to higher speeds on decent?
  • norvernrob
    norvernrob Posts: 1,447
    bobmcstuff wrote:
    NorvernRob wrote:
    On sighted, well surfaced descents I'm pretty fearless and will hammer it, but it's impossible on those types of roads unless you have a death wish!

    Indeed - if you look at some descents on Strava it's pretty clear that the only way to get near the top is to have absolutely no fear of dying. Personally I like to slow down coming up to tight blind corners, but based on some people's speeds they must be hammering them at full pelt.

    The descent we went down at the weekend was twisty with loads of unsighted bends, hawthorn hedges and fences either side, covered in gravel and averaged -14.2% with plenty of 20% sections. It's the first time I've gone down any hill with both hands constantly on the brakes and without pedalling once. Yet the top average speed down there is 25mph, that's absolutely suicidal.
  • davep1
    davep1 Posts: 836
    Going back to Dorking a week on Sunday with a couple of mates, will be trying again!
  • I do see where you're coming from- when I was living in Guildford (for a fair while!) I had a route that was somewhere between 25 and 30 miles long and took in a bit of lumpy stuff (not any 'proper' climbs though) and also some fast flat stuff towards the end. I would do that route if I was short of time or just fancied a spin after dark in the winter months and it's interesting to see how the average speed changes.

    However, as others have said it depends so much on conditions etc that you'll only notice big changes, so I might perhaps ride that route at the beginning and end of a season to see how I compare.

    Nowerdays I use my local 10TT course using my standard road bike without any of my TT gear to give myself a better benchmark, as you are far less likely to be held up by traffic etc along the way.
  • Ai_1 wrote:
    Ai_1 wrote:
    .....So I still reckon that what stuffs the average for a hilly ride is a) exponentially higher drag and b) braking on the descents.

    If it's something else, what is the physics behind that? Why is time being lost elsewhere?
    a) and b) matter but as I've tried to explain, perhaps not very well, they're minor compared to c)

    c) is Maths rather than Physics as per my previous comments.

    OK, so what is the Maths? I'm genuinely interested. Because IN THEORY, the laws of conservation of energy suggest that the potential energy that you have gained climbed will be given back to you when you descend.

    So what is c)? Where is the energy / time going?
    I've already gone through the maths in previous posts with various examples.
    In brief, travelling at two different speeds during a trip, rather than travelling at a constant rate, means the slower speed has a disproportionately large effect on average speed compared to the distance over which you are at that speed because the time spent at that speed is disproportionately longer. So, if you want to average 30km/h and you drop to 25km/h for 1km then doing the next one at 35km/h will not be enough. You'll be getting on for 7 seconds behind schedule. However if you drop to 25km/h for 1 minute and then rise to 35km/h for 1 minute, you'll be back on track. I think people are confusing the impact of time and distance on avergae speed.

    While I agree that potential energy stored by getting to the top of the hill will indeed be utilised on the descent I'm unclear why you would expect that fact to deliver an equivalent average speed if it weren't for the increased drag loses due to higher speeds on decent?

    And your confusing the calculation of average speed and the impact of time and distance with the actual reason why your slower. So you're right that lots of people confuse the distance they spend at X speed, when really it's the time they speed at X that matters when calculating average speed.

    However, that's not the the reason you're slower over a hilly route. The reason is when climbing you store potential energy as you work against gravity and you loose less energy to drag (as you're going slower). When you descend you get that sorted energy back, however, as air resistance is exponential you loose more energy to aero drag with your increased in speed when descending than saved whilst going at a slower speed whilst climbing.

    So, assuming a 10k time trial, 5k out at 3% and 5k back at -3%. You could put out 429 Watts on the way out and maintain 33kmh then do just 9 watts on the way back to to average 33 km and 219 watts over 10k, if the terrain was flat, you could still ride at 10 k at a constant 219 watts and still average 33kmh. However, this is where where human physiology comes in to it, as if 219 watts is FTP for a 10k, then you'r unlikely to be able to raise it to 429 for half the distance, even if the average over the full distance is still 219.

    If you're still with me, this means, that on any serious gradient it's impossible to raise your wattage on the incline to maintain a constant speed, so your speed drops for the uphill, and then on the decline you go faster... and then you loose more energy on through drag than you save by going slow because air drag is exponential with speed.

    So, if you've got a hilly course for your TT it will always pay to push a little harder on the hills and recover slightly on the downhills, but human physiology will never let you do this to such an extent that if it's a hilly course you're overall average speed doesn't suffer. That is the scientific reason why people are slower over hills.
  • ednino
    ednino Posts: 684
    Why are people so bothered about their average speed? Just ride. Mine goes up & down from 16-20mph depending on the ride. I'm only bothered about the average for TTs