Page 1 of 2 1 2 LastLast
Results 1 to 25 of 45
  1. #1
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070

    Interesting study on bike weight vs. performance

    I dont spend a lot of time in here, so if this is a repost, please disregard and accept my apologies. I just thought it was interesting.

    Power development in hill climbing as a function of bicycle weight

    General Conclusion: A 2 lb reduction in bike weight resulted in a 22 sec advantage for racers over a 1.8 mile steep climb.
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

  2. #2
    mtbr member
    Reputation:
    Join Date
    Sep 2010
    Posts
    874
    Great info!

  3. #3
    mtbr member
    Reputation: B.Trimble's Avatar
    Join Date
    Oct 2011
    Posts
    143
    A light bike certainly feels faster. Interesting reading

  4. #4
    mtbr member
    Reputation:
    Join Date
    Nov 2005
    Posts
    2,663
    Makes my wallet feel a lot lighter which in turn motivates me to pedal harder. Instant performance increase. As the Canadiens call it, "a win-win."

  5. #5
    mtbr member
    Reputation:
    Join Date
    Aug 2008
    Posts
    406
    Quote Originally Posted by Stumpjumpy View Post
    General Conclusion: A 2 lb reduction in bike weight resulted in a 22 sec advantage for racers over a 1.8 mile steep climb.
    It may just be semantics, but it actually says that a 2lb increase resulted in a 21 second disadvantage.

    I agree that a light bike feels better but I wonder what the true result of removing an addition 1kg from their standard bike would be. I assume that since they were racers they were already on fairly sorted out bikes.

    I guess wonder what the point of diminished returns are, and what are the variables that factor into that; bike weight, rider weight, rotating/non-rotating weight, course type, etc.

  6. #6
    DIY all the way
    Reputation: Mr.Magura's Avatar
    Join Date
    Aug 2010
    Posts
    2,411
    Keep in mind, that the study makes some pretty rough assumptions.

    Pedaling efficiency for one, is not going to be equal on a noodle 2 lbs frame, and on a super rigid 4 lbs frame.
    The same could be said about many other components.

    Another assumption would be the loss of power due to the condition of the trail is not playing a part, which is also highly questionable.

    So if two bikes perform exactly equal, under the conditions of the trail and so forth, yes, the lighter bike is faster.

    To need a study to reach that conclusion, well, ahem, is pretty much a joke.

    So if we can have components and frames, that performs equal, but are lighter, all is good.
    If not, the 22 seconds mentioned above,and then some, are going to be lost real quick


    Magura

  7. #7
    mtbr member
    Reputation: mtbnozpikr's Avatar
    Join Date
    Sep 2008
    Posts
    1,684
    Quote Originally Posted by Mr.Magura View Post
    Keep in mind, that the study makes some pretty rough assumptions.

    Pedaling efficiency for one, is not going to be equal on a noodle 2 lbs frame, and on a super rigid 4 lbs frame.
    The same could be said about many other components.

    Another assumption would be the loss of power due to the condition of the trail is not playing a part, which is also highly questionable.

    So if two bikes perform exactly equal, under the conditions of the trail and so forth, yes, the lighter bike is faster.

    To need a study to reach that conclusion, well, ahem, is pretty much a joke.

    So if we can have components and frames, that performs equal, but are lighter, all is good.
    If not, the 22 seconds mentioned above,and then some, are going to be lost real quick


    Magura

    Great point but simply the fact that this guy did this as his master's thesis is pretty cool in my opinion.
    2012 Intense M9
    2012 Pivot Mach 5.7 Carbon
    2008 Look 595
    2007 Custom Litespeed Sewanee

  8. #8
    LMN
    LMN is offline
    mtbr member
    Reputation:
    Join Date
    Sep 2007
    Posts
    3,469
    It would be interesting to see how controlled his study was.

    I have done a lot of testing of various bikes and components over the years. One of the things that astonishes me is how much my own weight can drop from trial to trial. If my trial are an hour apart, even after hydrating, I am using about a 1lb lighter for the second trail. (Which means I have to add balast to be accurate).

    I am curious because because I haven't seen that type of drop before from static weight in my testing. And if you put numbers in Analyticcycling.com on that length of hill with an average grade 10%, a 70kg rider who puts out 300 watts of power should only gain 8 seconds by losing a 1lb.
    "The best pace is suicide pace, and today is a good day to die." Steve Prefontaine

  9. #9
    Rock and/or Roll
    Reputation:
    Join Date
    Dec 2005
    Posts
    439
    Quote Originally Posted by Mr.Magura View Post
    Keep in mind, that the study makes some pretty rough assumptions.

    Pedaling efficiency for one, is not going to be equal on a noodle 2 lbs frame, and on a super rigid 4 lbs frame.
    The same could be said about many other components.

    Another assumption would be the loss of power due to the condition of the trail is not playing a part, which is also highly questionable.

    So if two bikes perform exactly equal, under the conditions of the trail and so forth, yes, the lighter bike is faster.

    To need a study to reach that conclusion, well, ahem, is pretty much a joke.

    So if we can have components and frames, that performs equal, but are lighter, all is good.
    If not, the 22 seconds mentioned above,and then some, are going to be lost real quick
    Magura
    I wouldn't make those assumptions. I have a 2 lb. carbon frame that is very stiff and efficient, and a steel framed bike that is very flexy and soft in comparison.
    I also have very lightweight carbon cranks, that are far stiffer than a heavier set of aluminum cranks.

    I would say that comparing my two bike examples, the lightweight bike is faster going uphill just due to better engineering, not just weight alone.
    An efficiently designed and built bike is going to take you uphill quicker, than a flexy, inefficient one. I would go even further to say that if the carbon bike was 2 lbs. heavier than the steel bike, it would still get me up the hill quicker.

  10. #10
    DIY all the way
    Reputation: Mr.Magura's Avatar
    Join Date
    Aug 2010
    Posts
    2,411
    Quote Originally Posted by turbogrover View Post
    I wouldn't make those assumptions. I have a 2 lb. carbon frame that is very stiff and efficient, and a steel framed bike that is very flexy and soft in comparison.
    I also have very lightweight carbon cranks, that are far stiffer than a heavier set of aluminum cranks.

    I would say that comparing my two bike examples, the lightweight bike is faster going uphill just due to better engineering, not just weight alone.
    An efficiently designed and built bike is going to take you uphill quicker, than a flexy, inefficient one. I would go even further to say that if the carbon bike was 2 lbs. heavier than the steel bike, it would still get me up the hill quicker.
    You missed my point

    It is a matter of comparing apples to apples. This meaning that a study that does not take this into account, is pretty much useless.


    Magura

  11. #11
    DIY all the way
    Reputation: Mr.Magura's Avatar
    Join Date
    Aug 2010
    Posts
    2,411
    Quote Originally Posted by LMN View Post
    It would be interesting to see how controlled his study was.

    I have done a lot of testing of various bikes and components over the years. One of the things that astonishes me is how much my own weight can drop from trial to trial. If my trial are an hour apart, even after hydrating, I am using about a 1lb lighter for the second trail. (Which means I have to add balast to be accurate).

    I am curious because because I haven't seen that type of drop before from static weight in my testing. And if you put numbers in Analyticcycling.com on that length of hill with an average grade 10%, a 70kg rider who puts out 300 watts of power should only gain 8 seconds by losing a 1lb.
    You got a good point there.

    I see no mention of that factor in the study either.
    This kind of stuff as basis for a master thesis, has to be a joke.


    Magura

  12. #12
    DIY all the way
    Reputation: Mr.Magura's Avatar
    Join Date
    Aug 2010
    Posts
    2,411
    Btw, it is real strange that nobody has pointed out the lack of keeping the test double blind?

    If you add a 1 kg water bottle to a bike, I would claim that most people would be able to tell, by just tilting the bike from side to side a couple of times.

    So now we have a flawed study, with lack of control, and the placebo effect working full throttle.


    Magura

  13. #13
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070
    Quote Originally Posted by Mr.Magura View Post
    You missed my point

    It is a matter of comparing apples to apples. This meaning that a study that does not take this into account, is pretty much useless.


    Magura
    By using the same exact bike and only adding 2 lbs, it IS comparing apples to apples. bike rigidity/efficiency remains constant. The only variable is weight.

    Anyone seeking to realize the benefits of a lighter bike in the real world has to modify the bike in such a way so as to minimize or avoid a loss of bicycle function/efficiency.
    Last edited by Stumpjumpy; 01-14-2012 at 08:54 AM.
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

  14. #14
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070
    Quote Originally Posted by LMN View Post
    It would be interesting to see how controlled his study was.

    I have done a lot of testing of various bikes and components over the years. One of the things that astonishes me is how much my own weight can drop from trial to trial. If my trial are an hour apart, even after hydrating, I am using about a 1lb lighter for the second trail. (Which means I have to add balast to be accurate).

    I am curious because because I haven't seen that type of drop before from static weight in my testing. And if you put numbers in Analyticcycling.com on that length of hill with an average grade 10%, a 70kg rider who puts out 300 watts of power should only gain 8 seconds by losing a 1lb.
    That's a great point.

    This study used a 16.5% grade. So that theoretically could be responsible for the more substantial delta in time
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

  15. #15
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070
    Quote Originally Posted by Mr.Magura View Post

    So if two bikes perform exactly equal, under the conditions of the trail and so forth, yes, the lighter bike is faster.

    To need a study to reach that conclusion, well, ahem, is pretty much a joke.


    Magura
    The study attempts to quantify how much faster a lighter bike is, not simply demonstrate that a lighter bike is faster.
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

  16. #16
    DIY all the way
    Reputation: Mr.Magura's Avatar
    Join Date
    Aug 2010
    Posts
    2,411
    Quote Originally Posted by Stumpjumpy View Post
    The study attempts to quantify how much faster a lighter bike is, not simply demonstrate that a lighter bike is faster.
    Analyticcycling.com gives the figure based on relatively simple math. The figure you get there, is exactly the figure you would see in a test, if the test was conducted correct, and the number of people involved, was big enough to make a statistic significant.

    To think an experiment like the one from the study, is getting a more correct figure, is at best flawed. As already mentioned above, the test is based on quite a number of assumptions, and is not even close to a double blind test.


    Magura
    Last edited by Mr.Magura; 01-14-2012 at 09:04 AM.

  17. #17
    DIY all the way
    Reputation: Mr.Magura's Avatar
    Join Date
    Aug 2010
    Posts
    2,411
    Quote Originally Posted by Stumpjumpy View Post
    By using the same exact bike and only adding 2 lbs, it IS comparing apples to apples. bike rigidity/efficiency remains constant. The only variable is weight.
    That would be true, if it wasn't for the fact that the test takes a situation, and turns it upside down, by adding weight.

    I for one have never met somebody whom were interested in adding weight to their bike.

    So, in real life situations, the part that is interesting, is when the curve of speed gained by loosing weight, crosses the curve of equipment performance reduction due to weaker equipment.

    If that study should have any relevance, it would have been a lot more complicated, and have involved quite a bit of measurements of the equipment.

    Just the fact that the energy loss due to frame flex, is measured as a percentage, should be enough for the thesis to have flunked.

    Take for instance the guy with the biggest energy loss due to frame flex, and compare him to the guy in the opposite end of the scale.
    Now adding 2lbs to the guy with the noodle frame, is going to make a significant impact on the result, compared to the guy with the rigid frame.

    If it was all that simple, I guess it would have been done a long time ago, and you would find websites with "speed calculators".

    Analyticcycling.com is the closest we are going to get to a simple answer, that is not flawed to a degree, close to wild guessing.


    Magura

  18. #18
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070
    I don't think the author ever claimed his study was definitive or better than anything else at quantifying performance vs weight.

    Like any study, it is just another datapoint, among many, to consider in drawing any conclusion re weight vs performance.
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

  19. #19
    DIY all the way
    Reputation: Mr.Magura's Avatar
    Join Date
    Aug 2010
    Posts
    2,411
    Quote Originally Posted by Stumpjumpy View Post
    I don't think the author ever claimed his study was definitive or better than anything else at quantifying performance vs weight.

    Like any study, it is just another datapoint, among many, to consider in drawing any conclusion re weight vs performance.
    Trouble is, that it has little to do with data.

    It's the test itself that is flawed, so whatever conclusion he draws from a flawed test, is of no value to anybody.


    It's "tests" like that one, being brought forward as scientific, that causes a lot of the religious opinions around.
    If nobody cares to point out the flaws and challenge flawed conclusions, the bike segment is gonna end up worse than the Hi-Fi guys, and that I think nobody besides the industry, will see any benefit from.


    Magura

  20. #20
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070
    Every study involving human athletic performance is flawed by definition. This does not render the results useless, however. Again, the studies are data points, that when taken as a whole (meaning all of the studies) provide useful information for drawing useful conclusions.
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

  21. #21
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070
    FWIW, I just ran the same parameters through analytic cycling.com and it generated a 19 sec delta instead of this study's 21 sec delta. Seems pretty consistent to me.

    But keep in mind that the following criticism applies to ANY estimate/study on weight vs. performance, no matter how calculated, so they are both equally totally invalid and worthless.

    Quote Originally Posted by Mr.Magura View Post
    Keep in mind, that the study makes some pretty rough assumptions.

    Pedaling efficiency for one, is not going to be equal on a noodle 2 lbs frame, and on a super rigid 4 lbs frame.
    The same could be said about many other components.

    Another assumption would be the loss of power due to the condition of the trail is not playing a part, which is also highly questionable.

    So if two bikes perform exactly equal, under the conditions of the trail and so forth, yes, the lighter bike is faster.

    To need a study to reach that conclusion, well, ahem, is pretty much a joke.

    So if we can have components and frames, that performs equal, but are lighter, all is good.
    If not, the 22 seconds mentioned above,and then some, are going to be lost real quick


    Magura
    Last edited by Stumpjumpy; 01-14-2012 at 06:49 PM.
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

  22. #22
    mtbr member
    Reputation:
    Join Date
    Apr 2006
    Posts
    2
    21 second difference does seem a bit high. When you add 1kg to 80kg rider+bike, the increase is about 1.25%. You would expect to go about that much slower: 1.25% of 10 minutes is about 7-8 seconds.The table listing the results shows that there are a few outliers, riders who went close to 1 minute slower with the extra weight.

    There is an interesting discussion after the table of results, where he says that people tend to go hard on the flats and take it easy on the steep parts. I always thought I go harder on the steep sections because I suffer more.

  23. #23
    mtbr member
    Reputation:
    Join Date
    Nov 2004
    Posts
    425
    That article was written to an audience that has never ridden a bike before. Then they threw in some big words. NOT impressed!

  24. #24
    mtbr member
    Reputation: brentos's Avatar
    Join Date
    May 2006
    Posts
    1,036
    To be fair, it's better than I would have done when in school. And it is interesting to read.

    The procedure is actually pretty good; if executed as described. A larger sample size would mitigate a lot of the variables some of you point out (mostly by randomizing run order). It really isn't written very well (a lot of filler), it was predisposed to an outcome, and there was no statistical rational...but the raw data is there, if not significant at these sample sizes.

    I'd like to see it repeated with a much larger sample size (5x to 10x larger), then evaluated for true statistical significance.
    Race Reports, PreRide Reports, and General Rambling:

    www.roostersbikersedge.com

  25. #25
    Helmetless Crasher
    Reputation: Stumpjumpy's Avatar
    Join Date
    Sep 2011
    Posts
    1,070
    Quote Originally Posted by kallee View Post
    21 second difference does seem a bit high. When you add 1kg to 80kg rider+bike, the increase is about 1.25%. You would expect to go about that much slower: 1.25% of 10 minutes is about 7-8 seconds.
    Your calculations seem to omit grade, which is obviously VERY important. Analyticcycling.com, which is apparently considered the best calculator by some in here, generates a 19 sec difference.
    '95 M2 StumpJumper FS
    '11 Cannondale RZ 120-two

Page 1 of 2 1 2 LastLast

Similar Threads

  1. My bike was stolen,need help on what to get
    By Realslowww in forum Clydesdales/Tall Riders
    Replies: 9
    Last Post: 02-19-2012, 06:15 PM
  2. I built a singlespeed. help me with this tensioner?
    By CommuterBoy in forum Singlespeed
    Replies: 72
    Last Post: 02-08-2012, 08:43 PM
  3. New here and looking to get a bike...
    By BigRedNole in forum Beginner's Corner
    Replies: 51
    Last Post: 01-17-2012, 02:16 PM
  4. I'm looking to take some weight off my bike....
    By Bro in forum General Discussion
    Replies: 32
    Last Post: 01-15-2012, 10:34 PM
  5. More Bike Mag atricles in Brevard
    By Logover in forum North & South Carolina
    Replies: 1
    Last Post: 01-10-2012, 08:54 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •