What kinda cadence would be 'ballpark average' for a competative rider going all-out up a long climb? I know, 'everyone is different' and all that, but what would be a good median target to train around if you were revising your climbing style, trying for better race results?

More particulars: (delete now if you find me too wordy) This off-season (until mid-Dec) I'm again trying to learn to ride a higher cadence. I've made half-hearted attempts to increase my climbing cadence before, but I've never really tried to 're-vise' my cadence with a serious program. I've been working on it now for about a month and it's becoming easier, spinning faster. Perhaps I've even improved a tiny bit in uphill speed, it is so hard to tell..My past attempts to change from a masher to a spinner, I have soon reverted to my slower cadence because I've felt a little faster as a masher.(but not quite fast enough to hold the leaders on climbs).This time, I am going to give it a serious go and see for certain whether there are any gains to be made, for me.

Most climbs I've been finding myself settling in at about 60 rpms in the past..When I upped my average climbing cadence in the past, my breathing seemed stressed..Lately, after about a month of cadence drills, I've been riding the same hills at around 80-90 rpms and feeling ok..same speeds, close as I can tell..My legs are not feeling so sore after a hilly workout. So it seems like it may be working..I need about a 2% uphill speed improvment to put me up in the leaders. Right now, on a downhill, I start to bounce(spin-out) at around 115 rpms. My bike fits properly. I just went from a 172.5 back to a 170 crankset.

So if everything was perfect and you had 'great legs' and an unlimited number of cogs, what cadence would be "perfect" for your best speed uphill? Or said differently..what range of cadence is everyone riding when climbing well?
Thanks,
Don Hanson

I haven't run a cadence meter in a couple years but when I did, I was around 75 rpm when climbing solo. That's the speed where I feel like I am "on top" of the gear and not struggling. In groups however I tend to ride at a higher rpm to be able to accelerate quickly should it be needed.

A couple years ago I did some experiments to see what cadence was most efficient for me. On a long stready climb I rode at the same speed with different cadence, then looked at the heart rate for each test. Below 75 rpms the heart rate went up. That's opposite of what's expected. My interpretation was that I am less efficient at lower cadences.

Since then my legs have gotten stronger and I have done some low cadence climbing to strengthen them. As a result I can handle lower cadences a little better than I could before.

I try to keep my cadence around 90 to 95. When it drops to around 85, I tend to shift to a lower gear. Of course I don't climb that fast, but I love climbing and keep working on it. Bring on the hills!

So if everything was perfect and you had 'great legs' and an unlimited number of cogs, what cadence would be "perfect" for your best speed uphill? Or said differently..what range of cadence is everyone riding when climbing well?

I can't help thinking you're trying to determine your best shoe size by asking everyone what size they wear. I'd suggest a better line of questioning would be how to determine the best cadence for you to climb at given the particlular goals you're trying to achieve. If you focus on the process rather than the result, I think you'll get a better answer.

A couple years ago I did some experiments to see what cadence was most efficient for me. On a long stready climb I rode at the same speed with different cadence, then looked at the heart rate for each test. Below 75 rpms the heart rate went up. That's opposite of what's expected. My interpretation was that I am less efficient at lower cadences.

You might have learned something about how heart rate responds to contraction rate, but there is no relation between heart rate and efficiency. The only way to measure efficiency is with a power meter and a gas analyzer.

You might have learned something about how heart rate responds to contraction rate, but there is no relation between heart rate and efficiency. The only way to measure efficiency is with a power meter and a gas analyzer.

You don't need a power meter when talking about a climb. Power is essentially proportional to speed.

-ilan

What kinda cadence would be 'ballpark average' for a competative rider going all-out up a long climb? I know, 'everyone is different' and all that, but what would be a good median target to train around if you were revising your climbing style, trying for better race results?

More particulars: (delete now if you find me too wordy) This off-season (until mid-Dec) I'm again trying to learn to ride a higher cadence. I've made half-hearted attempts to increase my climbing cadence before, but I've never really tried to 're-vise' my cadence with a serious program. I've been working on it now for about a month and it's becoming easier, spinning faster. Perhaps I've even improved a tiny bit in uphill speed, it is so hard to tell..My past attempts to change from a masher to a spinner, I have soon reverted to my slower cadence because I've felt a little faster as a masher.(but not quite fast enough to hold the leaders on climbs).This time, I am going to give it a serious go and see for certain whether there are any gains to be made, for me.

Most climbs I've been finding myself settling in at about 60 rpms in the past..When I upped my average climbing cadence in the past, my breathing seemed stressed..Lately, after about a month of cadence drills, I've been riding the same hills at around 80-90 rpms and feeling ok..same speeds, close as I can tell..My legs are not feeling so sore after a hilly workout. So it seems like it may be working..I need about a 2% uphill speed improvment to put me up in the leaders. Right now, on a downhill, I start to bounce(spin-out) at around 115 rpms. My bike fits properly. I just went from a 172.5 back to a 170 crankset.

So if everything was perfect and you had 'great legs' and an unlimited number of cogs, what cadence would be "perfect" for your best speed uphill? Or said differently..what range of cadence is everyone riding when climbing well?
Thanks,
Don Hanson

There cannot be one best RPM, because it will depend on the incline or grade. In general (I doubt there are exceptions), rpm's will go down as the grade becomes steeper. Even Lance Armstrong was not doing 90rpm on 20% grades.

-ilan

You don't need a power meter when talking about a climb. Power is essentially proportional to speed.

-ilan
Do much climbing with a gas analyzer strapped to your back? Anyway, all you're saying is you're using a speedometer as a very crude, inaccurate power meter.

The reality....try and maintain the same cadence you would on flat terrain going the at the same intensity level. That's basically where you are most efficient and where you have trained your body to be efficient.

Going higher or lower with cadence puts your body into a different situation that it's used to. With that said, some climbs you can't help but go with lower cadence due to lack of low gearing or too steep of a climb.

If you want to use a higher cadence start training for one, if you want to use a lower cadence then start training for one. Otherwise, just stick to what your body is used to.

So if you are a spinner and usually run 90-110 RPM on the flats going hard, shoot for that on the climb (and have access to gearing that allows it). If you are a masher and run 70-85 RPM on the flats going hard, shoot for that on the climb.

That's basically where you are most efficient and where you have trained your body to be efficient.

Can we please get away from talking about efficiency? It's at best tangentially related to performance and possibly counterproductive. Not to mention it's imposiible to measure outside a carefully controlled laboratory environment.

It has been well known for a long time that the most efficient cadence falls somewhere in the low 60 rpm range, but by any measure of performance the best cadence will be much higher than that.

Can we please get away from talking about efficiency? It's at best tangentially related to performance and possibly counterproductive. Not to mention it's imposiible to measure outside a carefully controlled laboratory environment.

It has been well known for a long time that the most efficient cadence falls somewhere in the low 60 rpm range, but by any measure of performance the best cadence will be much higher than that.

Remember....I never said where you are most effecient in a lab, I said where your body has been "trained" to be most effecient. There is a difference.

If you are used to pedaling at a specifc RPM, it's best to stay at that RPM while climbing.

Remember....I never said where you are most effecient in a lab, I said where your body has been "trained" to be most effecient. There is a difference.

Right. One is the proper use of the word "efficiency" (work produced per energy consumed); the other is misapplication of the term to mean "better" or "best."

Using tools commonly available to normal cyclists, how would you design a test to determine what one's best or most efficient climbing cadence is?

Using tools commonly available to normal cyclists, how would you design a test to determine what one's best or most efficient climbing cadence is?

You could simply time yourself up a known climb at "constant" heart rate in different gears/cadences. However, that would be of only partial use, because you adapt to the cadence you use, and the only way to find out if you are faster with a higher or lower cadence is to train that way, and train that way for months (if not years). Such a test has been reported repeatedly in the literature, but it has no meaning. Your best bet is to look at what top level racers do. 80-90 rpm is most common. Next time you're watching a climbing stage in a big race, time the cadence of the riders. You'll get some data about what people who have optimized their pedaling have concluded.

You could simply time yourself up a known climb at "constant" heart rate in different gears/cadences. However, that would be of only partial use, because you adapt to the cadence you use, and the only way to find out if you are faster with a higher or lower cadence is to train that way, and train that way for months (if not years). Such a test has been reported repeatedly in the literature, but it has no meaning. Your best bet is to look at what top level racers do. 80-90 rpm is most common. Next time you're watching a climbing stage in a big race, time the cadence of the riders. You'll get some data about what people who have optimized their pedaling have concluded.


That'll just tell me what is best for other riders, not what it best for me. There is a large difference between (for example) Lance Armstrong and Jan Ullrich's climbing cadence.

My question was mostly aimed at asgelle, who is telling us that the sort of test I wrote about is not useful (I understand that it could be better with a power meter, but I did not have one at the time). I am curious what tests that an ordinary cyclist could perform would be useful for determining either "best" or "most efficient" climbing cadence.

This won't answer Don's question, or mine, but is a lengthy discussion on optimal cadence:
http://www2.bsn.de/Cycling/articles/cadence.html

You are going about the wrong way. The simple answer is put a heart rate monitor on, and pedal at a rate that allows you to stay at around 70 - 75% of your max, if the goal is to just climb without killing yourself. (If you can't stay in this zone, you need lower gears than what you got on your bike).

If you are training then go ahead and pedal where your heart rate gets to close to 99% of your HR max :) for as long as you can take it (i.e. interval training). After a few months of doing that you will be able to climb better without exploding.

Using tools commonly available to normal cyclists, how would you design a test to determine what one's best or most efficient climbing cadence is?

The only way to determine most efficient climbing cadence is to measure energy consumed and power produced. There are many publications describing the protocol for doing this but they all center around controlling power and cadence and measuring oxygen consumed. By looking at power produced divided by energy consumed (VO2), efficiency can be calculated as function of cadence.

As to determining "best" cadence, that really depends on how you define best. Kerry pretty well covered the case where best is defined as fastest, but that is not necessarily the only definition. In general the test would be to measure performance (however that is defined by the individual) as a function of cadence carefully controlling for other factors and performing enough tests to minimize statistical errors. Then alter cadence during training for a period and repeat the test to determine the effect of adaptation on the results.

You don't need a power meter when talking about a climb. Power is essentially proportional to speed.

-ilan


You pretty wrong here.

My power numbers can vary quite a bit at the same speed depending on if I am standing or sitting, on the hoods or on the tops and what gear I am using. Aside from 20 min intervals and the like training my climbing with power is where I find the PM most useful.

Thanks.
I'm not asking for "THE magic NUMBER" that will make me climb like Rasmussen. Just the ballpark cadence. 80-90 is probably close enough. I went back over my Garmin TC data and see that I've picked up my cadence about 10 rpms average since I began my revised cadence training.

Here's the thing..You can gear your bike to let you pedal (~) any hill at any cadence. I see a lot of different climbing styles, but I don't usually have time or the will to count other rider's cadence while I am chasing. To confuse the issue further, one fellow I ride against, a record-holder on some substantial uphill time trials, grinds along at near 40 rpm in his big ring and still beats us all, while others spin at 120 rpm in maybe a 34-27 and do well, too.

Yesterday I rode one of my benchmark local hills, 7 miles with a 1700' elevation gain and had a cadence average of about 70 rpms against a stiff headwind. Previously I've usually done this same hill with about a 55-60 average cadence. So it seems I am making some headway at upping my cadence. My times were close, using my HR to put in my max sustainable effort..So perhaps I am on the right track. With a bit more high rpm training, I probably see some small improvment..which is what I am trying for.

I've tried timing myself up a known climb for data. Last season I couldn't decide whether my 585 Look or my DeRosa Dual was the better climbing bike, so I did 12 runs up an 800' section of a climb, alternating bikes and using my Garmin's Lap timer to start and stop the clock each time. Averaged the times for both bikes...Virtually a tie..3 seconds faster on the Look over a total of an hour..not significant..or informative.

A power meter would certainly give me more data, but at my level of racing, the gains I might find using one, weighed against the expense and learning curve make that seem a bit of a distraction.

Tjhanks for the input..Shooting for about 80 rpms now on climbs that allow that using a compact and a 23-25 cog..Steeper=slower pedal stroke..
Don Hanson

You could simply time yourself up a known climb at "constant" heart rate in different gears/cadences. However, that would be of only partial use, because you adapt to the cadence you use, and the only way to find out if you are faster with a higher or lower cadence is to train that way, and train that way for months (if not years). Such a test has been reported repeatedly in the literature, but it has no meaning. Your best bet is to look at what top level racers do. 80-90 rpm is most common. Next time you're watching a climbing stage in a big race, time the cadence of the riders. You'll get some data about what people who have optimized their pedaling have concluded.

Thanks.
I'm not asking for "THE magic NUMBER" that will make me climb like Rasmussen. Just the ballpark cadence. 80-90 is probably close enough. I went back over my Garmin TC data and see that I've picked up my cadence about 10 rpms average since I began my revised cadence training.

Here's the thing..You can gear your bike to let you pedal (~) any hill at any cadence. I see a lot of different climbing styles, but I don't usually have time or the will to count other rider's cadence while I am chasing. To confuse the issue further, one fellow I ride against, a record-holder on some substantial uphill time trials, grinds along at near 40 rpm in his big ring and still beats us all, while others spin at 120 rpm in maybe a 34-27 and do well, too.

Yesterday I rode one of my benchmark local hills, 7 miles with a 1700' elevation gain and had a cadence average of about 70 rpms against a stiff headwind. Previously I've usually done this same hill with about a 55-60 average cadence. So it seems I am making some headway at upping my cadence. My times were close, using my HR to put in my max sustainable effort..So perhaps I am on the right track. With a bit more high rpm training, I probably see some small improvment..which is what I am trying for.
Don Hanson

I usually naturally settle to a cadence between 105 and 115, regardless if I'm climbing or gong on flat, at 120 I start bouncing on the bike, and if I increase the cadence even more (I can pedal up to 145 for a few minutes) I stop bouncing. But 105-115 feels perfect to me.

My power numbers can vary quite a bit at the same speed depending on if I am standing or sitting, on the hoods or on the tops and what gear I am using.
If your power varies with body position, then aero drag is providing a significant retarding force rather than gravity dominating. Most would say that doesn't qualify as a significant climb.

If your power varies with gear selection, clean and tune up your drivetrain.

Everyone seems to have an opinion on this subject, so I'll give you mine:

Cadence makes very little difference in how efficiently or how much power you can generate in a climb so long as you're pedaling within a reasonably comfortable range, say somewhere between 60 and 110 rpm. The bottom line is you can only sustain the power that you can sustain and cadence (within a reasonable range) has little affect on power output.

If your cadence gets so low that you have to stand, that's a whole 'nuther ball game since your aerodynamics and ability to generate/sustain power may change.

To answer your original question, I seem to climb best with a cadence around 80 to 90.

If your power varies with body position, then aero drag is providing a significant retarding force rather than gravity dominating. Most would say that doesn't qualify as a significant climb.

If your power varies with gear selection, clean and tune up your drivetrain.

You and I will have to disagree the climb i use for tests is over 6% and so is a climb my drive train is very clean and my PT calibrated.

Aero drag is always a factor over 15mph but flutucations of 10 watts or more are possible at the same speed due to body position due to the efficancy of the person in that position. I am very ineffeciant when standing so it takes more watts to keep the same pace up that I would sitting.

You and I will have to disagree the climb i use for tests is over 6% and so is a climb my drive train is very clean and my PT calibrated.

Aero drag is always a factor over 15mph but flutucations of 10 watts or more are possible at the same speed due to body position due to the efficancy of the person in that position. I am very ineffeciant when standing so it takes more watts to keep the same pace up that I would sitting.

If you're seeing a change in power with different gear ratios, then power loss in your drive train is excessive. All drivetrain losses should be 1-2% of total power and the change in drivetrain loss with gear selection should be an order of magnitude lower. Are you really saying you can see changes in power of less than a Watt?

The grade of your hill has nothing to do with whether ot not it's a significant climb. What's a major climb to a beginner is barely a bump to a Pro. The point is if you are going fast enough to see power changes with position, then by definition aero drag is significant and the grade is not steep enouh to be a significant climb since the definition of one is where gravity dominates.

Finally, whether or not you are efficient standing or sitting has nothing to with how many Watts it takes top go a certain speed. You might check on the definition oef efficiency.

You pretty wrong here.

My power numbers can vary quite a bit at the same speed depending on if I am standing or sitting, on the hoods or on the tops and what gear I am using. Aside from 20 min intervals and the like training my climbing with power is where I find the PM most useful.

Well, I don't think I'm completely wrong (obviously power and speed are not the same, you can climb by repeatedly coasting which gives vast variations in power but not speed). However, if I qualify the statement by considering average power, then I think I am essentially correct. From repeated rides over a number of climbs, my average power up the climbs was quite clearly proportional to the average speed up the climb (AKA 1/time up hill). You will note that I did use a power meter, and yes, it was very useful. I can now estimate my average power for these climbs before I got the power meter.

However, I don't understand how you can register different power levels going at the same speed over the same terrain by pedalling at a diferent rate.

-ilan

you can climb by repeatedly coasting which gives vast variations in power but not speed
You don't slow down when coasting uphill?

You don't slow down when coasting uphill?

For example, if you coast for 1/10 of a second then you won't slow down very much, but your power output will drop to zero for 1/10 of a second. In other words, your power output will have changed by a factor of 1/0 (which is larger than any positive real number), while your speed will have changed by a less significant factor (that is, close to 1, e.g., 1.01). I don't see why I have to justify every correct statement which I make.

-ilan

If you coast for 1/10 of a second then you won't slow down very much, but your power output will drop to zero for 1/10 of a second. I don't understand why I have to state this explicitly.

-ilan
The reason I thought you should state it explicitly is that in writing it out you might see that you are wrong. Now I have no idea how big "very much" is, but I do know that if it's a steep climb, you will decelerate at a = mg sin(theta). I also know that even though you might not see it on your speedometer, all those tenths of a second add up. In fact, they add up so that the change in speed is directly proportional to the change in power and in the end P= mgvsin(theta)

The reason I thought you should state it explicitly is that in writing it out you might see that you are wrong. Now I have no idea how big "very much" is, but I do know that if it's a steep climb, you will decelerate at a = mg sin(theta). I also know that even though you might not see it on your speedometer, all those tenths of a second add up. In fact, they add up so that the change in speed is directly proportional to the change in power and in the end P= mgvsin(theta)

OK, thanks. It's hopeless.

-ilan

I usually naturally settle to a cadence between 105 and 115, regardless if I'm climbing or gong on flat, at 120 I start bouncing on the bike, and if I increase the cadence even more (I can pedal up to 145 for a few minutes) I stop bouncing. But 105-115 feels perfect to me.

What sort of gearing do you run?

For example, if you coast for 1/10 of a second then you won't slow down very much, but your power output will drop to zero for 1/10 of a second. In other words, your power output will have changed by a factor of 1/0 (which is larger than any positive real number), while your speed will have changed by a less significant factor (that is, close to 1, e.g., 1.01). I don't see why I have to justify every correct statement which I make.

-ilan
Actually, it's the incorrect ones you should focus on. If you were climbing at 250W and then coasted at 0, the fractional change from your steady climb would be Delta P / P. Delta P = 250 W (250-0), baseline P = 250 W so fractional change is 250/250=1.0. Now if you want to talk about accelerating from coasting, we can explore that, but that's not the profile you described. You're welcome.

What sort of gearing do you run?

53/39 and 12-25. On my old (early 1990s) steel bike I have 53/42 and 12-25.Oh and by the way, I'm rarely using the 12.

Geeze, it must be pretty tough terrain that I ride most often because when I run 53/39 and a 12/23 or -25 I find it almost impossible to maintain much over 60 rpms. I can sprint there for a minute or five, but then I die...I recently mounted a compact for a 'sick' race, 29,500' of climbing over two days of racing, and even there, with a 27 tooth cog as a lowest ratio, I had to stand on the steepest pitches of about 15%, and on the average grades of about 10% I seem to have been stuck at around 70 all day, both days...

Yeah, at 6% with a 25 cog I can hold 70 rpms up a long long climb, but if people are spinning up climbs with similar gears at 115 rpm, then either the "climbs" must be shorter or flatter than I see regularly..

Anyhow, If I shoot for 80-90 cadence for a ballpark, that will be upping my 'strokes' about 20 rpms and maybe I will see that needed 2% increase in my uphill speed that I need to be at the front..Thanks
Don Hanson

The only way to determine most efficient climbing cadence is to measure energy consumed and power produced. There are many publications describing the protocol for doing this but they all center around controlling power and cadence and measuring oxygen consumed. By looking at power produced divided by energy consumed (VO2), efficiency can be calculated as function of cadence.

As you point out elsewhere, measuring energy consumed is difficult. Normally in a lab one would measure oxygen consumed and extrapolate energy assuming aerobic respiration, and that works pretty well. But an individual on his bike can't make that test. That is why I used heart rate. HR does go up with increased oxygen demand, if you account for the lag effect. I don't say that it is a test that will give you a quantified result, but it should tell you if one of two tested cadences are more efficient for you. It'd be better to do it with a power meter so you can hold power constant. Of course you need a steady grade and you need to climb with the same style.

Note that the link I posted elsewhere in this thread shows that while efficiency is generally best at 50-60 rpms in both untrained and trained subjects, with trained subjects it does not change much as rpm varies.


As to determining "best" cadence, that really depends on how you define best. Kerry pretty well covered the case where best is defined as fastest, but that is not necessarily the only definition. In general the test would be to measure performance (however that is defined by the individual) as a function of cadence carefully controlling for other factors and performing enough tests to minimize statistical errors. Then alter cadence during training for a period and repeat the test to determine the effect of adaptation on the results.

The hard part is defining "best". Knowing that Don rode the Everest Challenge and that race being a special interest of mine, I was assuming "best" meant best over a stage race with a lot of climbing. That's the sort of thing my test was intended to tell me. "best" for a single race ending with a 500' climb would be different.

Geeze, it must be pretty tough terrain that I ride most often because when I run 53/39 and a 12/23 or -25 I find it almost impossible to maintain much over 60 rpms. I can sprint there for a minute or five, but then I die...I recently mounted a compact for a 'sick' race, 29,500' of climbing over two days of racing, and even there, with a 27 tooth cog as a lowest ratio, I had to stand on the steepest pitches of about 15%, and on the average grades of about 10% I seem to have been stuck at around 70 all day, both days...

Yeah, at 6% with a 25 cog I can hold 70 rpms up a long long climb, but if people are spinning up climbs with similar gears at 115 rpm, then either the "climbs" must be shorter or flatter than I see regularly..

Anyhow, If I shoot for 80-90 cadence for a ballpark, that will be upping my 'strokes' about 20 rpms and maybe I will see that needed 2% increase in my uphill speed that I need to be at the front..Thanks
Don Hanson

The hills I have here are shorter indeed. Steep at times, but very short. Just so you don't think I'm some sort of superhuman (or dishonest :) )