Changing gearing after changing crank length

I’ve recently changed crank length from 167.5mm to 160mm based on the advice of my bike fitter.

That’s also involved a change of drivetrain. I’ve changed from SRAM 10-33 / 48-35 to Shimano 11-34 / 52-36 but the easiest ratio is identical, so I don’t think it’s a factor.

When climbing higher gradients in the easiest gear, it feels more difficult, I’m guessing because the peak torque figures are higher because of that shorter crank, hence the gravitation towards higher cadences with shorter cranks (I’m up ~10rpm)

What I’m trying to work out is the torque figure at a given weight/gradient/crank length/gear ratio/cadence to understand when I head off on a climbing holiday whether the weight I’ve lost is enough to compensate for the lower crank length or I need to go for smaller chainrings.

I’d rather avoid smaller chainrings and losing even more top end, hence wanting to quantify rather than just take a blind stab at it.

I know torque is pretty easy to calculate given power and rpm, but it’s that weight/gradient component I need. Is there a calculator/modelling software out there that can work this out for me?

I think this will do what you want. Enter in weight, gradient, speed, etc to get required power. Then use your old gearing to calculate cadence, then crank length to calculate torque. Then enter lower weight and use the new lower power with new gearing and crank length, to calculate torque.

1 Like

Basic physics says if you shorten your cranks 10%, you should reduce your chainring size 10% to keep the same overall gain ratio.

But it turns out it’s not just one simple ratio because the leverage of your joints changes within the stroke, and what seems to happen with shorter cranks is that you use more of the more powerful part of your stroke and trim off the less effective ends.

So that then leads to a rule of thumb that you only drop your gearing by half the change. This was covered in Episode 9 of the BikeFit Podcast which unfortunately seems to be hard to find. Dr. Jim Martin, the researcher behind most of the crank length studies, mentions this near the end of the podcast.

However in episode 11 of the same podcast, John Cobb of time trialing and seat fame talks about using shorter cranks and says the people he puts on shorter cranks needed bigger gears! He also mentions it here.

So what should you do? The first thing is to give it a couple months to get used to it. Second is to not get hung up on higher cadence - you can pedal at a higher tempo but you don’t need or want your foot to be going at a faster linear velocity.

Also consider that shorter cranks effectively steepens your seat tube, shifting muscular load to the quads and away from the glutes and that adjustment might not work in your favor when climbing.

When I set up my short crank test mule and went from 175 to 155 I kept the same 38 front ring but the wheels I used had an 11-42 instead of an 11-46. Despite the total loss of 20% leverage, I actually found climbing steep stuff easier! But it was also weirdly hard to pedal easy for a couple months. Now it all feels normal.

So I’d say don’t try to engineer a solution, give your body a chance to adapt. Also try your seat back a bit, or maybe a little lower so you ride a bit more like a weight lifter than a ballerina.


Thanks for the information, that’s all super helpful and interesting.

I’ve been riding these shorter cranks for 3 months and 3000kms or so now, but there’s little in the way of climbs here, hence trying to solutionize in advance of my two week cycling holiday where there will be plenty of sustained 30-60 minute climbs.

Interesting what you said about the quads, I really struggled with the additional load on them initially.

Yes, but you don’t need complicated math. Going from 170 mm to 160 mm corresponds to a reduction of 6 %. If you want to put the same power to the pedals at the same cadence, you need to increase torque by about 6 %, too. If you adjust your gearing and keep torque constant, then you can raise the cadence by the same ~6 %.

I went from 172.5 mm to 165 mm, and I definitely took that into account when choosing gearing: I made my gearing a little easier, I opted for a 42-tooth chainring as opposed to a 44-tooth chainring on my 1x12 road setup. I noticed that my self-selected cadence drifted up by a few percentage points, pretty close to the 4 % difference.

In a sense, it is the deciding factor: you did not change gearing even though you changed crank length. With a 2x setup, it is harder as you have less gearing options. But I’d go for a compact crank. It isn’t enough to compensate for the difference in crank length exactly, but covers 2/3 of the way.

Yes. It is important to distinguish between gear ratio and the overall ratio that determines the torque you have to exert (I reckon that’s what you mean by “overall gain ratio”).

1 Like

They’re all first order effects so if the holiday roads are 10% steeper but you lost 10% of total bike/rider weight those cancel out. If you want to maintain the same force at the pedal then you’d change the leverage to make up for the difference between the grade change and the mass change.

The thing is though that the shorter cranks do tnot automatically mean less power. So while you can do the math, using just leverage isn’t really putting a real number into the model. So if you have a local small hill as a baseline use the difference in grade only - you’re already at your new weight and decently adjusted to the crank arm length.

However if you want to gear down based on the numbers just for peace of mind, why not? It’s your holiday and you shouldn’t be stressing having too big a gear.

The trouble is, I don’t want to gear down. I’ve already lost some top end losing the 10t of the SRAM, I don’t want to lose more!

It’d be an easy choice otherwise. The problem is I think I need to gear down, but want to avoid it if possible.

Margins are tight, so it’s not like it’ll ruin my day if it’s wrong, I was just looking to quantify my decision to maximise the chances of making the right choice :slight_smile:

You’ve lost 0.9kmh at 100rpm and 1kmh at 110rpm, not enough to lose sleep over.


A further 2.5km/h to be lost by going down to a 50x11, but then that’s regained by just an extra 6rpm, which I’ve gained in self-selected cadence.

It sounds so irrelevant put in those terms, but the last time I had a 50x11 (on 165mm cranks) it was noticeable and frustrating to feel like I’d be spinning out on some hills I’d usually keep pedalling on.

I’ll give it a try

1 Like

As my grandpappy used to say, you can’t coast uphill! (1)(2)

If it makes you feel better, pedaling downhill is a waste of energy, you’re blowing a bunch of energy fighting a power rule.

(1) he didn’t
(2) at least not for very long


Completely agree.

I’m going to buy some smaller chainrings and try it, just a shame 12s chainrings are so pricey.

Is that a matter of pride or a serious problem? Even at a leisurely 90 rpm, not accounting for the fact that your self-selected cadence did increase according to you, we are talking about 53 km/h in 50:11 vs. 55 km/h in 52:11. How often are you at those speeds on relatively flat ground?

Plus, the thinking is flawed, because you are forgetting that, like you reported in your post, your cadence increased proportionally to the difference in crank length (= leverage).

Climbing gears are much, much more important than overdrive gearing.

I don’t know about your hills, but “my” hills are of two types: The first kind has lots of curves and I have to break either to stay safe (traffic) or to take the curves. The second kind is the “any gear will spin out” hill.

Before going for a 1x12 on my road bike (with a 42-tooth chainring, i. e. the hardest gear is 4.20, way “easier” than 48:10 = 4.80 or 52:11 = 4.73), I had a compact crank and a few loaners with 52:11. It made precisely zero difference in practice.

My local terrain is rolling, so it’s not unusual to spend a decent amount of time close to that top gear.

But you’re right, it’s not that important and more an issue of how it feels than worrying about losing speed.

1 Like

It’s a very real feeling that you’re missing out on something, no doubt! I can give you a pile of anecdotes about why you’ll actually miss very little speed, but in these situations I also like to say it’s your bike and it should work the way you want it.

If you want a 53-11 big gear even it’s just “so I can slow pedal to get blood back in my legs” or “I love pedaling down hill because it’s the closest I can get to flying” if you’re willing to put the time/money/patience to get the results you want, it’s your bike!

1 Like

If your spider pattern is the 105/Ultegra/Dura-Ace X110mm BCD for 11-speed chains, you can go down to 33t. I’m running 52 & 33 on a 105 system without trouble, though I have put a 0.5mm thickness shim washer between the 52 & the spider. That’ll give you better low gears without losing any more top end. This is the one I’m using. You should be able to find a seller nearer to you:

I did also for a while try replacing the 165mm 1-piece crank on my cheap 90s mountain bike with a 115mm from a kid’s BMX. I had to pedal at 140rpm in a suitably lower gear to keep the same foot speed & mechanical advantage on the back wheel. I was 16. It was the cheapest way to get bigger gears (or the illusion thereof) when I was stuck with 48/38/28 chainrings & a 5-speed 14-28 cluster. Gave me a pretty mad top speed though. And it meant I actually used the granny ring.

Is there any reason you can’t use the shorter Shimano-system cranks & the Sram wheel, cassette, RD, right shifter, & chain? :smiling_imp:

Sounds similar to the terrain I am riding now. What cadence do you top out at? Personally, my limit is 120ish rpm, which is good for 63–65 km/h. Would it be nice to have another gear? Always! But climbing gears are more important to me.