I've been working on my cadence for several months now, and I feel like I've gotten to the point where I'm more comfortable using high cadence instead of standing up when I want to accelerate hard.

At stoplights if I'm really gonna accelerate hard, I'll start in first gear and shift up when I get to 140-150 rpm.

I stay seated the whole time. I feel like I'm faster this way than I am standing up, but I'm not sure. Until recently, I've always assumed that standing up in a higher gear was the best way to put on speed fast.

Can someone provide data or logic that points one way or the other?

  • 2
    It depends on the individual, but generally if you're over about 110 RPM you're wasting too much energy in simply spinning your legs. But this can get to be a contentious issue. (Saw a study of this about 25 years ago in the LAW magazine and lab tests indicated that over 100 RPM is probably not justified.) – Daniel R Hicks Jun 29 '15 at 21:42
  • I've been also working on rising my cadence and in keeping it for longer period of time. I can accelerate at almost the same rate while seated than ride buddies that accelerate standing on pedals. The same is true for short climbs: I do it faster seated and on high cadence than my mates who do it standing. I guess that if we used each others technique for a test, the results will be very different. – Jahaziel Jun 30 '15 at 0:58
  • 2
    If you are comfortable using a technique and you get good overall results, then don't change it but continue to improve it. – Jahaziel Jun 30 '15 at 1:00

The only way to be sure is to measure.

You can use an app such as Strava during the ride, then look at it's analysis later. It will show you how fast you were going at each point, and also gives an approximation of your power output.

We don't know what algorithm or assumptions it uses to calculate the power, but since it uses just one algorithm you can compare results for seated and standing sprints. Strava doesn't know what the wind speed is, nor does it know what your frontal area is (which changes when you stand), so the power values are not correct in an absolute sense. The speed measurements are more useful.

I would expect that you'll be producing most power at the higher end of your cadence range, 120-140 rpm.

In my experience a standing sprint is good for a short burst, rather than the extended acceleration you describe.

Using the higher cadence will be better for your knees over time.

  • To get a decent measure of acceleration from a standing start with any software you'll need bike-mounted sensors. At a minimum a wheel rotation speed sensor like on a standalone bike computer. GPS accuracy and especially update frequency aren't enough over the timescales you're interested in. – Chris H Jul 5 '15 at 8:15
  • @ChrisH Certainly bike mounted sensors would be more accurate; it depends how detailed the requirements are. It's a question of what is good enough for the OP's needs. – andy256 Jul 5 '15 at 9:08

Through personal experience, I have found that the higher cadence method will always get me ahead of the group of commuters at the lights. I sometimes see the standing grinders but by the first downstroke of their crank arm, I am already gaining much more acceleration. The grinders are also wobbling all over the place because their centre of mass is much higher and the bike is harder to keep in a straight line.

I think you will need to experiment with which gear you start in while accelerating in the saddle.

I commute on a Trek FX 7.2 and have found my most effective starting gear is using the middle front chain ring (the 38 of 48/38/28) and the largest rear sprocket (the 32 of 11-32, 8 speed).

Once I am spinning at a cadence which is around 80% of my physical limit, I then momentarily relieve the tension in the chain and change one gear (change up rear sprocket). It takes a bit of practise to get this right but it avoids the gears clunking while you shift.

Another benefit of starting in the 38 front chain ring is that I can accelerate to beyond my cruising speed in this chain ring. If I were to start in the 28 chain ring, then I would need to lose some acceleration while making the shift to the 38. A shift between front chain rings is much more time-consuming between a shift between rear sprockets.

I also make sure that my saddle height is correct and on the limit of being too high. With a lower saddle, it will be near impossible for your legs to transfer power with a high cadence. When I am at the lights, I need to fully extent my leg for my toes to reach the road. While I do this, I keep my other foot on the pedal in the ready-position. As soon as the lights change, I apply power and my extended leg has to 'catch' the spinning crank arm.

  • +1 you have several good points there but I think andy256 gets closer to answering my question. – BSO rider Jul 2 '15 at 22:32
  • No problem. I was just trying to elaborate, qualitatively. – Scott Jul 3 '15 at 7:35

The answer is probably the combination of standing up and high cadence.

I assume that when you say acceleration you mean hard and short burst of effort. The maximum power in that short period of time will be produced by the fast twitch muscle fibres type, (i.e. when you stand up, low cadence). This type of fibres work anaerobically so you can only sustain it for very short time.

If you need to accelerate longer than what the anaerobic system can provide then you will need to switch to aerobic using the slow twitch muscle, i.e. high cadence.

Even if you measure and find that high cadence is faster, it could also probably mean that you haven't train your anaerobic system enough to get the expected results. That's why sprinters have huge thighs, they go to gym for resistance training.

There's a table with muscle characteristics here: Fiber typing


Get a power meter and see what output you are getting at various rpm / loads. There is a big difference of 120rpm at 200w or 600w. This relationship will most likely not be linear.

150 rpm seems much too high, but everyone is different.

I personally find great power at 105-120.

Timing yourself or using HR can be a measure as well, but power is the only way to know for sure.


As already stated, measure. But maybe not the whole commute, because it introduces confound variable like traffic light timing, other road users, hot or rainy weather. I would rather get a friend with a stopwatch and spend a couple of hours on an empty parking lot (if that simulates your most common riding conditions, maybe a forest clearing or unoccupied forest track else).

Have markers for where you start from and about 30 meters later the "finish line".

Do factor in the final gear you end up in. Id est if your steady-state gear is X, after this sprint you are going to need to spend time to get to that.

Do factor in that high forces wear your knees out. In the extreme case, you can stand up and pull the handlebars towards you, allowing crazy acceleration. And knee wear.

Do make 5-10 measurements for every type of riding so that you can compute mean and variance.

And get back to us. This experiment sound fascinating!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.