Sy Reene

So I should know this, but I don't.. simple question. Is Average Cadence as reported by a Garmin unit, based on time, or based on distance traveled?

fietsbob

Cadence is the tempo of your feet..

Sy Reene

Right, but I'm asking about how AVERAGE cadence is calculated over a ride when reading that stat on a garmin

Chris Bamford

Would it not be based on time?

Average Pedal Revolutions per minute, as opposed to Average Pedal Revolutions per mile.

Sy Reene

I don't know.. is it just time, or time and distance? Maybe an example illustrates the question?
I ride on a flat course for 2 miles. Pedal consistently at 85rpm. Simple answer, my 2 mile ride, the average cadence was 85rpm.

OTOH.. I cycle up a 2 mile hill at 85rpm at 10mph and it takes me 12 minutes. I then coast/bomb down the same 2 mile hill at 40mph, no RPM (because I'm coasting) so it only takes me 3 minutes. Is "average cadence" then calculated as:
85rpm x 2 miles, and ZeroRPM x 2 miles = 42.5rpm, or
(85rpm x 12 minutes) + (0 x 3 minutes) = 1020rpm/15 minutes = average of 68rpm

Chris Bamford

While I profess no direct knowledge of the Garmin Cycling system, I have to think that Cadence, being expressed as rev/min, has to be time-based.

Simple: Count all the revolutions following Start, divide by all the minutes passed since Start.

Imagine the computation required to base it on distance... 8 rpm for 7 ft, 12 for 7', 19 for 7', 27 for 7, 38 for 7, etc etc for every one revolution of the front wheel. Figure them out for every revolution, remember them all, add them up at the end of a 40 mile ride (30,000 +/- data points) and come up with an average. Not very likely!

Sy Reene

Makes sense I guess. I'm just trying to get it in my head how much I value the "average cadence" statistic. I think ideally, I would want to at least have a statistic that told me that when I'm pedaling, what I average for my cadence then (0 rpm not part of equation). Or alternatively, over an entire ride, what my cadence averaged over that entire distance. I would guess (?) that just calculated over time, the tendency is to overinflate or not really provide a good picture of how much or how often you pedal at a given cadence (unless your rides are over really flat terrain). Or other statistical possibilities.. eg. mode -- the most common cadence that one cycled at.

Cychologist

On my Garmin 800 the average cadence is pedal strokes per minute while moving. There is a setting in tools where you can have it exclude the time you were coasting.

So it is a measure of rpm's, revolutions per minute, over the entire ride.

Homebrew01

How is avg cadence over a ride useful ? Real-time cadence is barely even useful for a few people.

deapee

I could see real-time being useful for someone trying to get a feel for what 80 or 90 RPM feels like...but yeah, the use of average cadence over a minute -- I wouldn't find particularly useful at all.

Cychologist

I don't keep my eye on all the data my Garmin gives me over a ride, but analyse it after the ride. My goal is for my average cadence to be above 75 rpm's.

For me, average cadence (and power) are more importance than average speed.

Sy Reene

Exactly; I'm talking about post-ride analysis. Looking back I've had comparable rides at similar average MPHs, but with more greatly differing average cadences (eg. 70 vs. 85). I was trying to get a handle on how average cadence is looked at. I'm trying to get rid of a tendency to spin up to good speed, coast down a bit, spin up again and so on, but rather instead just keep the pedals moving more often. Knowing how garmin calculates avg cadence with the first ride style is helpful.

Marcus_Ti

Depends on what average you use.

If you're talking median pedaling average, like Stravistix shows, it is fairly useful.


Ah, the good-old pedal/coast shens.

What did it for me, was having someone or something yell at me when I stopped pedalling. Garmins these days can start chirping an alarm if your cadence drops to 0. Stravistix is also nice in the extra cadence tool info possible. The cadence avg on Garmin isn't that informative, I'm honestly not sure how it calculates its number either, even with 0-averaging turned off.

Having something like Stravistix that actually will tell you by %age exactly how much of a workout you are "cheating" yourself can be fairly motivating.

intransit1217

I find it useful when I connect it to another thread about where most pro cyclists generate their best power within a range.

Granted, I'm no pro, but I can still use it to hone my own skills. I can compare rides and see that usually a higher cadence corresponds with a higher speed.

I find it very useful.

BenPS

afaik, it's average cadence based on the amount of time the cadence sensor is being triggered. So if you are coasting, that time is not factored in to your average cadence

Marcus_Ti

Depends on how you have the device configured. There's zero-averaging: on and zero-averaging: off.

What average Garmins display, mean/media/mode and the method to derive it is AFAIK anyone's guess.

wphamilton

Ride up and down a single hill and see what it says! Or better yet find a flat road, pedal at 90 for 30 seconds then coast for 30, and see if it says 90 or 45.

FWIW I believe from observation that the instantaneous readings for both the speed and cadence is likely calculated with a Gaussian smoothing over the past several seconds, with "zero" readings discarded. Smooths out erroneous spikes that way, better than just averaging 3-4 seconds.

redcon1

Correct. My Garmin Edge devices, if paired with a Cadence Sensor, exclude zero readings when you are not pedaling by default in the averaging. You can change the settings to include zero readings in the average.
(I know it's that way for 500 and 520 - I would assume the others but it can be found in the user manual).

Also from the Garmin manual:

Cadence: Revolutions of the crank arm per minute.
Cadence - Avg: Average cadence for the duration of your current ride. (So also time based)

Seattle Forrest

Hubs full of angry bees.

Marcus_Ti

Have had King hubs and Campy hubs. Never bothered me.

Did bother me once I saw real data graphs of my pedalling and how much "pedal and coast" actually is "coast and pedal a little bit". Having auto-warning of 0 cadence from a bike computer drills it in how little one actually pedals. One might *think* you are mostly pedalling and only coasting a little bit-but IRL the numbers don't lie....odds are at best pedal-and-coast results in 60%+ coasting


If you commute in MUTs or streets with lots of stoplights....very easy to condition yourself into pedal/coast, due to the constant slowing/stopping.

Seattle Forrest

Really?

I just exported my last ride into TCX format, queried it in SQL Server, and the average I calculate excluding zeros matches what my Garmin says exactly.

Mine is set to exclude zeros (for cadence, include them for power) so my query excludes zeros.

Sy Reene

Interesting info everyone. Thanks. I guess it would interesting if there was a way to get raw data and analyze any way you want (ie. after the ride). For example, take a ride of a certain distance, let's say 60 miles, and then have the raw data of how many revolutions total you spun over that 60 miles, in theory you could calculate a average cadence per yard or whatnot.

I'll have to check into how I have my Garmin treating 0 rpms. I haven't found a way to change the setting in the Garmin Connect PC analysis or it's export to Zwift -- so am I correct in assuming this can only be adjusted in the actual GPS unit, and not in the software?

CliffordK

A lot of devices will zero out zeros. So, stopping at a stop sign shouldn't impact your average speed.

Coasting while moving would be a bit different, but it would seem like adding it into your average cadence would give you bad data. Or, ideally it should be configurable. Average moving speed vs average trip speed. Average pedalling cadence vs average moving cadence vs average trip cadence??? Does standing at a stop sign while not spinning count?

I tend to pedal faster going downhill than uphill (although sometimes slowly going over the peak).

Sy Reene

As I mentioned previously, I'm trying to consciously make a better effort at pedaling more often and over more of the ride. Eg.. I may have had a tendency to spin up to 100rpm and say 20mph, then coast (0 rpm) down to 16mph, and so on. At the end I might get an 80rpm "average" cadence with an 18mph average.. while I want to strive to pedal more often right at 85-90rpm or so almost all of the time. I'm not exactly sure what I'd need to read consistent results that tell me how well I'm moving in this direction.

EDIT: I'm sure this is where someone tells me to get a power meter. Yeah, I know..

CliffordK

Do you use Strava (or a similar app), and are you able to upload power/cadence?

With Strava, you can select time or distance, and look at changes in power, speed, etc.

For Speed, I think Strava looks at both overall (in a segment) and moving (ignoring segments).