I was wondering ... most if not all cyclo computers measure the speed when moving, correct? Well, I tend to roll up slowly to red lights, hoping to get the green before I have to unclip. Would that result in a lower average speed vs going up to the light more quickly and having a longer complete stop? I'm still going to try to time the light but just curious.

Thanks

It can affect your "average speed": for every second spent at X-mph below your average speed, you have to spend a second X-mph above your average speed to maintain the average. If you zoom up to the light (above your avg speed), then stop, your "average speed" will be higher (every computer I've seen stops counting when the speed drops to 0-mph). You could also roll up to the light (at X-mph below average), and then when it turns green, bump up your speed to X-mph above average, and then settle back down to your average to maintain it.

However, this is only the average speed when you're moving. To more accurately calculate your average speed over the whole ride, you have to take the distance you traveled and average it out over the total time of your ride. For this "true average speed", it will make no difference (or very little difference) whether you stop at the light, or roll in and then take off again when it turns green.

All I know is that my average speed is too damn low.

Don't worry about it too much. It is a crappy metric to use.
My favorite saying is: average speed makes an average rider.

That being said, many computers will stop when you drop down below say 1 mph. Others have settings to enable this.

unless your average speed is 23+mph, you have no place on BF

On some computers you can set a speed to go on auto pause. I've set mine to 3mph so when I walk with my bike it does not count as moving time.

Different computers stop recording at different times -- a matter of seconds. Don't sweat it. But if you want a true measurement of average, get a computer with a manual option -- meaning to have to start the measurement when you begin your ride. It will then actually adjust AV, including the times you are stopped and give a true average, according to total riding time, rather than omitting those pauses.

We'd see a lot lower "average speed" reports if everyone used manual settings.

Randonneurs clock rides by this method. Now if we could just get a manufacturer to make a computer that compiled data for, say, 100 hours.

Usually when I'm recording information about my rides, I'll note the time that I started and the time that I finished, figure out how long I spent on the ride, and then calculate the actual average speed from that. Not that difficult, really...Would be nice to have it done automatically in the cyclocomp, of course, which would avoid a little bit of tedium in the calculations (converting ride-time to minutes, etc.).

I have mine set to count stops in my average speed. I've been doing lots of longer riding and the total time and actual average speed is whats important for planning future rides.

Several CatEye comps have manual setting; Enduro 8 for instance. 10 hours recording.

The Edge 705 lets you turn off the time when you drop below a user definable speed. I have mine set to 3.5 mph.

Whether or not you want to include stops in your average depends on what kind of ride it is, what your goals are, and what you want the information for. Although as people have pointed out, average speed is not a good metric for performance, if you are trying to see if you are "improving" on a given course, including stoplight time is kind of nonsensical. If you have to wait 2 minutes one day and 3 minutes the next day, the amount of rest you are getting is not going to meaningfully impact your performance, but the extra minute is going to make your total take longer.

Also, for example, if you are on a group ride, the amount of time you may have to wait at a regroup for slower riders isn't really meaningful.

Mine has two averages. Actual average and BF average. The BF avearge can be set from +10% to +50%. It also gives bonus credit for "cranking it up to 400w" when you see attractive members of the opposite sex.

I agree, that's why I use 2 magnets.

my wireless computer drops 10 mph or so every time i ride under a high voltage power line... avg speed is a fairly useless measurement anyhow...

I have to disagree that average speed is a useless metric. Sure, one cannot compare average speeds across different routes but it's a decent way to gauge if one is improving.

My routes are in urban areas so there are many stop signs and red lights. Thus, my question is probably more relevant than for riders in mainly rural areas, or where there are fewer interruptions. Unfortunately my computer is an el cheapo model where I cannot define a slow-speed cutoff point. I think next time I ride, I'll try slowing down more abruptly to see if my average goes up.

The real reason for my post is not so much to change the way I ride but to make sure comparisons with others are accurate.

Thanks for all the feedback.

We didn't say worthless.
Crappy <> worthless
Just that there are far better metrics to use.

Exactly

I commute so I have the same route every day. I don't obsess about stops at lights or higher wind days. The most important metric is looking at a graph of my average speed over time and seeing if it goes up as my endurance/strength improve.

If you have a ride you do frequently, measuring your average speed during that ride over time will give you a sense of rate of improvement. There are just too many factors that influence speed to have the speed of any single ride be meaningful.

Not sure mine ever goes up. Sure goes down, though, after a few days off the bike.

My computer records average speed, total ride time, and total distance. Dividing distance by time gives me an average speed that is different than the one the computer reports. Go figure. It does stop the time when the bike stops.

Average speed is useful for me - a 51 yr old newb...

i do more or less the same route about 4 or 5 times per week - 25 miles with some decent (for connecticut) climbs.

when i started out on that route in about march or april, i was struggling to be in the mid 13 mph (moving average) now, i'm nearly at a 16 mph moving average. at my level, i'm not sure what else i need to know...

all the stopping and slowing for lights etc. averages out. i would say it more or less averages out for each ride, but even if it doesn't, it surely averages out over a few rides.

i use a garmin etrex vista hcx, and i figure differences of 0.2mph in average speed are significant.

one reason average speed is useful for beginners, is that group rides are organized by average speed, so it's easy to know which group you should go with - i find i'm faster with a group anyway.

sure, advanced riders and racers may really get into a more detailed analysis of performance data, but i can't really be bothered.

This is a good reason average speed should not be used. I've seen allot of strong riders who have poor group riding skills. If your new to group rides, start with a slower group to learn the skills before moving up.

Average anything (Speed, Cadence, HR) would only be of use if you did the same thing over and over. Even with uncontrolable enviormental variable like stoplights, traffic, and wind, it you did the same thing over and over long enough in the end a properly plotted graph could yeild data.

Change the course, time of day, etc, or look at a small data set, and the information is useless.

My club does not use "average speed" to classify groups. We use distance, course difficulty, and regrouping to designate. I get asked all the time what the average speed of the A Group is, and the answer is always same - "too fast for you if you have to ask."

This graph is the plot of the ratio of moving average speed to elapsed average speed for my 8.5mi commute for the past 5.5yrs. It means nothing, just TMI.

A few notes: Each blue point is a single commute. Dark red line is 10 commute average. Average ratio for the past 6yrs or so is 1.08. Over the long term the step changes where you can see the ratio changes is due to route adjustments. The occasional peaks are due to longer stops for errands, flat tire, etc.