's Forum Archives - General

Archive Home >> General(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )

Calculating Average Speed?(23 posts)

Calculating Average Speed?JIM LUKENS
Jul 16, 2001 5:39 AM
When people are quoting their Average Speed, is it most often a rolling average or an overall average? When I ride, I take breaks fairly often because it's very hilly here and I'm still getting in shape. I just calculate my average speed as Miles Ridden divided by Total Time on the ride.
Then I look at some of the average speeds that people post and think "Man am I pathetic". Which, compared to most on this board, I am. But compared to when I started, it's light years' better. But I'm not sure how those average speeds are being calculated.
I kinda like the total time approach because I can see improvement just from getting in better shape and not needing as many breaks.
re: Calculating Average Speed?Len J
Jul 16, 2001 5:45 AM
Most Bike computers calculate avg speed as Total miles riden /total time the Bike is moving. (Ignores rest stop time). Some people also will not turn the computer on until they are warmed up. You hit the nail on the head though, and that is consistancy. As long as you always measure it the same way you can tell if you are improving.

Good luck.
re: Calculating Average Speed?david l
Jul 16, 2001 5:47 AM
I calculate it the same as you...and I have the same reaction when I read about posted ave. speeds. But then I AM pathetic. I may not live long enough to get back in shape at this rate. I just enjoy the struggle and I like improving, too.
re: Calculating Average Speed?Lone Gunman
Jul 16, 2001 6:23 AM
Not pathetic at all. Varies from computer to computer also. I found out recently when forced to ride the backup bike with the old 'puter attached that it was measuring my ave by overall time, not time in motion and was really slow. Then I switched on the auto start/stop feature and my ave went back to normal. My flight deck on the regular bike does the ave calc for you without choosing a mode. It really starts out with the calibration before use. You could put in a larger tire size (I think this is correct) into the calibration and presto, you are as fast as LA. :-(). And absolutely, turning the 'puter on late and shutting it down early will give you a faster average time. Overall, ave time is time in motion
Don't worry...Jim Burton
Jul 16, 2001 6:14 AM
Quite a few "averages" posted are, as a weeks prior post stated, sort of like fishing stories. It's easy to say, "I averaged 25mph for the entire ride," but not so easy to do as people let on. If you are riding by yourself, you have no help or rest from the draft. Sharing the wind resistance will dramatically increase your attainable average. If you've never ridden in a group before, you would have no way of knowing this. Quite a few people also calculate (or more like guesstimate) a different average than thier computer is displaying. And, like another post said, some people don't set the computer until after warm-up, which means after they have to slow down for red lights and stop signs and are on open road.

Really, don't worry about your seemingly slow averages. If you are improving, that is all that matters. Be patient.
and some of them only turn 'em on on downhillscyclopathic
Jul 16, 2001 11:13 AM
doesn't matterDog
Jul 16, 2001 7:23 AM
As said above, this is a figure that doesn't matter too much. Even if we agreed on a common way of calculating, that tells you nothing about hills, wind, drafting or not, etc.

It is just as valid to quote rolling or total average speeds. But, they are telling you two different things. I quote total average, as that is the only thing that counts in any event, unless I quote both, e.g., 100 miles in 6 hours total, with rolling average of 20.5 mph. That's pretty common.

Don't reduce this bike thing to all numbers, especially if you are sort of new to it. That shifts the focus from where it should be, IMO (even though I do it a great deal, too).

doesn't matterJIM LUKENS
Jul 16, 2001 7:42 AM
I can't help but get caught up in the numbers - I'm an accountant and numbers make me feel warm and fuzzy. Seriously, I like using it to track my progress, and I can't help comparing myself to others sometimes, as hard as I try not to. It also gives me an idea of how long a certain route will take so's I can tell my wife about when to expect me back. That way if I end up as a hood ornament, somebody can come looking.
It used to matter to meterry b
Jul 16, 2001 8:45 AM
I really like to play with numbers and I religiously (in Excel) keep all the mileage and time data from my rides. I used to think average speed (time spent riding, excluding stops) was a decent indicator of my improving condition. Then I got in the habit of entering a "comments" field in my spreadsheet to explain ride results, both good and bad. It finally dawned on me as I was entering comments that how fast I was riding was pretty meaningless - more effected by wind, traffic, route, temperature, roller bladers, dogs, etc. Instead, it became obvious I was gaining fitness by the mileage I was adding to my typcial ride. Used to be 25, now it's 55. Used to be tired, now it takes 100 to really tire me. To date, I have over 200 rides in my DB and the average is 17.677 mph - interesting since I spend most of my time riding between 19 and 23 mph. That number no longer changes except in the 3rd or 4th decimal place. Being an accountant so you understand how averages work (I'm a statistician) and realize that eventually averaging your speed over lots of rides will bring you to a number that will never change. If you truly want to measure your improving fitness, you need to pick a fixed route that is absolutely immune from interruptions and ride it many times taking the total time for your ride (including breaks) and plot the total time. And, as an adder you could include your average mph for time ridden (only) as an indicator of whether or not you are getting stronger and hence able to propel the bike faster. I like to look back on when I first started riding again (age 40) and think how I used to be seeing spots and gasping at the end of a fixed 10 mile loop around my neighborhood. Now I barely crack a sweat on the same ride. To me, that's the only way to look at this problem.
doesn't matterdavid l
Jul 16, 2001 10:32 AM
You're right - I rode my old ['89] C/dale today for about an hour. It has no computer. Great ride.
Compare yourself to me and feel better ...Humma Hah
Jul 16, 2001 7:30 AM
... I ain't fast, and don't mind admitting it. My rolling time average (from my cyclocomputer, which stops timing when I stop moving) is about 15 mph on a GOOD day on my favorite suburban training ride, a little over 20 miles, with 25 traffic lights and a couple of stop signs. I've done an informal 20-mile "TT" at 16 mph (no traffic lights, fairly flat, but much jogger/blader/stroller traffic to avoid).

I can cruise at 18 mph, but just a few traffic lights or other stops will blow the heck out of average speed.

If you use overall time, that really throws in a lot of variables. On some routes, overall can be almost identical to rolling time (saddle time). For most urban/suburban routes, however, overall time can be several mph slower.
A Bike Computer Program in the works...Mabero
Jul 16, 2001 9:40 AM
As a mathematician I too spend too much time dwelling in the numbers calculating rates of change, interpolating my distances to find patterns...and have I found? A lot of consistencies and a lot of inconsistencies mainly with Avg. Speed.

I dwell in Avg. Speed and is the first thing I want to look at the end of a ride, but as I have found it doesn't incorporate the variables. Therefore having some kind of ride description for the day really helps you see your progress. I know that I can go 60-70ish mile route and ride faster than I would if I went a different (hilly route) for a 50ish loop.

So it really comes down to me looking at what I did for that day, how much did I really deviate from my average from that day, etc, etc...

I like to look more at my Avg. Distance per ride. This is interesting cause you see your monthly averages grow over that's a good indication of how well you are riding (in the general scheme of things).


On another note I am writing a windows based application for Logging our data. Considering how much I hate excel I figured it would be so much easier for bikers/runners to have a program specifically for them that would store and give results of our data. Plus it will (has) the ability to plot the data and do numerious statistical analyzation on the data for all us number geeks.

Anyone interested? I am looking for my suggestions as in for input fields...anyone interested in demoing a finished copy in a month or so?
Ride Environment has a dramatic impact on speed.........Len J
Jul 16, 2001 10:33 AM
Sat I did 70 miles in generally flat terrain. The first 35 was dead into a 20 mph headwind. I averaged 15.4 mph and my heartbeat never got below 160 (85%of max) Turned around at the 35 mile mark & now had a 20mph tailwind. I "averaged" 23.5 mph coming home (same route,same lights(very few)) however my heartbeat stayed between 152 & 158 the entire ride home. So I actually worked less but had a higher speed. Overall I averaged 19.5 but clearly I had two different rides.
Exactly...Two completely different rides on same terrain (nm)Mabero
Jul 17, 2001 4:14 AM
A Bike Computer Program in the works...rtolle
Jul 16, 2001 11:49 AM
Mabero, the bike program sounds cool. I would like to demo w/my bike team if the opportunity becomes available.
Sure thing!Mabero
Jul 17, 2001 4:08 AM

Sure. Most likely when I get it done or at least a working specimen of a finished product I'll just post it here. I don't think it's going to be a big application so it would be quick to download. Basically once I get the basic functionality sound and "bug" free I will just post. Unfortunately at that point it won't be finished cause I know the graphical extraction from our bike data will take me a lot longer.

Either way it should be an easy program to just use without the hassle of using/learning excel.

Even if it is used as an analyzer and excel is used to display the data, I still think it would be a good program to use. As it is right now I have a program that I use that I wrote in a program called Matlab. The program is great but I couldn't tell someone how to use it!


As a decidedly "left brain" individualPaulCL
Jul 16, 2001 2:20 PM
I would love to demo your program when available. Since I am in the finance/investment business, I deal with numbers all day long. My wife says I need to write poetry so that I can develop my "right brain". NOT! I am addicted to my bike computer output.

As for fields: the obvious first: avg speed, distance, time, about altitude? terrain (scale of 1 to 10, 1 = flat, 10 = tomorrow's tour route), weather, alone or in a group, how many alcoholic drinks you had the night before?????
Good ideas...but altitude?Mabero
Jul 17, 2001 4:13 AM
I have incorporated the basics with a lot of statistics. I also have a field for a ride description to be able to write in "60 degrees, sunny" and stuff like that.

Having a food field for what you ate and how much before, during, and after is a good idea, actually is a great idea! Also it would be simple to rate the terrain with a scale, but with altitude are you saying having a field that would store the amount climbed?

ps. Hopefully I can have a demo version out in a few months as I am the doing this in my spare it gets kind of hard!
A Bike Computer Program in the works...mk_42
Jul 16, 2001 3:33 PM
I too am a number geek--let me know when/if you come up with something and I'll be happy to beta it or if you want a co-conspirator I could write some code. What are you writing it in?


To support my claim to geekness I'm using this post to test how this bulletin board treats images--please just ingnore any image or HTML you may see below.
There may be an exciting post by me coming up.
----Ignore stuff below here----
I forgot to mention...mk_42
Jul 16, 2001 3:56 PM
I forgot to mention anything about me. I know perl, and I do DB stuff for work. I also have some graphics experience. Again, if I have anything you're missing on this project let me know.


I'm (still/again) testing how this bulletin board treats images--please just ingnore any image or HTML you may see below.
There may be an exciting post by me coming up.
----Ignore stuff below here----
A Bike Computer Program in the works...Mabero
Jul 17, 2001 4:01 AM

I am writing it in MFC so consequently Visual C++ 6.0. Right now I am testing the basic functionality and will start to incorporate screens later this week. I have coded right now about 15 classes to deal with, speed, distance, time, date, ride descriptions, daily input, stored input, stored statistics, getting stats such as standard deviation, variance, average...and a bunch of weighted averages.

Once I test some the backdoor code, I am going to do screens on the side. But what I really need to work on is loading the data and developing a class that would handle files. This way you could load a month or a week or a year. That will be the long part for me.

I don't think in the demo version I will have any graphical options cause that is going to take me some time.

So far it's been a fun project and you know what? I think it's about time to have a program for us!

Bike Happy.
Some of my best rides...coonass
Jul 16, 2001 4:07 PM
were when the 'puter WASN'T working....I never knew when I wasn't reaching my normal 'average speed' or how slow I was going up the hill....just enjoying the ride and BEING THE BIKE!! If you're riding with your 'buds, then staying with them is the ride, plus you'll less likely burn-out when you DON'T know that you're riding about 2-3mph faster than you're accustomed to....Try it....
Stop the Clock.Dutchy
Jul 16, 2001 5:52 PM
Half of my rides are in the city (~20 sets of traffic lights), riding to work 2-3 times per week and the other half are in the country. I always stop the clock at traffic lights, busy intersections, train crossings. This way I get an idea of my true average speed. If I kept the clock ticking on my weekly commute, some days I would drop 5kph due to all the traffic lights and trains. When I ride in the country, the ride is uninterrupted except for 1-2 (5min food stops), the only reason I stop is so don't get run over while trying to open a powerbar with full length gloves in winter. So this is why I keep a rolling average, because I try to avoid stopping.
I don't normally ride more than 110kms/70miles so I don't stop as often as the guys that do "mountain centuries/doubles".
Whether you stop the clock or not, as you get fitter your average will still increase.