It’s a quark powermeter so dual sided. And you have “difference between what the powermeter reads vs true power” which I don’t really care about and you’ve got “identical rides = identical numbers (or not) which I really care about. Do you think the precision is enough to suss out a 2% difference in power? How could I even test that?
My off road lap is about 850 seconds…..is 1 second resolution enough for a lap that long? I had to drop undergrad stats so it’s not exactly my thing lol.
Joe
It’s not just about the total time. It’s about how big of a difference you are seeing between the tires vs the time resolution and accuracy. The difference was <3 seconds on average. Since your resolution in the time measurement is 1 second, the error just from the time measurement can be as big as 1.8 seconds. So a 3 second measured difference might be 1 or 5 seconds actual.
You generally want your difference in results to be around 10 times your resolution. So if you are using a time resolution of 1 second, then you need about 10 seconds difference in the results of the two tires. If we assume the time difference scales linearly (it might not due to differences in the new route’s terrain vs current route), you need about a 2400 second (40 minute) route (since you want to overshoot especially since some tires will have less difference). A longer test of course makes it even harder to control other variables.
Alternatively, if you can get more accurate timing (0.1 seconds resolution and accuracy), you could keep your current test route. But that would require some automated timing solution to not have reaction time be a significant noise factor.
It’s so much easier to find something wrong with a an experiment or analysis than to do them correctly. I’m no expert in statistics, but I do lots of testing and am good at finding flaws.
OK so that makes sense…I hit the button early one lap and late on another lap and I’ve lost 2 seconds of my 3 second time difference. And 1 second is well within statistical insignificance (or however you say that). I think I’m stuck with 1 second resolution…at best.
OK so on my off road lap is about 850 seconds and my time resolution is 1 second. I ran 3 laps with a total spread of 6 seconds (859, 856, and 860) yesterday (although power varied from 162 to 164 watts). Let’s say I can run 4 laps and I’m within 10 seconds for tire A and 10 seconds with tire B, all at the same power. Is there a way to know how much different the lap times for the tires would need to be to be able to show a difference?
Ha ha so you much go insane looking at rank amateurs post their youtube videos about how fast one tire is over the other! I know I do!
Just as an aside, I’ve had many days where the times seem to get faster as the hours go on. It could be air density or the track drying out or any number of things but it’s not unusual for the “slow” tires to run at the same speed late in the day as the “fast” tires early in the day. If you compare lap to lap the “fast” tires are always faster but if you compare lap 8 vs. lap 2 the “slow” tires can be faster. Realistically I am going to take 25 minutes to do a 14 minute lap (getting back to the starting point, swapping wheels) and usually the first lap is abnormally slow so if I am going to get 4 laps on each set of tires it’s gonna take like 4 hours. And that is assuming no laps are throwaways (people, dogs, really craptacular line choice, etc.) where I have to slow down.
I guess the first question is … is this even possible? It seems like it should be but the more I learn about statistics and the more laps I run the more questions I have.
Joe
Everything points to really wanting 0.1 second data collection. So you could see if there’s a phone app that can record GPS data at that rate and include a small amount of overlap in your lap start and stop. Then you could get very precise lap times in post analysis of the GPS data rather than rely on reaction times.
You would still record with your head unit as normal so you have power data, which can be 1 second data collection.
Something like this, announced by Garmin, explained by 5krunner?
That’s a very interesting article, thanks!
I was not aware that the FIT file format itself had a limit of 5 Hz, though I highly suspect that that’s an artificial restriction to ensure a maximum file size dies not get exceeded. Obviously there’s many ways to record data faster than that. And the existing Garmin GPS chips are capable of 10Hz, so presumably phone GPS, used at much higher speeds, is as capable or better.
The current discussions on tire testing (either XC or gravel) always remind me of something. I happen to work somewhat frequently with aero testing, mostly for road/gravel (not an aero expert, mind). Once, during a wind tunnel session, one of the tunnel operators, a renowned guy that worked with big teams, said something in the lines of “if something is that hard to test (as in finding drag differences) in the wind tunnel, then it might not be worth to test at all”.
His point being, if some difference in equipment drag is so hard to measure in a controlled environment, then just go with the setup you’re most comfortable and confident with. This from a guy that made a living selling test sessions and consulting for teams.
If tire drag differences are so minute that finding them in testing conclusively is difficult, then just go with the ones one which you feel more confident and more comfortable
Totally agreed. And with things as complex as aero or tires, even if something is significantly faster in your test, doesn’t mean that it will be faster for you in a different situation or faster for me in any situation.
That said, I see claims all the time of “this testing shows that X is better than Y” and more often than not, the data doesn’t say either is better. So I wanted to point out what it takes to actually make that claim. And how Joe could improve his testing or maybe it’s not worth the effort if the results don’t say what he thought they said.
On a mtb if both are equally fast tires I’d rather have something I trust the grip with. For the front I’d always go with a slower tire I trust the grip with.
BRR Dubnital Race Rapid 2.4 results are available
Good results all around, but seems to be yet another severely undersized tire… Wonder if it grows significantly under pressure like some Schwalbe models
Looks Like a good alternative to the Racing Ray.
I can not honestly imagine having a different front tire than the Ray but the DUB may be similar but slightly faster.
I showed up to a very muddy slick race last weekend with the Ray and it did “well” enough to be competitive and was just as predictable as the dry. It does fantastic in the dry.
I guess, as a front tire I can put it on and forget it - just use it for whatever the conditions are and do my best and if the DUB can do that and be slightly: faster, wider, better puncture - its a great choice!
TL/DR - The Dub looks like a good mild upgrade to the Racing Ray with possibly similar “feel”.
The measurements are captured with the tire mounted on a 17.8mm rim. My pair of tires is ~62mm on i27 rims @22-24psi.
My comment was based on an apples to apples comparison with other tires also mounted on the same rim at BRR
When I first got mine out of the box I thought they looked small. Mounted them on 29mm internal rims and blew them up to 50psi and left them over night.
They measured 2.4 and when let down to 17psi retained that width.
So far very happy with them.
I echo that. Have yet to catch up to my PR on my local XC loop that I have set in 2020 on an enduro bike with Assegai front / DHR2 rear. I was much fitter back then and also had a lot more confidence. Now I‘m runnjng tires that are about 0.5kg lighter and I‘m sure they roll substantially faster.
Not saying this is applicable to everyone or any trail. I wish we had some mellow blue or green trails where I‘m optimizing for rolling resistance rather than finding a balance between grip, puncture protection and rr. I like fast tires but there’s more to it than just RR imho.
Would like to hear how it compares to other tires you’ve tried. Currently like the Ray as a front, but they never measure more than 58 on a 30mm rim. I think if they made a Ray that just measured 60, I would use it for everything. Been thinking should try a Barzos or fast track t7, but they’ll definitely be a good chunk slower.
Just had my first race and rides with the DUB’s. This was in garbage weather (wet mud and red clay) and the dubs were crap and compare to the ricks in these conditions. The rear had no grip on inclines and would just spin up. The front just tracked and didn’t give any confidence albeit I’m not sure many tires would in these type of conditions.
I did take them on a different trail system yesterday and they felt good on dry hard pack and rocks. I have the final race for this series in Tennessee this weekend so hopefully it’s dry.
I feel like the gps location trigger isn’t all that precise…although I don’t know for sure. It seemed like I was getting better consistency when I hit the button manually.
Is that coming or available? I honestly couldn’t find it on the page.
Joe