I’ve just tried the latest version of the android app and it all went really well apart from the speed from power.
At 103w my speed was averaging 19.1mph.
My average speed for the entire workout was 21.5mph at an average watts of 141w. My normal average speed outside at 141w is generally about 15mph.
The app is version 3.4.5.0 (171).
Does anyone know if I can adjust the speed from power in anyway?
Thanks
I’ve given up on how speed/distance gets calculated for indoor rides. 
The only way it could work using the standard formula would be to factor in your weight (including bike & gear) plus any slope changes if riding a sim.
Then you would get a proximate speed to riding outside on a calm day (no wind).
I haven’t tried experimenting with changing body weight in Xert to see if that affects the calc. It should based on any change to W/kg.
Perhaps they could add a setting under ‘Speed from Power’ so you can tweak the calc for those who’d like more realistic speed/distances recorded for their indoor workouts.
As it is now it definitely over estimates.
Thanks Riderider2, its definitely way out compared to my normal speeds. I’ll try changing my weight next time. 
Calculation is usually done on the basis of a flat and windless course, rider in drops position (no aero bar). Rider weight has little influence in that scenario, speed mainly depending on the Watts you produce.
Still, 140 Watts should not give you much more than ~18 mph, give or take an ear - comparing it to outdoors is only meaningful if the ‘indoor’ conditions are met.
Definitely out by quite a bit. I mean 19.1mph at 103w?
That’s off by ~3 mph, as is the overall. I was stunned myself when I found that at 70 Watts, you are capable of moving at ~13.5 mph - I thought you wouldn’t move faster than 5 mph at that power 
What happens if you fill in your weight and bike (with gear) at this site then enter watts?
https://www.gribble.org/cycling/power_v_speed.html
Perhaps the Xert calc is not that far off and you’d get a similar figure outdoors if you were able to ride a completely flat windless circuit (or velodrome) on the drops with your stem slammed.
But as soon as you factor in outdoor conditions with stops, slope, wind, and your aerodynamic profile the same effort would drop your average speed to what we are used to?
Aero bar makes ~1.5 mph difference, so if they use ‘best case scenario’, we’re closing in 
Have a look at that calculator and play with weight - as I said, it doesn’t make as much of a difference (about .1 mph per 10 lbs or something) as does your (aero) position…
Not at this time. I can see if we could consider adding a bias setting that would allow you to scale the speed from power up/down.
It’s likely not a pressing change for us, since most of the analysis in our system focuses on the Power-Time relationship, while distance (in our system) is not overly relevant. However, I know many people like to track their annual distance, distance between tuneups, etc. so it’s good to have a reasonable estimate from indoor rides.
And as we have established that the calculation is - best case scenario - not that far off, somewhere between 1.5 and 2 mph maybe, for a one hour session, it doesn’t really add up…
And as I’ve mentioned, it’s kinda difficult to compare in- to outdoor…
Enter some remark about how our estimations are performed assuming you’ve freshly shaven your legs and are riding the lightest, most aero, newest road bike 
Forgot about the leg shaving - that’s also a few seconds on a 40 km ITT
Sorry but it’s not difficult at all for me to compare, when the app has me going at 19.1mph at 103 watts. I’d have to weigh about 40kg to manage that 
Depending on the variables used and your being able to accurately enter them, it’s calculating an approximation of what you could achieve outdoor.
For an indoor ride, all these variables do not count, so the average may be off, compared to what you ‘normally’ achieve outside, but it’s not that significant.
Have a look at another tool: http://bikecalculator.com/
At 40 kg, producing 103 Watts, you’ll be moving at 18.49 mph and at 80 kg you’ll be moving at 17.32 mph.
Your weight is not a big factor, if you do not double it at least
Drag, drive train loss and other stuff is, which makes the earlier mentioned calculator a bit harder to work with.
You can turn it around and think your position on your bike is not optimal, when cycling outdoors. That may slow you down, so you enter parameters that are not correct and can’t match the predicted speed.
All fine in theory but unfortunately nothing can be changed within the xert android app to reduce the speed it’s coming up with.
I posted the other calculator link because it shows the math. Not that I fully understand it.
If I use 150 lbs with 20 lb bike/gear both calculators indicate 103 watts should be in the 16 mph range, not 19.
I think what bothers most of us is we end a smart trainer workout and think “I couldn’t possibly have gone that fast outdoors with the same effort”.
There is a big difference between averaging 15.x mph versus 18.x on your favorite loop. The higher the numbers, the bigger the difference in relative effort.
As @ManofSteele mentions Xert could add a setting under ‘Speed from Power’ that allows adjusting the calc similar to the freshness slider. IE, skew the calc to reflect what you feel would be your “normal” outdoor speed for same effort.
With speed overestimated trying to establish equivalent miles per week stats becomes nominal. Of course Xert’s stance is that is another useless figure to track. It’s all about XSS and focus minutes which don’t necessarily relate to distance or average speed so why bother? 
However, many of us are accustomed to “miles per week” as a gauge or “average mph” on a ride to judge relative effort and fitness level. I guess both are basically old school thinking and irrelevant as we spend more and more time training with power indoors.
So the moral of this post is just keep the needle pointing up and don’t worry about it.
Or Xert should just dial it down as I don’t hear any complaints about speed being underestimated.
Nix the best case scenario (lightest bike, aero bars, tuck position, no slope or wind) and give us a “normalized” road estimate of speed.
Nope and Scott mentioned it has no priority either.
In any case, I wouldn’t worry about it too much, but that’s probably just me
All good arguments and as I said, it’s not that I do not agree with speed apparently being off.
In my view, it’s not by that much if you average scenarios and not really relevant, as the only number I look at is power.
I often ride a virtual climb, grinding at 15 kph and averaging 250 Watts.
But the app controlling my trainer says my speed was 38+ kph.
Targets like miles per week are very valid, but if you mix in- and outdoors, you will most likely not overshoot that by a lot of miles because the (Xert) app calculates your speed a little optimistic
Its probably just me being overly picky and not wanting my friends to see my virtual mph on strava being so far out of wack
I was thinking about this yesterday and I think I know what the cause is (related to some other testing we’ve been working on). I think we’ll have a fix for this, giving more realistic values for Speed from Power. I’m going to test today and we’ll add it into the production version of v3.4.5.
