and if you think that 17cm taller height is enough for a worse CdA that could explain a rider loses 1min 46 seconds on the flat in a 17,8km ITT, what I could say! 
Also you said that clould be frame+wheels effect… even when I said before, that some people proved that the differences are really small.
But well, here you have the differences in a 3100seconds (51 min) test:
See, even if winner used the best frame and the best wheels and tha pro’s used the worst… differences are maybe 20 seconds? In any case 1 minute and 46 seconds 
But, why are you supponsing the pro’s are using standard set ups? 
Guys, I will stop here, you have to be kidding me, no doubt!
I will tell you easier, you only have to do that many maaany (pathetic)people are doing.
Please go to Zwift, try this:
180cm 80kg and push 300W, EXACTLY same bike, same FLAT route
then go and edit your profile to:
180cm 60kg and push 300W, EXACTLY same bike, same FLAT route
and…
YOU WILL START TO FLY
and, of course, you don’t have to come back here and say to me that I was right.
First you are looking at frame only data, for road bikes not TT.
Wheels make as big or bigger difference than frames
On one lap it’s roughly a minute difference between the best and worst frames. Factor in wheel difference and it could be even larger
It is more that Cd that play a roll on the flat.
Cd - Aerodynamic Drag - Bigger rider has more frontal aria.
Cr - power to overcome rolling resistance goes up as your weight go up
So what are you saying,
is the lighter rider faster or slower?
I’m quite certain the physics model uses power against aero and rolling drag on the flat. Aero modeling uses weight and height to calculate the CdA and frontal area, rolling drag uses a Cr and weight - all these are assumptions, I don’t have any visibility into the model, but it’s consistent with the tests done on ZwiftInsider. So higher weight would penalize you on the flat at a given power. Of course, if you look at equal W/kg, that more than offsets the drag penalty of a higher weight, making heavier riders faster on the flat at equal W/kg. The real question is whether the penalties of higher weight are realistic; three of the relationships used (CdA vs weight, CdA vs height, frontal area vs weight and height) are possibly arbitrary, it’s pretty hard to get a cohort of riders where these are the only variables, and human bodies come in all sorts of shapes that don’t lend too well to simple modeling.
Quite evident @Gerrie_Delport 
If you are 180cm and push 300W with 80kg on the flat in Zwift you will go at 39km/h
But if you change the weight to 60kg, then you will go 42km/h, that’s not realistic in any way…
So, that’s why people put in their profile unreal weights (25…40kg) to fly in every terrain, even on the flat with 160W…
Thanks @S.C
What is the source of your data?
Graphics are taken from the tests of Zwiftinsider people: Speed Tests: Steady w/kg Across Various Body Weights | Zwift Insider
But I also tryed to change my own weight and ride same power, same bike and wheels and same road…and it’s like that.
That’s why I told you everybody that can do your own test of this FACT. 
The thing is you look at w/kg you need to look at power.
So why do we see this on a flat race? TTT with normal bikes on Greatest London flat.
112kg had to do 306w but 80kg only 243w
Your assumption was that the Power (w) remain the same, but in fact the power does not, as shown below.
80kg = 41km/h @4w/kg - 327w
60kg = 38. km/h @4w/kg - 236w
So as your weight goes down so does the power requirement to maintain the same speed.
That is because rolling resistance does have a mass (weight) factor, heavy riders have more rolling resistance than light riders, as per the article I posted above. Also heavier riders have larger frontal aria than smaller riders because they are generally bigger.
Races? but what about drafting?
Better if we consider solo rides or ITT as an example of race…
Susun is right. Reducing your weight will make you faster in a flat TT. It’s because lighter people have a lower drag coefficient. same 300W, same bike, just change the weight from 80 to 60 and you go faster in Zwift and in the real world (it’s a lot harder to reduce your weight in the real world though). It’s why, in the real world, you draft off the fat guy and not the skinny guy. The fat guy is punching a bigger hole in the air.
Gerrie’s link to Bicycle Speed (Velocity) And Power Calculator shows it too. Adjust only the rider weight and the CdA reduces which means faster speed.
Now, as for how accurate Zwift’s calculation of drag is based on weight is a matter open for discussion. Personally I think Zwift has tweaked their calculations a bit giving lighter riders far less drag to help them out on flat ground otherwise it would render the whole w/kg race category system useless.
I never disagree with that, it work the same IRL less weight and same power = more speed. That is why F1 cars use carbon fiber to save weight. Weight also impact acceleration.
I can’t comment on Zwift’s calculations, I would love to see the formula. But I don’t think they tweak it to give any type of rider an advantage.
Man, you have to be kidding me, aren’t you?
Oh my God!
Really? Do you think that aerodynamics if the rider is heavier, and rolling resistance can explain that?
At 300W, if your weight is 80 instead of 60kg, your CdA and drag increases like if you were a TRUCK?
Is like that?
Nobody who now something about cycling in the real life would say that if you push 300W and your weight is 60 you will go at 42km/h but if you are 20kg heavier you will go only 39km/h…
You only have to use the Bicycle Speed (Velocity) And Power Calculator that YOU also used early in this topic, please put there 300kg, same height and switch 60 or 80kg, press “calculate”.
As you know, Bicycle Speed (Velocity) And Power Calculator is using changes in CdA and rolling resistance (Cr). They assume heavier has worst CdA even when that can’t be true always, but well, no problem.
And nothing considered there about the inertia forces, kinetic energy helping to maintain speed that favors the heavier as everybody knows.
Well, even like that… speeds are similar, not 3km/h more for the lighter rider…
I just used the data set you provided. That is what I get with your data.
True, but is really slightly faster, not 3km/h!! 
That is…
Could be a mistake, or intentionally to help lighter riders… I don’t know.
But the fact results in a not realistic simulation that causes big LOL things, like a skinny guy moving 300W could beat a pro ITT specialist (even the best) if they use Zwift
You are still stuck on a fixed wattage (average), which is not what happens in reality. Variations in the wattage over the TT CAN make a difference in overall time, even if the 2 riders have the same average over that time.
I ride TT/Tri bikes in Zwift and also IRL so I might know a little about this.
And, please remember to be civil on the forums.
So then, another thing to fix…
What difference? maybe small, but not until to compensate such big difference.
And, in th other way, we are talking about “instant” speed now. Not about average in a TT.
If you do the test I suggest (weight down 20kg) your speed will be bigger at same power, that has nothing to be with variations/average…power.
