HTupolev

If you want a specific study that gets into bicycle performance, "Validation of a Mathematical Model for Road Cycling Power" (Martin, 1998) is a good place to start. This paper basically assembles a theoretical model for performance, and does some on-bicycle power tests, showing that the model predicts the measured power fairly well if reasonable parameters are provided for the bike+rider system.

If I simplify the model a little (combining static and rotating drag coefficients and assuming that the wind velocity is 0), it assumes that force from aerodynamic drag can be modeled as:
.5 * (air density) * CdA * (v^2)
CdA is the aerodynamic drag coefficient of the bike+rider multiplied by the frontal area of the bike+rider. It's basically a number describing the dragginess of the aerodynamic profile. The air density at sea level tends to be around 1.225 kilograms per cubic meter.
To get the power required to overcome that force, we can just multiply the formula by velocity, so:
.5 * (air density) * CdA * (v^3)
As an example, a typical road cyclist might have a CdA somewhere in the ballpark of .32. If they're riding along at 20mph (8.94m/s) on level ground in calm conditions on a mellow day at sea level, we can predict their power loss due to air drag as:
.5 * 1.225 * .32 * (8.94^3) = 140 watts

Force from rolling resistance is assumed to be constant with respect to velocity, but linearly proportional to the load on the tires. So:
9.8 * (mass of the bike+rider system) * Crr
Crr in this case is the "coefficient of rolling resistance", which depends on the tire setup (such as what tires and how they're inflated, although things like rim width and choice of inner tube also play a role). Similar to before, we can multiply by velocity to get the power required to overcome the force:
9.8 * (mass of the bike+rider system) * Crr * v
So for example, if we have an 80kg bike+rider going 20mph (8.94m/s) on tires with a Crr of .004, we get:
9.8 * 80 * .004 * 8.94 = 28 watts

Rolling resistance is a bit more squirrely than air drag. Notably, how many physical phenomena are actually reasonable to encompass under "crr" and the degree to which they're invariant with respect to speed is a little bit fuzzy. But as the paper demonstrates, it's still a *good enough* approximation to do useful things with the model.

Crr is traditionally measured by loading a tire against a surface (usually a steel drum), and observing how much power is required to rotate the tire against that surface at a given speed. The Crr can be inferred from this. Drums aren't roads (and so Crr needs to be adjusted for the shape of the road surface to get precise results), and this also doesn't account for effects relating to the tire's performance as suspension on a given surface, but it's still a useful and repeatedly start for predicting which tires are going to perform better than others, and by approximately how much.

If we go by data measured from bicyclerollingresistance, 25mm Gatorskins have a Crr of .00659 at 80PSI, and 25mm GP5000s have a Crr of .00363 at 80PSI.
Let's go back to the examples earlier, with the 80kg cyclist with a CdA of .32 doing 20mph at 168W. If we swap his tires out to those GP5000s, that total drops to 165W. If we then swap in the Crr for the Gatorskins, how fast will he be going at 165W? It turns out to be about 19.06mph, so nearly 1mph slower.

Now, this is admittedly a very approximate example. There are a few factors I didn't account for, and obviously the Crr values are an extremely rough guess. Consider, though, that we've got two posts in this thread that corroborate that the difference is of this magnitude if not larger.

Anyway, if we assume a difference of around 5% for a 20mph roadie, and compare this to rubik's example of a century ride? That's a 5-hour ride, so 5% would add 15 minutes, which is usually plenty of time to fix a flat.

I don't think that oleg's point about the repair kit makes a very bit dent on this judgement. Partly because I haven't known Gatorskins to be anywhere near tough enough to justify not bringing a repair kit in a situation where you would for a racing tire, and partly because repair kits just aren't that big or heavy. My main pump is 88g; add a tube and tire levers (and perhaps a Park GP-2 for good measure) and I can still be under 200g. Which is peanuts. Like, it takes about 40 seconds for a mere 150W power source to lift a 200g mass by ten thousand feet.