I've never thought about it.
Quick back of the envelope calculation (actually, I'm not even using an envelope...
): rear F1 tyres have a 67 cm diameter. So,
for each cm of wear you will get a delta V of 100*(67/66 - 1) percent, or 1.5 percent decrease in speed. If top speed is, let's say 320 kph,
that is 5 kph. That would be true no matter the gears, the mass or whatever, so I disagree (slightly) with RH and Jersey Tom.
Certainly it would be noticeable, because that's comparable with the difference in top speeds among teams (perhaps half the difference).
Maybe, perhaps, it's possible, you could theorize, there is the chance
that the thinner tyre would stretch a little more (I imagine it would be
very little, the walls wouldn't change much, unless really battered on kerbs), compensating something of the estimation I made, but I don't think THAT would be noticeable.
Am I wrong? I was too lazy to use a calculator or "draw the problem", as I usually do, so I could have made a mistake. Thanks if someone have the time to check it.
Warning, crazy Ciro idea: if the difference is noticeable (and that's a big if, even if the quick estimation is true, I don't know if the tyres wear 1 full cm), the regulations allow you to have a bead with a 4 mm tolerance ("between 328 and 332 mm"). So,
Can any of you imagine a design for a wheel whose bead diameter change in that amount to compensate for the wear while racing, thus keeping some of the top speed? Yes, yes, I know it sounds like Rube Goldberg, the "magical expanding wheel" and I don't even know if regulations allow for it... besides I guess it would be forbidden in ten minutes. There is no practical application for that and if any team "develops" that kind of wheel the others would follow suit in a minute, if it's not forbidden immediately.
Anyway, teams invest a lot more in aero to get smaller differences in top speed, so...