Specifically in regards to training at altitude as means of overspeed, we must still consider the sprinters sport form at the time of the altitude session as I’m not sure that sprinting at altitude guarantees a supramax performance. If not, then the idea of peaking for a meet or session at altitude, in order to attain a supramax performance, seems counter intuitive.
If, on the other hand, we were to accept that sprinting at altitude ensures, with high probability, that a supramax performance is likely to happen, even if the sprinter is not in top form, then your idea would stand to reason IF one accepts that one or two single exposures to a supramax stimulus is sufficient to provide a strong enough residual to build upon with subsequent ‘conventional’ speed training, back home, that would serve to bring ‘unaided’ speed potential closer to the previous supramax performances at altitude.
I’m pretty possitive altitude results in faster times, but I agree with you that the circumstances surrounding a performance at altitude may not be conducive to produce that faster time.