Well I don't even know if this should be called 'drift' or 'deflection'.
In my mind, aerodynamic deviations(active) to a bullet's fall should be referred to as deflection. Those deviations that are not taking energy from the bullet's condition(passive) should be referred to as drift. That's just my thinking.
There was an incredible discussion on wind deviations at BRC that went roundy roundy for a long time, ultimately with zero conclusions about it.
One camp thinks a bullet points into wind and gets sucked sideways like a wing. These folks think bullets fly..
Another camp thinks bullets simply point toward a new path of least resistance. These folks think bullets fall.
I think it may be both, in that a bullet will point forward even with a drift as this takes least energy(for a gyroscope), and point forward into relative vector amounts to least resistance aerodynamically. But I believe the actual deviation is still passive, for falling bullets.
Fire control RADAR could show it.
The formula for wind deviation is based wind component applied to time lag:
D=W(T-R/V) where
D = bullet deflection in feet,
W = crosswind in fps,
T= time of flight in seconds,
R/V = range in feet divided by muzzle velocity in fps.
This means no lag = no deviation. So,, if you drop a bullet straight down from 100yds up, into a 10mph crosswind, does it deviate in the wind direction? There would be no time lag as the bullet is accelerating the whole way. Would the BC of that object(hi or lo) even matter given any acceleration(like a rocket)?