Imagination is running wild here.
Given the data so far, the only issue might be that the reed switch might not have sufficient time to close before it passes the magnet, unless the magnet is rather large. A 1-inch magnet is passed at 300 ft/s only in under 0.3 ms. That appears to be close to the minimum closure time of most relays:
It's also unclear how you intend to keep the switch oriented correctly. If the switch is rotated 180º, the swithc will not work. There is also a question of switch bounce as well.
Since we're talking design sheet concepts, I would recommend a Hall Effect switch as an alternate. They can be gotten in bipolar forms:
which means that rotation is less of an issue, but not eliminated, although one could imagine placing two of the oriented 180º apart, or two magnets can be used.
Their switching times are under a 1 microsecond, since they're ideal for motor indexing and speed measurement applications. There's no switch bounce. They are substantially more sensitive than a reed switch, as there is no physical mass to move with the magnetic field. They can be specified for -40ºC to +150ºC operation. They are also more g-shock insensitive. Package size is 2.75 mm x 2.9 mm x 1.1 mm.
So, I would vote for a HE switch over a reed switch.
TTFN