I am fairly sure you do not mean 10 MHz of Doppler shift. 10 MHz of Doppler at 436 MHz is roughly a closing or opening relative velocity of 7,700,000 MPH (yeah, that is seven million plus).
OK, I realize it was typo or some such, just had to say.
One of the reasons that FM is more forgiving in this case is that if you are tuned 2 kHz off on FM you might loose a little clarity, maybe induce a little bit of noise, but still the signal is generally usable as long as the signal strength itself is adequate. However, if you tune off freq (or have a Doppler shift, it is the same affect) by 2 kHz with an SSB signal the audio frequencies are actually shifted up or down that 2 kHz, making the audio unusable.
For reference a Low Earth Orbit (LEO) satellite (most AMSATs are LEO) would have a maximum velocity just under 20,000 MPH, and the higher the altitude the lower the velocity will be. 20,000 MPH produces a maximum Doppler shift of 8.6 kHz at 145 MHz and 26 kHz at 436 MHz. The altitude of several popular LEO AMSATs require a velocity of under 8000 MPH, and so the maximum Doppler shift would be just over 3 kHz at 145 MHz, and about 10 kHz at 436 MHz. And that is the worst case Doppler, assuming the sat is making a pass headed directly towards you (thus the highest Doppler possible just as it comes over the horizon) and will pass directly overhead.
So, the rule is, the lower the altitude of the satellite the higher the potential Doppler shift will be, while very high orbits require less Doppler shift. For example Geosynchronous sats require no Doppler shift, because their relative velocity is very near zero to the Earth station.
T!