That is why theoretically, you should be able to land safely on an horizontal surface, and without any transversal speed, when jumping from an height about one time and a half what you can jump on your bare feet (also depends on frame geometry).
Explanations to that....
The damping structure
If we consider your centre of gravity (a virtual point G on which would be concentrated all your mass), it combines with the bicycle frame into the knee-like mechanism defined by the levers G-to-C and C-to-H; where C, the crank axis, acts as an articulation.
Just before impact with the ground, you should be fully extended, with G laying forward, thus opening the G-C-H structure to create maximum shock absorption potential. When the back wheel hits the ground, the impact is absorbed by shifting your weight backwards, controlling the rate of this transition with your arms and legs. That move brings G down and backwards, folding the overall structure and progressively absorbing the shock.
The curves I, II and III indicate the paths of G relatively to the crank axis C and to the hub axis H, for different combinations of the arms and legs actions.
G is taken as the initial position of your centre of gravity corresponding to a maximum extension on the bike before landing. Curve I illustrates a weight shifting backwards, keeping your legs straight. No Good! All the damping action is concentrated in your arms, you'll tend to be ejected backwards, 'cos you won't be able to hold the bars (too much load).
For curve II, there is no weight shifting backwards, only your knees and legs will absorb the shock, the G-C-H structure
didn't fold. No Good! You could virtually land with the hands off the bars, but you'd need to absorb the impact as if you
were jumping on your bare feet (obviously not the best option, as the bike is still in your way and may upset some precious
Now, the green curve (III) in the middle illustrates the ideal combination
of your legs and arms' efforts to fold the G-C-H structure. In curve III, G ends up a bit further from C than it actually
passes by it. That is because the weight shifting first starts with the knees folding under the legs' control,
before the arms take the relay by a controlled extension. It ends up just above H, which means that you are
perfectly balanced on the back wheel after landing.
A straigth line between the points G and III would be in the case of an equal load control in the arms and legs, which isn't ideal, because the legs are much stronger and should bear most of the shock absorption. Above the ideal curve, there is more than necessary load in the arms, whereas below this curve, the impact is mainly absorbed in the legs.
The angle alpha (H-C to horizontal) should be between 10° and 30°, leaving you with some margin for the balance, and avoiding to bang the front wheel on the ground.
If alpha= 0°, then you'll land flat. You can still do the correct move relatively to the bike, but you'll take some of the impact directly into the wrists (avoid that).
Also avoid alpha < 0°, landing on the front wheel is hardly pleasant.
By blocking the back wheel, you'll get a longer lever on the rear, tyre touch-point to crank axis, instead of the H-to-C lever, which means even more shock absorption. The tyres add some damping, but it's no big deal, 2" travel at most.