This needs to be quoted again.Why is it that people think they can forego the ten years of math and science courses and just skip right to theorizing about the universe?
If you don't understand exactly what it is that is know about the behavior of the universe today, you can't make any guesses about the causes of that behavior.
Yes, as an example of the type of unhelpful, contemptuous and completely inappropriate response that is unfortunately far too common here when someone expresses a view (or, in this case, makes a suggestion) that's scientifically naive (or, worse, suspected of being woo).
What's particularly silly is that it completely misses the point. No maths (and scarcely any physics) is needed to see that martu's model is a total non-starter – just a modicum of everyday observation. It fails to account for gravitational phenomena in a number of crucial and obvious ways – some of which have been pointed out.
martu, you have just considered one gravitational effect – that an object decelerates when moving away from a mass (i.e. appears to be 'attracted' to the mass). Your model is that the mass has displaced the gravitational particles, which have therefore become crowded in the vicinity of the mass (presumably this crowding thins out further away from the mass). These particles are supposed to obstruct the motion of an object moving through the crowd. There are serious problems here – in particular (as Reality Check pointed out), your model has the gravitational force depending on the object's size, not its mass. Also, you offer no explanation why the gravitational particles should 'spring back' rather than continuing to move away.
Even worse, your model fails to account for the fact that objects don't need to be moving to feel the gravitational force, and is completely wrong about the gravitational influence on an object moving towards the mass. Why doesn't it decelerate as it meets more densely crowded particles? Why does the effect of the obstructing particles work differently on objects moving towards the mass than on objects moving away from it? Your model ignores the directionality of gravity – rather a serious omission!
Other objections have been mentioned - the number and complexity of the assumptions required, and the implication of a granular, fixed space.
The lack of mathematical rigour at the concept stage isn't necessarily a problem. If it seemed at all a viable model (it isn't!) then you would need to develop a mathematical model that explains, for example, the observed distribution of gravity. Creating a descriptive model first and then developing the maths (or getting someone else to develop it) is a perfectly valid way of doing theoretical physics.
Loss Leader, you did have a point. It's mindbogglingly unlikely that a non-physicist could develop a useful descriptive model of gravity that generations of theoretical physicists have missed (and it does rather trivialise physics to suggest it). You could have said this in a friendly way (having pointed out some of the logic and physics errors - assuming you're capable of doing so).