Gross simplification of a complex issue
Being able to control the dice after hitting the back wall is nothing short of complete bullshit.
No, "complete bullshit" is not a scientific answer.
Throw a die with exactly the same trajectory and velocity each time and the outcome is completely deterministic. Such consistent delivery is not required to turn the odds decisively in your favour. Greg Limes published an analysis of just how little influence is required for an advantage, which I archived in the Gambler's Research & Development Newsletter below.
Many individuals have claimed to possess the ability to influence the dice in this manner over a large number of trials to a statistically significant extent-though actually that is not required. Provided you can throw the dice within certain paramaters associated with the initial release, then, regardless of your actual results, you have a positive expectation.
The controversial point is whether human ability can influence the dice to the extent that the house edge can be overcome. A strong body of anecdotal evidence exists to support this position-Marvin Karlins, John Scarne, Darwin Ortiz and Edward Thorp all testify to the existence of dice controllers. Individuals such as Steve Forte have demonstrated the skill publicly.
Two things may settle the issue publicly in the near future: a Canadian university has developed a dice-throwing machine and is running tests to assign precise values to acceptable error terms associated with die-throwing in order that a positive expectation may be achieved. Further, "Sharpshooter", author of the forthcoming "Get The Edge At Craps", one of the individuals who claims to be able to control the dice, claims to have footage of thousands of controlled throws. Presumably, you would accept this is as conclusive evidence of the efficacy of the method? If so, I could try to arrange the video footage to be made available online. If not, then I'd question whether your comments on this matter is personal or scientific.