Sorry, can't follow what you are saying John.
Begging the question is phrasing your premise and conclusion in such a way that they circularly support each other regardless of the validity of either.
For example, you eliminate anyone involved in a fight although being involved in a fight doesn't mean that a person can't legitimately claim self-defense. In fact, the first TX CHL shooting could accurately be described as a roadrage fistfight that ended in a fatal shooting. Yet the shooter was acquitted and rightfully so.
Had he been convicted, your screening process would have eliminated the case and therefore you would not have examined it to see if expert witness testimony might have made a difference in his case.
By also trying to eliminate LEO shootings from consideration, you further reduce the pool of shooting trials to draw on and the result is that your premise follows--there is no need (or only very rarely a need) for expert witnesses in self-defense shootings.
This is very similar to the technique used to paint gun ownership in a bad light by saying that a gun in the house is more likely to kill an occupant than to kill a criminal. Focusing only on situations where criminals are killed rules out all the cases in which a criminal is only frightened away or injured. The result is that the usefulness of guns in self-defense can be made to look quite minimal. The fact is that 80% (or more) of the time self-defense doesn't involve firing the gun and about 80% of the time when someone is shot with a handgun they survive. Which means that by virtue of how the question is couched, perhaps 95% (or more) of self-defense gun uses are eliminated from consideration and the premise (guns don't stop crime, they're a danger to the owners) seems to follow as a consequence.