I think that it makes for better discussions and even better understanding that way.
You provided the property crimes and murders numbers as your metrics.
Metrics note. 13,716 sounds like all murders. I'd assume that only using murders that occur during property crimes would reduce the odds by half or more which would further your case.
Your statement "with someone just deciding to ignore facts and decide to murder someone" is only defensible if your decision is only based on a slightly flawed 1 out of 600 aggregate metric. You've basically stated the victim murdered the home intruder. Therefore, we need to use a more appropriate metric to evaluate the case for a murder.
Our only relevant metric now is the number of murders that occur in cases where three gun carrying home invaders are encountered inside of the house.
I'd say without question your 1 in 600 odds have shrunk down quite a bit. Victims in this situation are probably going to think that things are about 100 to 1 against them, but they don't understand the big picture statistics like you do. I'd also say that if two police officers just happened to have been in the victim's house that it's highly likely that they would not only have fired on on the intruders, but probably gotten commendations for it.
If the victim's statement that the intruders "...pulled a gun on him and told him to 'give it up,' " is true than your 1 in 600 is probably down into playing one number in roulette if not worse.
The only way you can stand by your murder charge in this case is if your core belief is that any potentially lethal self defense action during home invasions and I'd presume in most other defensive situations is murder. Or is it more like any lethal defensive gun use is murder?