Isn't there a DRS rule about how far down the pitch you are?
The 2.5m rule that's been quoted only relates to the limitations of the technology. Basically, the software doesn't have enough data to calculate a projected trajectory based on the limited amount of data. So if you're able to get that far down the wicket (it's possible) and you're given out, the technology can't reverse the decision because it hasn't got enough data to predict the path of the ball.
The key section of the rules to consider with advancing down the wicket is:
...it is to be assumed that the path of the ball before interception would have continued after interception, irrespective of whether the ball might have pitched subsequently or not.
So basically, if it pitches in line with the stumps, and it's going on to hit them, you're out. Irrespective of where you are the wicket. It's VERY difficult for the human eye to account for the discrepancies of a player being in an abnormal batting position* in such a short amount of time it takes a player to advance down the crease. Traditionally, the benefit of the doubt lies with the batsman as it's almost impossible to judge the "height" of where the delivery would end up based on the player being down the wicket, regardless of if it pitched in line or not. That's why they're not usually given out.
Not enough information for a computer program and not enough for a human.
...Now, since we have a "perfect" system that's error-free by design (and errors are subsequently programmed in to create the illusion of error) then those 50/50 decisions in real life, will be 100% absolutes in a computer game. I would caution BigAnt to dial back on that too much, because then you're opening up the game to even weirder LBW decisions.
Does all that make sense?
LBW is such a curious law and it's always open to personal interpretation. I'm always over-cautious when it comes to it and when you bring it down to the rudimentary numbers and law of averages when it comes to the actual AMOUNT of stump vs. the size of the player vs. the odds on hitting them, most LBW's are not out, by default.
What DRS is does is backs the umpire as the "benefit of the doubt" rather than any individual player and it says if the umpire thinks it was happening, it is happening, regardless of the percent of the stump the ball may or may not hit based on the technology. You'll see many decisions being called "good" when the ball is only hitting a half of the leg stump and it's not given out, for example, then those same decisions being called "bad" when they ARE given out. It comes down to three things:
The Player
The Match Situation
The
Other decisions during a game.
If it's a good player on a roll and he's batting well, the umpires will be less-inclined to give a 50/50 LBW decision.
If the match situation is precarious and it's going down to the wire, those LBW decisions are given more weight and the umpires will be less inclined to give those 50/50 decisions. The opposite applies to games where they're done and dusted and it's a shitty player batting. If there's been a tonne of LBW's during any given game, they become less and less likely as the match goes on. Law of averages applies there. There's always exceptions to this of course, but as a general rule, you're taught to be aware of all these elements when umpiring.
*Switch hits is a huge issue for LBW's because the batsman will often take a completely different stance with his back leg becoming his front one, etc. So the rules applied change in milliseconds for the umpire and it's almost-impossible to apply the normal laws in that instance, which is why if you're hit on the pads doing a switch hit or similar, you're going to be out more often than not.
EDIT:
All this is of course my interpretations of what is taught during umpire education courses.
