I have been told that the best way to understand it is to look at it as two separate buckets. There is a bucket for in-game fatigue and a bucket for post game fatigue. For in game fatigue, if the pitcher started the game with 100 energy, 100 is 100 no matter if he has been at 100 energy the last four games or just barely made it to 100 because he pitched recently. However, once the game ends, if he had pitched his normal 100 pitches with an 80 stamina rating, and had just pitched recently, his post-game energy would be a lot lower than if he had pitched his normal 100 pitches and had been at 100 energy for a a few days before game day. This tells me that post game energy can get above 100. Combine this with other posts that have hypothesized that post game energy can be below 0, and all of the above makes sense. However, just because it makes sense, doesn't mean that it is true. None of this has been confirmed. It is just a theory. Nonetheless, I play by this theory now.
With the above stated, I think you need to be careful with the above strategy regarding pulling a starting pitcher early for a game 7. I am not certain, but I don't think in-game fatigue comes into play until the pitcher is below 0 energy. I can't confirm this, but I have seen too many times where a pitcher gives up no runs and once the game is over, he is at 1 energy. I would think that if fatigue worked progressively (i.e. if in game energy is at 50, the pitcher has X amount less effectiveness), then we would see more pitchers get pounded late in their appearance.
Another reason is that, on many separate occasions, I have seen a starting pitchers give up 3+ runs in the first or second inning, but then give up none between innings 3-7. Last thing you want to do is pull your starter after he gave up three runs early, only to have your bullpen guys give up more booty!