The thing is, trading can't be the only successful way to make money, or even more successful than other traders on average -- because trading implies buying more as the market confirms your views, and if that's profitable, it implies that traders will own all the world's assets, and every asset will either rise infinitely or collapse to zero.
If it helps, you could think of him as an arbitrageur betting that the mortgage market would converge with reality. That thesis relies on believing that the market misprices assets, so it seems that further mispricing would confirm it rather than belie it.
If you trade on value, you shouldn't really care what the market is doing. If risk is mispriced, buy more, even if the price is going down and you are losing money in the short term.
I find it more intellectually appealing to make my money taking a stand than to make it by saying "Me too, me too!"
Most successful traders emphasize that adding to a position that is showing you a loss, because you believe you are right and the market is wrong, is a losing strategy in the long run. That suggests that Paulson won with a losing strategy, which suggests that timr has a point.
Perhaps you are correct that there is something other than "trading" in which people routinely add to losing positions and end up making money when the market changes to confirm their judgments. It would certainly be something other than winning trading.