I did not mean to offend you personally, if I did please accept my excuses.
Suppose that just-dice had not 1% of edge but 99%, so that if we toss a coin there is 0.05% of winning two times the bet. In that case players have a bad life and we agree that investing/divesting (daytrading) at any time (even knowing the bankroll of the players) does not make sense, we would be investors of a quite greedy casino' indeed. Do you agree? If you agree suppose that the edge is 98%, has something changed? Will you make daytrading? No. If it is 97%? 96%? ..20%? Is there an edge where you start to "daytrade"?
I see your point, that's a good argument. What I would say is that your argument can be made even simpler looking at %-win instead of the question of edge: if Nakowa was spamming 98% bets, betting 40,000BTC per bet to win ~200btc per bet, you'd have to be invested all the time because you don't know when that 1 in 50 huge loss is coming. Same goes if he's betting 2btc over and over at 1%-- You never know when that big loss is coming. But my argument before holds for 50% bets with a small edge because his bankroll follows a sort of random walk and you have enough time to invest divest as he is reaching one end or another of his Goal <---> Last Deposit spectrum.
Again, I'm not advocating this, I'm just saying it's a possibility that the day traders under those conditions
could make EV+ decisions. But only under those conditions. For example, if you don't know how much nakowa has deposited, you don't know where his lowerbound is, and by divesting you could lose out on a huge loss by the player.