Based on the bet history, most bets that are higher than 100 bits were lost bets. I did not check all pages of the bets history but I can see some 10,000 bits losing bets. Shouldn't you make some profit from those 10,000 bits losing bets? That's even 3x bigger than your bankroll.
If anything, the majority of those bits would be going to the
bankroll, not his pocket - Only house edge profits (they are usually subtracted from the total bet whether it's a win or a loss) can be pocketed by the house.
So that would make to be several hundred bits at the highest.
In this case, the bankroll is effectively his pocket. While his expected profit is the house edge multiplied by the bet, his actual profit depends on how the bets end up going. If he did profit 10,000 bits from a losing bet, there's nothing stopping him from taking them as profit. Expected profit is simply the expected value of each bet and doesn't represent exactly what will happen as users play bets.
I noticed you posted a 50 bits promo code here just now but removed it. I'm the one that used it (I think they're still one-time use?).
I wanted to see how your bankroll would withstand a few bets so I was gambling 10 bits at a time for most hands.
As I expected, I eventually lost all that bonus money after going as high as 110 bits at one point. So it at least makes me feel better that you have some kind of house edge.
Losing all of your balance doesn't mean that there's a house edge. It could very well be a result of variance, especially if the house edge is close to 0% (either positive/negative). You also had a very very small sample size.
Mmm, the cards are picked randomly, any ideas of how demonstrate that the house does not have any edge?
I want to increase the bankroll and the max bet but I'm a little bit afraid
I can't seem to find the rules used by your game of blackjack. If you know what they are, you can plug all of the values into
a blackjack house edge calculator to figure out what kind of edge you have. If the set of rules is unique, your best bet might be to run a very large number of simulations to determine what your house edge is approximately.