So if I understood you correctly, then most bots that are allowed to use are aimed only at a superficial result that will not be outstanding for the user. Then I believe that it is better for the player to work and polish his skills himself, without using bots. What's the point then in bots? Or are they just for the psychological satisfaction of a gambler?
The point that bots exist are: 1) if you play with built-in auto dice, there is not enough flexibility (settings) for the player to play, 2) browser based game cause memory leak over time (nature of it) and it can crash your game anytime after few hours of play, 3) not all gamblers know how to program/write code of the bot himself to have metioned feature, that's why public bots exists.
There are two types of bot: standard bot with feature such as multiply by x after each loss (Dicebot or Mydicebot has this type of calculation), custom bot in which multiply x1, x2, x3 or add y1, y2, y3 after each loss, or continously add one satoshi to the next betsize to suite targeted condition (loop of +1 satoshi until it fits). The later one is rare in which human can't do it, or built-in/well known bot can't or will not figure out how to do it. Given the following complex case:
- Given that for a set of 1000 test sessions with specific odd/ win chance over 100,000,000 rolls of each test session (100 billions bets made), the summarized statistic from the tests are "common - 99%", "rare - 0.75%", "extremely rare - 0.25%" losing streak, you want the bot do the following functions: adjust the betsize that "deposited amount" could sustain rare case and the "reserved amount (additional willing to deposit/lose)" could sustain extremely rare case, ajust the bet size that at the specified rare losing streak (let call it r1, r2, etc..) if you win the r1 streak the profit after that win can help you bet 3 more bets, r1+3 = r4, and so on if you win r4 streak the profit can help you bet 3 more bets r4 + 3 = r7. Or the total profit after winning of 100 common losing streaks can help to sustain up to r1+1 streak.
A question for you to test:
- Given a site at 1% edge, Player A total willing to lose budget is 1.0 BTC and he is betting on 0.5% win chance, scale his budget so that it can sustain 1800 bets.
- Given the same scenario, but for Player B he is richer and his willing to lose budget is 6.0 BTC playing on the same 0.5% win chance, scale his budget so that it can sustain 2400 bets.
I don't think this can be done using standard known calculation method, and certainly, rely on your skill alone can't help you acomplish above calculation or human can't do it. Finally to have the above complex case of betting strategy, you need something which come out from statistic (after numerous of time to do specific test), thus human can't collect that kind of statistic other than using a bot which can log the events.