I've been having a long and constant problem with using the APT software, and that is because the bots are just not realistic. The thing that bothers me the most by far is how often they donk bet and how it is almost always for value. The OOP flop checking range for the bots even in the KGB dungeon is about 90% weak. I'm super surprised that this has not been brought up in the past because this is a tremendous flaw in the bot coding. Donking is not a common play in real life, live or online, but I see it on a constant basis here with the APT bots. If you are skeptical about this, then I will be more than happy to show video evidence of this occurring. I don't have any right now, but it will be easy to reproduce this issue.
It looks like you're new here. If you want to get involved, click one of these buttons!
Comments
The bots at the KGB level use AI and play different styles depending on what works for them against your game/stats. If they are doing a bunch of donk betting, it probably is working for them and was the strategy they are choosing to combat some pattern they detected in your play. They should adapt out of it if you catch on. They definitely are not perfect, but of the long-term weaknesses in their play that I know about - things that are tough to fix without messing something else up - this is not one of them. They tend to play styles that might be considered "bad" when it works, which is really the point of the site. Good players tend to mix it up and try to confuse their opponents. If they played a consistent "standard" style, they would be easy to beat. But overall, the bots crush our members at the KGB level. Part of getting better is learning to adjust to styles that you may not like playing against. If I could win by donk betting, I'd do it also.
I looked at both of your accounts and it appears that I'm correct about the bot strategy working. Both of you are getting beaten in most sessions and overall by the bots at the KGB level. I'd recommend trying to punish them for the strategy - they will have to change.
Do the bots at the hardest level use you session statistics, your lifetime statistics, or some mix in between these two?
Am I correct in assuming that the bots balance their donk betting ranges instead of just betting value?
In my live tournaments when I see a donk bet I assume it's for value by a weak player. I don't know that to be a fact, because I never see a showdown because everyone folds to the few donk bets we see.
Hmm... maybe I should throw a donk bet or two into every tournament!