Could human beat GTO bots in Fast Fold poker?
Fast Fold poker obviously plays different in long run strategy. You can select the hands you want to play and adapt this
No one lives forever, so we can't play infinite number of hands. And since it's impossible to determine this number precisely, it's kind of not fair to glorify the bot as UNBEATABLE.
I usually think this when I watch the boxing. ‘And introducing…. theeeeeeeeee UNNNNNDISPUTED… ‘
Well.. he hasn’t been disputed by me yet. Maybe I can kick his ass? Nobody even asked.
So far, we've proven here that GTO bot is less unbeatable in Fast Fold 😀
Guys, come on, there's no such thing as "less unbeatable", it was a joke. Something can be only beatable or unbeatable. All I meant was to show where this thread has led to - to a strange conclusion.
But beating a perfect GTO Bot with the help of variance is really uninteresting. You could beat the perfect GTO Bot over trillions of hands with nothing but luck. Variance never goes away completely.
The only interesting thing is EV. Show us a way to gain EV on the perfect GTO Bot and people will get excited.
The only interesting thing is EV. Show us a way to gain EV on the perfect GTO Bot and people will get excited.
The only thing that comes to my mind is that combining categorized statistics could help. But nobody can do this perfectly.
There are countless number of categories you can make stats from, for example:
- completing a flush draw on the river (%)
- ace appearing on the river (%)
- 2, 3 or 4 appearing on the river (%)
- river pairing the board (%)
- opponent completing a straight on the river (%)
You can never predict what card will hit the river, but if you had all these 5 mentioned statistics from the last 25 or 50 or even 100 hands you could then choose the most appropriate ones for particular situation and minimize the risk before the river card appears. You'd have to adapt the selection and combine different categories with each other and this wouldn't be easy. Maybe if you played very fast against the GTO bot you could then count this more precisely. But still it would be super difficult to execute multiple times.
I think when you're in the right state of mind, you are estimating some of these calculations subconsciously, but I'm not sure.
Those categories could help the player see whether he has already used his "luck credit" or not. For example, if you completed flush draw on the river 4 times in a row, it would be not smart to think you can do it 5th time in a row.
Those categories could help the player see whether he has already used his "luck credit" or not. For example, if you completed flush draw on the river 4 times in a row, it would be not smart to think you can do it 5th time in a row.
ITryDeuces, why do you think competing a flush last hand makes it less likely to happen next hand?
I'm curious, how does the deck remember that it owes you a flush?
ITryDeuces, why do you think completing a flush last hand makes it less likely to happen next hand?
I'm curious, how does the deck remember that it owes you a flush?
I'm talking about a few hands in a row combo. Often during the game you will get lucky in some situations. Using a combination of categorized statistics you could switch from one luck category to another during a match or just simply estimate how likely is the next situation to happen. It's good to know how lucky you can be, and how lucky you can't be.
Let's say you get AA pre flop once, then in the next hand you get AA again, so in the next hand you would need a miracle to get AA for the 3rd time in a row.
Same thing here. It's not that the deck owes you a flush, like you said. It's exactly opposite - it owes you a miss after a few hits.
In stock markets, when you have a graph and the stock price rises continuously, at some point(s) it must do a "correction" (I'm not sure if that's the right English word for it) and the graph will go down there to gain some overall balance. It's almost impossible that the price will go up without correction points. So when the randomness gives you lucky situations in poker, you need to estimate when to stop "milking" your luck or switch to a different luck category.
When my yellow line gets too low I like to lure luck back out with some impressive, inebriated calculations. And then, guess what? It starts going up again.
Coincidence? Or strong, fact based evidence that you can switch luck timelines with vibe distorting attitudinal shifts?
Switching luck category is the key to trick randomness and use what NLHE mechanics offer.
Switching luck category is the key to trick randomness and use what NLHE mechanics offer.
There are different mechanics at work in the stock market, but for randomized cards, dice, or flipping a coin for example, what you are trying to allude to is a fallacy and is incorrect.
You are stating that if a fair coin is flipped 10 times and for those 10 flips they are all heads, then for the 11th flip it is more likely to be tails—this is false.
I'm talking about a few hands in a row combo. Often during the game you will get lucky in some situations. Using a combination of categorized statistics you could switch from one luck category to another during a match or just simply estimate how likely is the next situation to happen. It's good to know how lucky you can be, and how lucky you can't be.Let's say you get AA pre f
How does the deck remember that you've been dealt AA twice in a row? What is the physical mechanism that makes AA less likely on the 3rd deal?
Let's say you flip a fair coin and get heads 4 times in a row. By your logic, it's less likely that it would be heads on the 5th flip, correct? But if so, how does the coin "remember" that it has been heads 4 times in a row, and how does it alter its own probability of landing heads? What is the physical mechanism?
I'm aware that hitting a combo doesn't change the probability of the next deal, coin flip, etc. But when the combo accumulates and is for example 4 successful hits in a row then the next try becomes more important, because if you hit it, it will be considered as unusual luck, miracle, etc. So combo is equal to situations where you try to hit a one timer - for example, everyone has hit a Royal Flush at least once, but if you don't play poker and just try one hand dealt and you hit a Royal Flush then it's an extraordinary situation. Well, if something I'm reaching for is said to be "extraordinary luck" then I'd rather not try it.
You can look from the perspective of a single deal and get normal "%", and you can also look from the perspective that considers previous tries, it also does provide "%" for certain combos. And poker should get a more overall perspective instead of focusing on each hand separately, in my opinion. In Fast Fold poker you play faster, so it's a bit easier to grasp that long run perspective on what is and what was happening at the tables.
What's very important - "switching luck category" doesn't necessarily have to connect with cards appearing on the board. It may as well refer to certain actions (like making opponent fold). Then the % value of the next try is no longer measurable. So the bigger the combo, the less likely you'll be successful in the next try.
There's a complete freedom whether it comes to "switching luck category" or "combining categorized statistics". Countless options allow different ways of executing. It may as well be impossible to control this all the time properly due to the infinite number of ways you can use it. Someone would have to organize it more precisely, then maybe it would become easier to use.
You can compare a Fast Fold match vs GTO bots to a tournament, in terms of stacks accumulation. This kind of approach I'm explaining here probably works better in tournaments. Luck is never infinite, so you should know when to stop when there is more money on the line.
^^^^ All nonsense.
You are stating that if a fair coin is flipped 10 times and for those 10 flips they are all heads, then for the 11th flip it is more likely to be tails—this is false.
Wait… but is it?
If we can predict the % chance over the next 100 flips with 87% (or w/e) certainty that an x% of those next flips will be tails (using the same logic that gives us the 50/50 if we flip one coin in a vacuum, just spread out over a larger dataset where variance is ironed out), then the actual ‘informed’ odds (which now recognises the context of the probability) are in fact very very very slightly increased in Tail’s favour. Maybe it’s only 0.0000000000000003866%, depending on your timeframe, but nevertheless, a clear hypothetical edge (that could theoretically be exploited by an immortal primate?).
If a coin comes up heads ten times in a row, I'm betting on heads for the next flip.
Well you have a 49.999999999999999999997244% of being right, so not a huge error.
I've explained the gambler's fallacy to dozens of people and it's always interesting to see how they confront this paradox:
Those categories could help the player see whether he has already used his "luck credit" or not. For example, if you completed flush draw on the river 4 times in a row, it would be not smart to think you can do it 5th time in a row.
Imagine there's a machine called "Money Milking Machine" that gives you 60% of winning and doubles your bankroll each time you win. You start with $1 and you can hit the MILK button only 5 000 times total.
With your type of thinking, you'd never stop before getting something like $10 000 000 in one combo. It wouldn't make any sense for you to stop earlier, because you have 60% each time. In other words, you'd never win anything on this machine.
It's quite obvious that you should cash each combo that got big enough, like $1024 or so. Then maybe you could start from $1 again and try to repeat that.
So the conclusion is that the desired combo sequence should be divided into smaller parts instead of being happy with magical 60% and going full front ahead.
Imagine there's a machine called "Money Milking Machine" that gives you 60% of winning and doubles your bankroll each time you win. You start with $1 and you can hit the MILK button only 5 000 times total.With your type of thinking, you'd never stop before getting something like $10 000 000 in one combo. It wouldn't make any sense for you to stop earlier, because you have 60% e
Do you lose your entire bankroll if you lose? If so, where do you get the $1 to start again?
Also, see "risk of ruin"?
So the conclusion is that the desired combo sequence should be divided into smaller parts instead of being happy with magical 60% and going full front ahead.
But the 60% remains 60% every round regardless of past wins or losses.
Sounds like you are talking about bankroll management. Something like kelly criterion could be used to calculate the optimal bankroll strategy.
Do you lose your entire bankroll if you lose? If so, where do you get the $1 to start again?
I assume the game only allows us to start with $1 even if we have much more money in our pocket. And the game only allows us to either double our bet or quit and start over from $1 every round.
It's a pretty interesting game to be honest. Strategy would depend on how rich the person already is.
A poor person would use a more conservative strategy and a billionaire would play a very risky strategy because the utility of money is different.
