Female Player

[shadroch]

I think what Scott is trying to express is that the advantage of any bet is a function of the parameters of the gaming device and not the amount of money in the player's pocket. When you are a theorist and calculating a game's advantage, you don't consider the amount of money being wagered because you aren't wagering any- this is theoretical. RJT is looking at it from the perspective of a professional advantage player where the advantage is defined as how much advantage you can bring to bear on your own bankroll thus it is bankroll dependent. No need for anyone to get upset.

If we were playing theoretics,I might agree. But we are not. We are discussing playing BJ,with real money.Im fairly sure that there is a great theoretical math board somewhere on the internet where lively discussions on moebious strips,and prime colors occur. This board is dedicated to the former.
When somone needs to go into the super-theortical to defend his postions and attack some of the worlds best players because they base there game on the real world,they have crossed into the pale.

 

[KenSmith]

Looks like I've been missing out on quite a thread here.

I agree with ScottH and AutomaticMonkey on one key contentious point here. While overbetting your bankroll is disastrous for the individual player, the casino CAN NOT make money off legions of overbetting APs. A few very lucky players will win back all the money that their poorly-informed comrades have lost, and then some. In fact, in the long run they'll win exactly as much as the aggregate advantage over all the players.

While each player is overbetting their individual bank, their combined bank is easily large enough to handle their elevated betting level. Heck, at triple Kelly, you only need three players to make the math work. Think of the world full of underbankrolled counters as a massive team, where they never break the bank. Get it yet?

If you're still not convinced, let's play a fair coin flip game, continuing until one of us has all the money. We'll bet $10 per coin flip, but I'm underfunded with a bankroll of only $100 compared to your $1000. Do you have an advantage in this game? Of course not. Yes, you'll usually get my $100, but I'll win your $1000 in one out of 11 trials. My expected value is $0, and so is yours.

 

[RJT]

I completely agree with you Ken, but the issue is real world gambling and as you only have a 1 in 11 chance of winning my bank compared to my 10 out of 11 chance of winning your bank as an individual player, you're not playing with the advantage. In fact as a one off, i'm playing with a might advantage, even though if i keep playing against people like yourself in the long run my EV is 0 (technically my EV is 0 even in the short run, but if you were to offer a one shot go at this, which side would you rather take?)
In the real world, these players aren't playing together, and while blackjack litrature is filled with storys of players making a killing after starting dramatically undercapitalized, these people are the exception. For most of the people that try this, they will lose and and not have the money to make it back to the point where their bankroll equals their expectation.

RJT.

 

[Sonny]

It looks like RJT explained this pretty well, but I think I can sum it up in simple terms.

That just means you have a close to 100% ROR. It DOES NOT mean you don't have an advantage.

The problem is that you’re only looking at one hand, not the big picture. What good is having an advantage if you don’t have any money to bet on the next one? If you go broke then you lose your advantage forever. You can backcount all you want, but if you’re not making any bets then you do not have an advantage. You may have had an advantage on that one hand, but your bet had a direct impact on all future advantages.

You never have a 100% ROR, even betting it all every single hand. You just have a really high ROR, but your advantage is still there, even though you are overbetting.

It’s true that your ROR on that one hands is not 100% because you have a small advantage (although it is close enough to 100% for most people), but if you consistently bet all of your money on every hand then your ROR becomes a full 100%. You will eventually lose all of your money, no question. And once you go broke you lose your advantage forever. That is how you lose your advantage by overbetting.

Although you’re right that the size of your bet does not affect your advantage on the current hand, it has a huge affect on your overall advantage, which is what most APs are concerned with.

ROR and advantage are completely different things that people are getting confused.

No, RoR is a function of your advantage, variance and bankroll. They are all directly related. You’re thinking of advantage as a short-term gain and not something that must be managed in the long-term.

-Sonny-

 

[EasyRhino]

Think of the world full of underbankrolled counters as a massive team, where they never break the bank. Get it yet?

Yep, I'm going to need to back off from my previous musings.

I still think there might be some situation where a casino would tell the players to "bring it". I think it would require:

1) A large, but finite, number of gamblers
2) Continuous overbetting
3) Very small starting bets, or a very large casino bankroll.
4) (Optional, but highly recommended) 86 a player when they hit the max bet

... and even then, you'd have a casino which would burn out a large number of players. I'm sure they would prefer that those players play a losing game over a longer time. (Rather milk the cow than eat the hamburger, or whatever).

 

[Automatic Monkey]

I think it might be helpful to clear up the different uses of this word "advantage."

1) The expected gain from placing a bet. For example, in one game I play, when there is a true count of +8 there is an advantage of 2.2%. Any player who places any bet on that next hand (assuming he uses the correct playing strategy and has enough money to double and split) will experience an advantage of 2.2%. It has nothing to do with bankroll.

2) The expected gain from playing a game. In this same game, my spread and strategy give me an overall advantage (IBA) of 1.25%. As long as I have enough money to spread over complete shoes, my average advantage on my action is 1.25%. It has nothing to do with bankroll.

3) The benefit to an entity from taking a particular course of action. In the case of gaming, it means placing bets with an advantage and at the same time sizing your bets strategically in order to maximize one's chances of actually benefiting from this advantage, given the variance inherent in the game. It has everything to do with bankroll.

We're using "advantage" in both the mathematical and colloquial sense and the two senses are being confused.

 

[Sonny]

I think it would require:

1) A large, but finite, number of gamblers
2) Continuous overbetting
3) Very small starting bets, or a very large casino bankroll.
4) (Optional, but highly recommended) 86 a player when they hit the max bet

I think those two elements are the most important. The casino must have a large enough bankroll to sustain the swings, and the gamblers must continuously overbet their bankrolls. Even when a few of them get lucky and reach “escape velocity” they will resize their bets so that their RoR returns to 100% and I am guaranteed to get their money in the end. Or, alternatively, the casino would need to make sure that the gamblers are not reinvesting their winnings back into their bankroll. If they are spending all of their winnings in the restaurants and strip clubs then their bankrolls will eventually be lost.

-Sonny-

 

[bluewhale]

i'm not going to point out individual names but the bottom line is this...

casinos DO NOT make money on people who play with an advantage (even if it is small and they dramatically overbet)!

that being said, people who overbet their bankroll and claim to have only a small advantage (due to imperfect counting) almost always don't actually have an advantage at all. they make too many errors, tip too much, and buy too much stuff (food, stripper's time, rooms, drinks, etc.) to completely negate their puny edge and are rightly welcomed back into casinos by the staff.

also, is ignoring RoR and dramatically overbetting a good strategy from an AP perspective?? obv. not, neither scott or AM are claiming this.

 

[KenSmith]

We're using "advantage" in both the mathematical and colloquial sense and the two senses are being confused.

I think that is exactly the reason there is disagreement in this thread.

 

[ScottH]

I think that is exactly the reason there is disagreement in this thread.

It is obvious that not everyone is referring to advantage as the exact same thing, and this is part of the argument.

I did think of a possible reason for the disagreement. I think the people who are saying that overbetting decreases your advantage really mean to say overbetting decreases your lifetime winrate.

So we have one player that always resizes his bets to 2x kelly, and one who always bets exactly regular Kelly. The one overbetting is going to lose his bankroll after a while, but the other player will play forever since he is always betting a percent of his bankroll and thus will make more money in the long run.

By overbetting the player lowered his lifetime winrate because he is sure to lose his money and stop making money. I still think that they both have the same advantage, but the overbetting player eventually will not be able to use it since he is broke. I think he still has the advantage since he is a good player, he just won't have a bankroll to apply it to. Also, the overbetting player will have lowered his lifetime winrate.

So everyone seems to be right, it's just a miscommunication on terms.

So I think advantage is a function of the game, something that is independent of bankroll. I also think other people are trying to tell me overbetting decreases your lifetime winrate, which I also agree with. So everyone could be right here, it's just that we are all making different points.

What does everything think about this?

 

[Kasi]

I think the people who are saying that overbetting decreases your advantage really mean to say overbetting decreases your lifetime winrate.

OK - I meant to say consistently overbetting your advantage decreases your lifetime winrate - to ZERO. Oh wait, I did say that.

Also, I think consistently betting 2X Kelly is a break-even proposition, not a losing one as you say in your example. Betting >2X is a losing one.

And I'm glad that a "bunch of ploppies" like us and a pro like you can, after all, all seem to be right. What a beautiful world we live in.