What is Pabrai's Blackjack Secret?

[Mephistopheles]

In this 5/21/20 Fireside Chat video, Pabrai discusses being banned from playing Blackjack at a Vegas casino. He says that it did not involve card counting but it was a winning strategy and that he made around $100k.

 

I am very familiar with card counting but am surprised to hear there is another system? Anyone know what he could be talking about?

 

 

Starts around 15 minutes

[ActOfWill]

He mentioned bj21.com so I think it has perhaps to do with some rule variations that differ per table that are exploitable. Perhaps some side bet rules.

 

But the way he speaks about it "51% edge vs 80% in stock markets" (later on) seems to be slightly inconsistent as that are card counting odds to me.

 

Have to say he exploits it quite well for personal marketing as he latches on to the Ed Thorpe bandwagon. Personal opinion: virtue signalling.

[Guest cherzeca]

no one can get on the Ed Thorpe bandwagon. not even Fisher Black could.

[Orchard]

In many domains you have people signaling their intelligence by walking around and talking about their gambling exploits.

 

The question i would pose them is: If you're so good at gambling how come you don't do it full time?

 

The most likely answer is that if you gamble in a low-risk way with a strategy that gives you favorable odds you are making peanuts.

 

The only way to make big bucks is to play with big stakes, but due to the swings inherent in gambling you will most likely run out of money.

[Munger_Disciple]

He mentioned bj21.com so I think it has perhaps to do with some rule variations that differ per table that are exploitable. Perhaps some side bet rules.

 

But the way he speaks about it "51% edge vs 80% in stock markets" (later on) seems to be slightly inconsistent as that are card counting odds to me.

 

Have to say he exploits it quite well for personal marketing as he latches on to the Ed Thorpe bandwagon. Personal opinion: virtue signalling.

 

In the video (around 22:40 min mark) Pabrai actually says that he is "fighting to get 50.1% odds vs 49.9%" which makes me wonder even more.

[Foreign Tuffett]

I know very little about blackjack, but any casino game in which the 'house' is your counterparty is nearly always going to have a negative expected value.

 

Have their been some exceptions in the past? Yes, but they almost prove the rule since (a) they are so notable that people talk about them for years on end (b) they usually rely on the house making mathematical errors, which is rare.

 

The writer known as "Bob Dancer" (a pseudonym) supposedly made a mint back in the day because one or more casinos screwed up the expected payouts on some video poker machines.

 

Don Johnson (not the actor) made millions because a few casinos screwed up big time by allowing him to negotiate special blackjack rules so that the the odds were actually in his favor.

 

https://www.theatlantic.com/magazine/archive/2012/04/the-man-who-broke-atlantic-city/308900/

 

Pabrai didn't say anything in the interview about how his blackjack "system" actually works. He did mention he uses a $100 subscription service to decide when and where he will play to maximize his odds.

 

Can I believe he won some money playing blackjack in Vegas? Yes. Do I think he figured out some magic formula to win? Nah.

 

Finally, I would love to hear an explanation of his statement that "in the equity markets I can make bets where the odds are 80%." In what meaningful sense can he (or anyone else) precisely calculate the odds of a particular equity investment being successful?

 

 

Hopefully someone more knowledgeable about blackjack will chime in..

[Saluki]

In Fortune's Formula they mention some earlier iteration of the card counting system where (If I recall), you only counted the face cards.  I forget why they abandoned it in favor of the counting method.  Maybe it worked, but the counting system works better when you have a team approach?  I still have the book somewhere but am not interested enough to look it up since I don't gamble.

[ActOfWill]

 

Finally, I would love to hear an explanation of his statement that "in the equity markets I can make bets where the odds are 80%." In what meaningful sense can he (or anyone else) precisely calculate the odds of a particular equity investment being successful?

 

 

Pabrai is effectively saying that he is 80% directionally right on each investment. Now the comparison does not hold fully when you consider varying time horizons. But that would be the gist.

[Gregmal]

The original card counting systems were scrapped largely when casinos went from a 2-4 deck approach to using 8+ decks and reshuffling half way through the round.

[nickenumbers]

We are such math, probability, numbers people.  2-4 decks, 49.9% probabilities, 80% directional confidence...

 

You guys are my tribe! ;D

[Foreign Tuffett]

 

Finally, I would love to hear an explanation of his statement that "in the equity markets I can make bets where the odds are 80%." In what meaningful sense can he (or anyone else) precisely calculate the odds of a particular equity investment being successful?

 

 

Pabrai is effectively saying that he is 80% directionally right on each investment. Now the comparison does not hold fully when you consider varying time horizons. But that would be the gist.

 

I still don't get it. How is he able to calculate with any degree of precision that a stock has a 4 out of 5 chance of going up? What would such a calculation even look like?

 

[Gregmal]

 

Finally, I would love to hear an explanation of his statement that "in the equity markets I can make bets where the odds are 80%." In what meaningful sense can he (or anyone else) precisely calculate the odds of a particular equity investment being successful?

 

 

Pabrai is effectively saying that he is 80% directionally right on each investment. Now the comparison does not hold fully when you consider varying time horizons. But that would be the gist.

 

I still don't get it. How is he able to calculate with any degree of precision that a stock has a 4 out of 5 chance of going up? What would such a calculation even look like?

 

He's not. I saw a decent conversation on Twitter(a rarity) the other day making the same point about how one should run, and fast, from any money manager talking about using Kelly criterion as part of their strategy. You cant use gambling based, probability reliant inputs for things(and odds) that are not stationary. Its marketing bullshit.

[OracleofCarolina]

Pabrai has said many times that if he did revise " The Dhando Investor" book, he would leave out the chapter on the Kelly Criterion as it does not work in investing....he admits it was a mistake

[Parsad]

 

Finally, I would love to hear an explanation of his statement that "in the equity markets I can make bets where the odds are 80%." In what meaningful sense can he (or anyone else) precisely calculate the odds of a particular equity investment being successful?

 

 

Pabrai is effectively saying that he is 80% directionally right on each investment. Now the comparison does not hold fully when you consider varying time horizons. But that would be the gist.

 

I still don't get it. How is he able to calculate with any degree of precision that a stock has a 4 out of 5 chance of going up? What would such a calculation even look like?

 

He's not calculating the probability of the stock going up or developing some calculation for that.  What he's saying is that 4 out of 5 times, after he analyzes a business and invests in it, the stock will eventually move closer to intrinsic value 80% of the time.  And he will be wrong roughly 20% of the time.  Cheers!

[LC]

Edge sorting

Hole card

Front loading

 

https://www.888casino.com/blog/blackjack-tips/how-to-win-in-blackjack-without-card-counting

https://www.888casino.com/blog/edge-sorting/what-is-edge-sorting

[Parsad]

In many domains you have people signaling their intelligence by walking around and talking about their gambling exploits.

 

The question i would pose them is: If you're so good at gambling how come you don't do it full time?

 

The most likely answer is that if you gamble in a low-risk way with a strategy that gives you favorable odds you are making peanuts.

 

The only way to make big bucks is to play with big stakes, but due to the swings inherent in gambling you will most likely run out of money.

 

Even with big bucks, it's hard to beat casinos because of table limits.  If you can somehow play in a high stakes game, with no table limits, then a good player can do well. 

 

High stakes poker tables prove that there is a certain amount of skill involved and those abilities can distinguish between the good and great players.  But unlike statically picking stocks, there is an element of luck with gambling...that does not exist in stock-picking, unless you are throwing darts at a list of the S&P500 and hoping one of them will be a 100-bagger!  Cheers!

[Guest cherzeca]

kelly criterion is a heuristic device.  assuming that you know what (how much in probabilistic terms) your edge is, Kelly math tells you how much to risk. big assumption!  Thorpe used Kelly, but he also knew that he had an edge when he developed his warrant pricing model well before anyone else had a clue.

[Foreign Tuffett]

 

Finally, I would love to hear an explanation of his statement that "in the equity markets I can make bets where the odds are 80%." In what meaningful sense can he (or anyone else) precisely calculate the odds of a particular equity investment being successful?

 

 

Pabrai is effectively saying that he is 80% directionally right on each investment. Now the comparison does not hold fully when you consider varying time horizons. But that would be the gist.

 

I still don't get it. How is he able to calculate with any degree of precision that a stock has a 4 out of 5 chance of going up? What would such a calculation even look like?

 

He's not calculating the probability of the stock going up or developing some calculation for that.  What he's saying is that 4 out of 5 times, after he analyzes a business and invests in it, the stock will eventually move closer to intrinsic value 80% of the time.  And he will be wrong roughly 20% of the time.  Cheers!

 

If you are suggesting that he is extrapolating from his past experience as an investor to come to that 80% figure*, then he is just making a basic logical fallacy. Surely that isn't what he means?

 

 

* "80% of my equity investments in the past have been successful, therefore 80% of my equity investments in the future will be too."

 

 

[Parsad]

 

Finally, I would love to hear an explanation of his statement that "in the equity markets I can make bets where the odds are 80%." In what meaningful sense can he (or anyone else) precisely calculate the odds of a particular equity investment being successful?

 

 

Pabrai is effectively saying that he is 80% directionally right on each investment. Now the comparison does not hold fully when you consider varying time horizons. But that would be the gist.

 

I still don't get it. How is he able to calculate with any degree of precision that a stock has a 4 out of 5 chance of going up? What would such a calculation even look like?

 

He's not calculating the probability of the stock going up or developing some calculation for that.  What he's saying is that 4 out of 5 times, after he analyzes a business and invests in it, the stock will eventually move closer to intrinsic value 80% of the time.  And he will be wrong roughly 20% of the time.  Cheers!

 

If you are suggesting that he is extrapolating from his past experience as an investor to come to that 80% figure*, then he is just making a basic logical fallacy. Surely that isn't what he means?

 

 

* "80% of my equity investments in the past have been successful, therefore 80% of my equity investments in the future will be too."

 

If Buffett has historically over 60 years picked 95% correctly, you don't think he's going to be close to that percentage in the next 10 years?  It's not a guarantee of any sorts, but unless he is suffering from some sort of mental handicap or completely changes his methodology, you don't think he would be close to that percentage going foward?  That's not a logical fallacy...that's reality!  He's not rolling dice...he's developed an advantage through his methodology.  Cheers!

[winjitsu]

https://www.theatlantic.com/magazine/archive/2012/04/the-man-who-broke-atlantic-city/308900/

 

It's possible to gain a slight edge if offered loss rebates.

[clutch]

In many domains you have people signaling their intelligence by walking around and talking about their gambling exploits.

 

The question i would pose them is: If you're so good at gambling how come you don't do it full time?

 

The most likely answer is that if you gamble in a low-risk way with a strategy that gives you favorable odds you are making peanuts.

 

The only way to make big bucks is to play with big stakes, but due to the swings inherent in gambling you will most likely run out of money.

 

Even with big bucks, it's hard to beat casinos because of table limits.  If you can somehow play in a high stakes game, with no table limits, then a good player can do well. 

 

High stakes poker tables prove that there is a certain amount of skill involved and those abilities can distinguish between the good and great players.  But unlike statically picking stocks, there is an element of luck with gambling...that does not exist in stock-picking, unless you are throwing darts at a list of the S&P500 and hoping one of them will be a 100-bagger!  Cheers!

 

There is definitely luck in investing, even if you are the world's best stock picker. If it was all skills, we would see a lot more consistency among high performers which is quite the opposite.

 

Another way to look at it:

 

If you play 1-on-1 basketball against Michael Jordan at his prime 100 times, you will never able to meet him. Because it's 99.999% skills in 1-on-1 basketball.

 

On the other hand, if you have a stock picking competiton against the best stock picker in the world 100 times, you will be able to beat them a handful amount of times.

[Guest cherzeca]

Ted Williams explained in the science of hitting that you have to swing only at good hittable pitches to have a high batting average.  He also was known for outstanding eyesight, and could distinguish curves early in the pitch (could tell where the pitch was going).  Disciplined process and vision (clarity of applying the process) works well in investing too.  I would compare investing to baseball hitting rather than gambling.

[tede02]

no one can get on the Ed Thorpe bandwagon. not even Fisher Black could.

 

LMAO! Ed Thorpe is the man. His autobiography is one of my favorite books.

[dwy000]

There's a roulette system called Reverse Labouchere that claims to have broken the banks in Europe and the strategy is replicable in blackjack - probably even more so because there's no "0" or "00" so your odds can be better.

 

There is a book called Thirteen Against the Bank that describes how they did it (pretty interesting read).  Roughly speaking the Labouchere method uses the premise of a near 50/50 bet (i.e. a blackjack hand) and if you win, you pocket the money and if you lose you double down over and over until you win (its more complicated than that but that's the general gist).  The problem was that if you went on a losing streak you would hit the bet limit before you could win back the original bet - and the combined bets would make you broke.  So they reversed it and if you won you would double down over and over again until you lost.  If you went on a run of doubling your bet each time you would hit the casino limit at which point you would start over at the minimum bet.

 

I read it like 20 years ago and I'm shocked given all the blackjack card counting books and movies this hasn't got more light shone on it.

[Guest cherzeca]

There's a roulette system called Reverse Labouchere that claims to have broken the banks in Europe and the strategy is replicable in blackjack - probably even more so because there's no "0" or "00" so your odds can be better.

 

There is a book called Thirteen Against the Bank that describes how they did it (pretty interesting read).  Roughly speaking the Labouchere method uses the premise of a near 50/50 bet (i.e. a blackjack hand) and if you win, you pocket the money and if you lose you double down over and over until you win (its more complicated than that but that's the general gist).  The problem was that if you went on a losing streak you would hit the bet limit before you could win back the original bet - and the combined bets would make you broke.  So they reversed it and if you won you would double down over and over again until you lost.  If you went on a run of doubling your bet each time you would hit the casino limit at which point you would start over at the minimum bet.

 

I read it like 20 years ago and I'm shocked given all the blackjack card counting books and movies this hasn't got more light shone on it.

 

looks to me like there is an ergodicity problem here. and this goes against Kelly since there is no edge