Don Johnson - Blackjack Strategy
A lot of people say he had a good rebate plan. But that doesn't seem to matter. Since he won at all the casinos he played.
He was flat betting. So obviously no card counting in terms of initial bets.
Also, a couple of weeks later he played at a table with rules available to all customers. And won another $2m
Is there a new version of Basic Strategy that he used?
The blackjack rules were 6 decks, DOA, DAS, S17, LSR, RSA, with a house edge of 0.263%.
Quote: PinkJackFor those unfamiliar. Don Johnson was the guy who won upwards of $6m playing blackjack. While a lot of people argue that he had negotiated terms that lowered the house edge. The house still had an 0.26 edge over his play. I've tried to run simulations using the same game rules available to him. And something doesn't seem to add up.
A lot of people say he had a good rebate plan. But that doesn't seem to matter. Since he won at all the casinos he played.
He was flat betting. So obviously no card counting in terms of initial bets.
Also, a couple of weeks later he played at a table with rules available to all customers. And won another $2m
Is there a new version of Basic Strategy that he used?
The blackjack rules were 6 decks, DOA, DAS, S17, LSR, RSA, with a house edge of 0.263%.
The loss rebate still mattered even though he always won. The rebate is extra equity if he would have lost.
Because if he was using pure Basic Strategy he should've lost on the long run
Quote: PinkJackBecause if he was using pure Basic Strategy he should've lost on the long run
Why do you think that? He got 20% back for losing over $500k. That's definitely worth more than the house edge of the game.
So he basically would go into every session thinking, well, I'm gonna lose $500k this session (and get $100k of it back), or win a bunch before quitting. Luckily for him, all the sessions were apparently "win a bunch".
Also, I think at one of the casinos he played for 7 hours straight and won $6m.
So has to be more than luck in that case. I'm assuming a variation of Basic Strategy.
Maybe someone who's researching this could give a mathematical answer
It's not as though he won huge in a single short session - he was grinding several places at once, building those millions against his rebate, over some months.
There's about an hour-long thing on him on YouTube, (the one im referencing above) that he allowed them to follow him around in LV, and he talks some about it. It's probably 10-15 years old.
Mathematically, if you have a 20% loss rebate on a -.26% game, and you play perfect strategy, you would have about a 9% edge, right? So if you have the br to get 60m in play and can ride out the variance, you should end up 5-6m ahead. I'm not a math guy, so again, I could be wrong.
In that they mention that he was playing for 12-hours straight. One-on-one with the dealer. And he was betting $100,000 a hand. (Even his wiki page mentions a 12-hour marathon at the Tropicana casino).
Definitely not luck or loss rebates.
Also, I dont think someone would risk millions because they get a loss rebate. When you're going in with that kinnda money. You're expecting to win. Loss rebates dont explain how he won the game.
I think he had planned to stop at around $2.5m which would be around 25 hands.
I've also seen a guy play recently at a local casino. Came in with $10k. And was flat betting $500 on every hand (20-hands of play). And walked out with $100,000. They next day he came, the pitboss asked him to leave.
My guess is there's a new version of Basic Strategy that they're following. Just a guess. Guessing someone in the community has an answer.
Edit: Also I think the answer must be something really simple (like a BS chart). Because in one interview when the report asked him how he did it, his answer was 'if the sample size is large enough you win.' And in another interview I've heard him say, 'I won because the sample size was too small.'
Quote: PinkJackIs there a new version of Basic Strategy that he used?
No. If he wasn't up to some advantage play, then he is smart enough to use the proper basic strategy. It's my understanding that he used a mix of advantage plays, but loss rebates was the main one.
Also, I've seen a couple of people pull off the same stunt but with lower bet amounts at the casino. Flat bet. And enough bankroll for 20-hands. So wondering.
Edit: I've been practicing Blackjack at home. If I start with 200 units and bet 10 units a game (20-hands; 6-decks; cut-card). I eventually deplete my brankroll.
Quote: PinkJackHow about the 12-hour play at Tropicana where he won $6m. No rebates there. Seems like he won fair and square. Is it possible he used Basic Strategy with variations based on TrueCount and RunningCount? (Though he doesn't seem like the kind of person who'd be able to pull of counting. Since its really difficult to count cards).
Also, I've seen a couple of people pull off the same stunt but with lower bet amounts at the casino. Flat bet. And enough bankroll for 20-hands. So wondering.
Edit: I've been practicing Blackjack at home. If I start with 200 units and bet 10 units a game (20-hands; 6-decks; cut-card). I eventually deplete my brankroll.
I haven’t read too much about this but is it possible there is only reading material on his huge wins and not his losses?
Quote: beachbumbabsIt's been several years, but he said something (in a documentary piece about him) that he used having a lot of people around him, including beautiful women, to help distract from the cards, and benefited from some dealer mistakes. But I think the biggest thing was the house rebates on losses that overcame the HE, plus he ran well overall. I could be wrong.
It's not as though he won huge in a single short session - he was grinding several places at once, building those millions against his rebate, over some months.
There's about an hour-long thing on him on YouTube, (the one im referencing above) that he allowed them to follow him around in LV, and he talks some about it. It's probably 10-15 years old.
Mathematically, if you have a 20% loss rebate on a -.26% game, and you play perfect strategy, you would have about a 9% edge, right? So if you have the br to get 60m in play and can ride out the variance, you should end up 5-6m ahead. I'm not a math guy, so again, I could be wrong.
Here it is Babs :)
[edit] Initial intuition says to most of us, for the House, "a rebate on *actual* loss has to be better than a rebate on *theoretical* loss. No?".
Well, no. It is not, turns out.
Phil Ivey, whom I dislike, had a similar deal at craps in Las Vegas and took advantage of it,
Quote: odiousgambitI think the OP does not have a grasp of the loss rebate advantage in the right hands. It is counter-intuitive .
Initial intuition says to most of us, "a rebate on *actual* loss has to be better than a rebate on *theoretical* loss. No?".
Well, no. It is not, turns out.
I always assumed they gave him a rebate on actual losses. Are you saying they gave him a rebate on theoretical losses?
It may be there were never losses made on which to give rebate. That's part of the counter intuitive aspects of loss rebate bonuses: You don't have to ever make a loss to benefit from them.Quote: darkozI always assumed they gave him a rebate on actual losses. Are you saying they gave him a rebate on theoretical losses?
E.g. If you are offered a 100% loss rebate on a $1,000 wager on a coin toss and win the coin toss, then was the loss rebate valuable $:o)
Nowadays there are casinos in Vegas with lower house advantage than that on double deck BJ. All the MGMs and T.I. have as low as 0.19 and M Resort has 0.14 - are you sure Don Johnson was able to get it only down to 0.26 ? If so, the standard six deck at MGMs today ($100. min.) is 0.26 too.
Anyway, from what I have read about his play, it may have come down to something few on this forum wish to recognize - the power of getting up and leaving after a winning streak.
No, if they had, they couldn't have gotten hurt in the long run, assuming they don't give more rebate, including comps, than would ever be given.Quote: darkozI always assumed they gave him a rebate on actual losses. Are you saying they gave him a rebate on theoretical losses?
When I say a rebate on actual/theoretical is good or bad, I mean from the perspective of the House. That wasn't clear , I went back and edited that in the post you quoted.
I believe he also said in that documentary that he took advantage of the Great Recession, and subsequent desperate casinos, and that he couldn't have done what he did in any other economic environment (i.e., now, or any time before).
Quote: MDawg<<The house still had an 0.26 edge over his play.>>
Nowadays there are casinos in Vegas with lower house advantage than that on double deck BJ. All the MGMs and T.I. have as low as 0.19 and M Resort has 0.14 - are you sure Don Johnson was able to get it only down to 0.26 ? If so, the standard six deck at MGMs today ($100. min.) is 0.26 too.
Anyway, from what I have read about his play, it may have come down to something few on this forum wish to recognize - the power of getting up and leaving after a winning streak.
I like this comment
Quote: PinkJackI like this comment
His comment is incorrect.
Mr Johnson was playing at a tremendous advantage.
Suppose you and are flipping a coin. It's a 50-50 proposition. In the long run, we should break even. Now consider that on the days, I win, you get nothing, but on days you win, I get 20% of your winnings. Do I need do anything to make a long term profit?
Would you say the odds are still 50-50 long term.
Now, change those rules so I can quit at any point I want. Let's say I play for an hour and am even. I'm betting 100K a hand and suddenly lose five hands in a row. I choose to stop. I'm down 500,000 but get $100,000 back.
The next day, I'm down $50,000 when I win five hands in a row and leave up $450,000.
In two days, my real results are that I lost $50,000 but thanks to the loss rebate, I am up.
Rinse and repeat.
In the real world, sessions are meaningless. All life is a session. However, If you find someone foolish enough to rebate twenty percent of your session losses and let's you define the session, you've struck gold.
Quote: MDawg<<The house still had an 0.26 edge over his play.>>
Nowadays there are casinos in Vegas with lower house advantage than that on double deck BJ. All the MGMs and T.I. have as low as 0.19 and M Resort has 0.14 - are you sure Don Johnson was able to get it only down to 0.26 ? If so, the standard six deck at MGMs today ($100. min.) is 0.26 too.
Anyway, from what I have read about his play, it may have come down to something few on this forum wish to recognize - the power of getting up and leaving after a winning streak.
At that time, Atlantic City was only allowed to offer 8 deck shoes.
I dont think that is correct,I believe most casinos in Ac had 6 deck in hi limit,and some casinos had single deck.Quote: FCBLComishAt that time, Atlantic City was only allowed to offer 8 deck shoes.
Quote: WizardI attended a talk he gave several years ago. He mainly spoke of the loss rebates, but he never denied other advantage plays were involved. From word-of-mouth, I hear he was also counting cards (but not so aggressively as to get caught) and inducing dealer errors in his favor. I'm not big on quoting "word of mouth," but that's all I've got.
He was flat betting, to the best of my memory, but was making play decisions based on the count. He encouraged fast sloppy dealing, and stated that he was getting at least 1 average bet per hour in dealer mistakes in his favor.
He had mathematically figured out EXACTLY the proper win/loss points to end his session. His loss rebate was per session.
There are a lot of people making decisions to allow players to take liberties with either the math, or with the rules (I'm looking at you Phil Ivey). That is why it is so important to have table games management who know how the games work, and how and why the house makes money.
https://www.blackjackapprenticeship.com/don-johnson-blackjack/
From the article.....
I thought, “I’m a Ph.D. mathematician; I can do that!” I got to work to figure out how Don Johnson did it. The main advantage that Johnson negotiated was a “loss rebate.” Simply put, if he lost money during a trip, then a percentage of his losses would be returned. This is a common incentive for high-rollers, but it is usually accompanied by a requirement for a minimum amount of play. Typically the high-roller is required to play at least 12 hours to qualify for his rebate. This play requirement allows the casino to earn enough “theoretical win” to compensate for the cost of the rebates when they are given. What gave Johnson the edge was that his loss rebate incentive had no minimum play requirement. Here are the details:
◾The blackjack rules were 6 decks, DOA, DAS, S17, LSR, RSA, with a house edge of 0.263%.
◾ The table maximum wager was $100,000.
◾Don Johnson could claim a 20% rebate on his losses any time he lost $500,000 or more.
◾There was no minimum play requirement.
◾The loss rebate reset every day.
It’s easy to see that this structure can be beaten. If Johnson simply quit for the day after either winning $500,000 or losing $500,000, then on his winning days he would keep the full amount, but on the losing days he would only lose $400,000 (after his 20% rebate). With a $100,000 wager, these quit points (winning or losing 5 units) would likely occur after just a few hands. But blackjack is so close to an even game that playing a few hands is about the same as a coin-flip. It follows that with this trivial strategy Johnson would win, on average, slightly less than $50,000 per day. But could he do better? The key to optimizing Johnson’s winnings was to determine his best “win/loss quit points.” These are the profit-maximizing win/loss dollar values at which Don Johnson would leave for the day and either keep his winnings or collect his rebate. I dusted off my old “Stochastic Processes” book from graduate school and soon proved a sequence of three theorems I call the “Loss Rebate Theorems”. When the “Loss Rebate Theorem” spreadsheet was used to analyze Don Johnson’s 20% loss rebate program, assuming a $100,000 wager, it yielded the following:
◾Don Johnson should quit after winning $2,411,000.
◾Don Johnson should quit after losing $2,597,000.
◾The probability of hitting the win-quit point in any given session was 49.07%.
◾Don Johnson was playing with an effective edge over the house of about 0.26%.
◾On average, the number of expected rounds to reach a quit point was 481.
◾On average, the expected win per day for Don Johnson was $125,000.
Quote: PinkJackHowever, that's not what he did.
Exactly. When Don Johnson played for 12 hours for example and simply left ahead, there was no rebate involved. You gotta lose first to get a rebate.
But try to explain this simple concept, or much of anything else to billryan, well...I'll leave it to someone else to try.
Another thing Johnson did is have other players jump in to eat up bad cards. I've actually done that quite successfully, I've had my friend sit at first base, and play small, with me the big player at second. When I felt that I needed it, I would have him take hits.
Of course, here comes billryan, who hasn't played in a casino in years, here to explain why that doesn't or wouldn't matter....
Don Johnson won. I win. When someone comes in and explains Repeatedly why it couldn't happen wouldn't happen nothing the player did would make a difference, that tells me...that the someone, lost.
And then as far as the statement that a 20% loss rebate ensures a player win, we all get 15 - 20% comps against our losses, if we end up straight losing. Some whales get 10% off their losing markers. So if you think about our little equation along THEM lines, every gambler would be a winner over all, if you consider the net cash and prizes, IF this theory were correct. Every gambler would figure, "Well if can negotiate 20% comps against my losses no matter what I'll always be better off gambling than paying straight cash for my trips."
But in any case, rebates against losses had nothing to do with the big wins Johnson had, these were more about getting up and leaving when he was ahead. There was no rebate involved in those huge wins because he didn't lose in those particular huge win sessions in the first place.
Quote: WizardI attended a talk he gave several years ago. He mainly spoke of the loss rebates, but he never denied other advantage plays were involved. From word-of-mouth, I hear he was also counting cards (but not so aggressively as to get caught) and inducing dealer errors in his favor. I'm not big on quoting "word of mouth," but that's all I've got.
If we assume he was counting cards. How would that help while flat betting? Can the edge shift in the players favour just by using Basic Strategy deviations? (Illustrious 18 with Surrender).
Is there a new method of counting that he figured out that tracks 5's and Ace's? - He has frequently said that he hired PhD Mathematics to help him with his play.
Quote: PinkJackIf we assume he was counting cards. How would that help while flat betting? Can the edge shift in the players favour just by using Basic Strategy deviations? (Illustrious 18 with Surrender).
Is there a new method of counting that he figured out that tracks 5's and Ace's? - He has frequently said that he hired PhD Mathematics to help him with his play.
Given that he was flat betting and counting cards to make strategy decisions, there is no way that he simply counts high cards vs low cards.
For a 16 vs 10 decision, the presence of 5s in the deck increases the EV of hitting. The presence of 6s in the deck increases the EV of standing. So, you use a count system that gives opposite values to 5 and 6.
For a 10 vs 10 decision (Hit or Double) the only thing that matters is whether or not the remaining cards in the shoe have a surplus of Aces. If the surplus of Aces is large enough in the remaining shoe, you Double. It has nothing to do with the abundance of 5s or Tens, or low cards vs high cards.
I have worked this all out for many of the close call decisions -there are many besides the illustrious 18, which were originally picked because they are sensitive to the abundance of high vs low cards. I can write an article about it if anyone is interested. It is powerful for single deck or double deck BJ, but less powerful for 6-8 decks.
Just as if you're getting 12s and 13s all night you'll lose no matter what the count is, if you're getting 20s or hitting to 21 all night you'll win no matter what the count too.
One night I was playing a single deck and almost every hand I was winning, either getting blackjacks, twenties or hitting to twenty one no matter how crappy my first two cards were. The house tried everything to break my streak, changing dealers, changing decks, shuffling after each deal, nothing worked I just kept winning up to a peak of just under $70K. I wasn't even betting that much around $800. per hand but I just couldn't lose that night. If Johnson had a run like that or even close to a run like that easy to see how he'd win millions in one session.
I've seen a BJ player go from his last five thousand dollar chip to practically draining the dealer's tray, or Baccarat players go from ten grand up to a million. But more often than not they keep playing, and lose it all. Ever heard of the no shoes bandit (speaking of the homeless)? Look it up. Walking after a winning streak is key. If everyone in Vegas stopped playing after losing half his winnings Vegas would go dark by the end of the year. Ask any pit boss, they've seen it all: "Why do people lose?" - "Because they won't quit while ahead."
That Canadian mattress maker in Theroux's documentary, he was ahead $50K at roulette right after arrival, but in an Owning Mahoney-esque statement, declared that he hadn't flown six hours just to play twenty minutes. Not hard to see how or why the mattress millionaire ended up selling off his companies to pay gambling debts, eventually fired as an officer of his former companies, and finally, a broke Uber driver. The house advantage isn't what did him in so quickly - it was his inability to stop when ahead, making it so that he never had any winning sessions.
You just can't undervalue the value of quitting while ahead.
Quote: MDawg
That Canadian mattress maker in Theroux's documentary, he was ahead $50K at roulette right after arrival, but in an Owning Mahoney-esque statement, declared that he hadn't flown six hours just to play twenty minutes. Not hard to see how or why the mattress millionaire ended up selling off his companies to pay gambling debts, eventually fired as an officer of his former companies, and finally, a broke Uber driver.
Not only his losing ruined his life, that documentary put acid on his wound. I also saw the online screen shot of him driving for Uber. I always wonder if he got more credit and over did it on that particular trip to save his face from the documentary.
Since most simulations take millions of hands into account. If one was to run multiple simulations of a small sample size (eg: 1500 hands) and then determine in what percentage of those does a player come out on top. We might have a better answer.
So to reiterate. Instead of running a single simulation with million hands. We run 1000-simulations each with 1500-hands, Would this give a different result?
Quote: OnceDearIt may be there were never losses made on which to give rebate. That's part of the counter intuitive aspects of loss rebate bonuses: You don't have to ever make a loss to benefit from them.
E.g. If you are offered a 100% loss rebate on a $1,000 wager on a coin toss and win the coin toss, then was the loss rebate valuable $:o)
Yes because otherwise I wouldn't have made the bet.
@PinkJack -- he had a loss rebate, no upgraded basic strategy, etc. Getting the math perfectly accurate can be difficult, but the logic should be simple to grasp: If you win, you keep that money...if you lose, you get some of it back....add the two together, and you're ahead.
The more frequently you're getting a tie around 17s and 18s with the dealer. The closer you are to X-true count. Any data around this?
Quote: PinkJackIs it possible to guess the true count based on how the cards are being dealt?
The more frequently you're getting a tie around 17s and 18s with the dealer. The closer you are to X-true count. Any data around this?
Guess?
Yes it's dead easy to guess. Ploppies do it all the time. We don't aspire to guess, right or wrong.
Quote: WizardI attended a talk he gave several years ago. He mainly spoke of the loss rebates, but he never denied other advantage plays were involved. From word-of-mouth, I hear he was also counting cards (but not so aggressively as to get caught) and inducing dealer errors in his favor. I'm not big on quoting "word of mouth," but that's all I've got.
Quote: MDawgAnother thing Johnson did is have other players jump in to eat up bad cards. I've actually done that quite successfully, I've had my friend sit at first base, and play small, with me the big player at second. When I felt that I needed it, I would have him take hits.
I also heard this - don't recall where, maybe a GWAE interview? He would sit out bad counts and had other players around him to eat the cards (and had a great act of being fumed/pissed off that he was on a losing streak or whatever). So even if he was flat betting, he effectively had a 1 - 0 spread.
I'd assumed that he was playing by himself. Hmmm,
Quote: PinkJackSo Basic Strategy + Illustrious 18 + 0-1 Betting Spread ; what would the house edge be with these factors?
I'd assumed that he was playing by himself. Hmmm,
PinkJack, you seem to ask the same question over and over. You don't seem to be listening to the answers, and you don't seem to understand the theory or underlying math of blackjack, or the realities of playing at a table.
If you are playing one-on one with a dealer at a $100,000 high stakes table for 12 hours, then there is no real meaning to a "0-1 Betting Spread." If you Wong out to avoid a negative count and talk on the phone/hit the restroom -then the same negative count is waiting for you when you return. So your question makes no sense.
The so-called "Illustrious 18" comes down to Insurance and 16v10 and nothing else really matters when there are 4-8 decks - especially if you are using a High-Low count that is intended for bet size variations rather than for achieving computer-perfect decisions.
I think a fair number of people, including the Wizard, have told you that that Don Johnson's profit while flat betting was not a matter of grinding out a mathematical edge over the house achieved by better Hit/Stand/Double decisions.
Quote: mcallister3200Almost certainly no one not involved with the play will ever have more than a broad estimate what his actual advantage was, there was admittedly other advantage plays involved that he would be betraying confidences by talking about. Billy and commish/teliot showed enough that the edge would be substantial without anything extra though. I really doubt Johnson was the brains behind this play and bet his tax returns would show a substantial chunk of those winnings went to others involved, he was the clean name.
He did say that there were PhD Mathematics involved. Hence I assumed that they coached him in some simple techniques which gave him an edge despite him flat betting.
Quote: gordonm888PinkJack, you seem to ask the same question over and over. You don't seem to be listening to the answers, and you don't seem to understand the theory or underlying math of blackjack, or the realities of playing at a table.
If you are playing one-on one with a dealer at a $100,000 high stakes table for 12 hours, then there is no real meaning to a "0-1 Betting Spread." If you Wong out to avoid a negative count and talk on the phone/hit the restroom -then the same negative count is waiting for you when you return. So your question makes no sense.
The so-called "Illustrious 18" comes down to Insurance and 16v10 and nothing else really matters when there are 4-8 decks - especially if you are using a High-Low count that is intended for bet size variations rather than for achieving computer-perfect decisions.
I think a fair number of people, including the Wizard, have told you that that Don Johnson's profit while flat betting was not a matter of grinding out a mathematical edge over the house achieved by better Hit/Stand/Double decisions.
Okay. I think we can close this discussion. Seems like enough answers have been given. Thanks guys! :)
Quote: MDawg
That Canadian mattress maker in Theroux's documentary, he was ahead $50K at roulette right after arrival, but in an Owning Mahoney-esque statement, declared that he hadn't flown six hours just to play twenty minutes. Not hard to see how or why the mattress millionaire ended up selling off his companies to pay gambling debts, eventually fired as an officer of his former companies, and finally, a broke Uber driver. The house advantage isn't what did him in so quickly - it was his inability to stop when ahead, making it so that he never had any winning sessions.
Has this actually been verified? I've seen that documentary several times, and heard that story about what happened afterwards, and seen that Uber screenshot. But I've also heard didn't go broke, wasn't fired, and just retired and sold the company for millions, and he's doing fine and just drives Uber on the side for fun.
I haven't seen any concrete facts about what actually happened to the guy, just various conflicting word of mouth stories people spread on the internet.
Quote: TigerWuHas this actually been verified? I've seen that documentary several times, and heard that story about what happened afterwards, and seen that Uber screenshot. But I've also heard didn't go broke, wasn't fired, and just retired and sold the company for millions, and he's doing fine and just drives Uber on the side for fun.
I haven't seen any concrete facts about what actually happened to the guy, just various conflicting word of mouth stories people spread on the internet.
Possibly.
Would anyone drive an Uber for fun? Must be better ways to get out of the house.
Is he about to join the Uber strike for better wages lol
