just interested in finding out how often a progression will bust (or how quick) on average....(again, no posts about pointing out how progressions dont give an edge....)
hands in blackjack win roughly 43.5%, lose 48% and push 8.5% of all hands. To figure out the odds of winning two hands in a row you just multiply 43.5% x 43.5% = 18.9% You can figure out the probablilty of any winning or losing streak this way. (P of 4 losses in a row (not counting pushes) is simply 0.48^4 or .48x.48x.48x.48 or 5.3%
a winning streak of 10 wont kill me, id need a few bad streaks to do that, or win about 1/3 of hands (not including pushes) consistently...
basically, i want to find out the % of times ill bust in the first hr, in the 2-5 hrs, in 10hrs etc...
Roger Harris wrote a progression simulator. Could be what you're looking for. Link to the old Progressions Zone below.
otherwise the speed of your play would be the major factor. When running low on funds just slow down play and run out the clock.
I have written software (in plain vanilla "C") which tests progressions.
Built into it are the probabilities of:
...losing 6 base bets (i.e., splitting out to 3 hands,
...doubling on all, and losing all);
...losing 5, 4, 3, 2, and 1 base bets; and
...winning 1, 1�, 2, 3, 4, 5 and 6 base bets
which probabilities were determined by a run of approx. 1.4 billion simulated hands (i.e., approximately 280 million rounds with 5 players at the table and the results for each player recorded), with the following rules: 6D, S17, DA2, DAS, RSA3, SAP3, NS, 75% pen.
The program assumes you have a "trailing" stop loss limit and a win goal. For example, say you are willing to risk 20 base bets and have a win goal of 20 base bets: the "trailing" stop loss means that if you reach a peak of 19� base bets up, the program will record a win if you get >= 20 base bets up and record a push or loss if you get down to 0 or -� base bets (i.e. your loss is limited to 20 base bets down from your PEAK). It would stop at 0 (a drop of 19� base bets from the peak) because you wouldn't have enough money left to make another bet the size of the base bet.
Usually I use a stop loss value considerably lower than the win goal (typically 7 base bet stop loss, 20 base bet win goal). A "win" and it's value are recorded whenever the goal is met or the stop loss is triggered when the $ are > 0; a push is recorded whenever the stop loss is triggered when the $ equals zero; and a loss is recorded whenever the stop loss is triggered and the $ are < 0.
There are three arrays, which in this case would number 58 elements each, with each element corresponding to a value ranging from -19 to +19� base bets in steps of half of one base bet. These arrays hold double precision numbers, and are named the "luz," "bet," and "win" arrays.
The program starts by putting the probability of 1.0 in the "bet" array element corresponding to a dollar value of zero. Then the loss probs are distributed to the appropriate "luz" array elements and the win probs distributed to the appropriate "win" array elements. Whenever a "bet" prob is distributed, it is zeroed out.
Whenever a $ total reaches or exceeds the win goal, the prob and the money are recorded via a special routine which provides the equivalent of a double precision variable with 114 bits of precision. Whenever the $ total drops to the point where the trailing loss limit is reached, the results are recorded similarly. This 114 bit limitation requires that when "bet" probs below a certain value (roughly 7 * 10^-23) are encountered they are considered too small to f**k with; so are merely moved from the "bet" array to the "luz" array where the vast majority of the time they will be added to an existing value there large enough to distribute the next iteration.
Since at the end of this process the "bet" array will be empty, the pointers to the "bet" and "win" arrays are swapped, the index to the progression array is incremented, and the probs in what is now the "bet" array are distributed to the "luz" and what is now the "win" arrays. This processing of runs of consecutive wins is repeated until there are no more probs left to distribute.
Once that happens, both the "bet" and "win" arrays are empty; so the pointers to the "luz" and "bet" arrays are swapped, and the whole process is repeated until the sum of the probs for win, lose and push is at least 0.9999.
Then these probs are rationalized to total 1.0, and the program reports the prob of win, lose and push, the average win, the average loss, the total action, the ROI, and a few other stats relating to volatility and a few other things.
Thus, the program starts with the initial bet, and takes it through EVERY possible sequence of events, and reports the AVERAGE performance of the progression.
If you want to email me your progression and provide me with your win and loss limits, I will be happy to run it through the program and report the results to you.
Also, I have another version of the program with has already tested every 6-step "press on win" progression from 2-1-1-1-1-1 to 2-6-10 (with no bet allowed to exceed 5 times the base bet) but without the stop loss feature. It is a straight test of either win or lose 20 base bets, and the % of time it wins. FYI, using these criteria, the worst progression is 2-1-3-5-10 (wins 30.8337%) and the best progression is 2-6-6-7 (wins 35.8737%). What is obvious from these results is that no press-on-win progression can win; and the faster you get the money out the better chance you have to win.
Ergo, if you have an "average" progression with, say, a 1/3 chance of winning, you have a 2/3rds chance of busting out; and only 1 chance in 9 of getting two winning sessions in a row.
In addition to that, I also wrote a program which put "The Pro's 95% Solution" to the test. That system appears to be the blackjack adaptation of a craps scheme I once evaluated which uses a Fibonacci series where you always "follow the shooter" and go up the series on a loss, and down on a win so that two wins in a row yields a win amount greater than the sum of all previous losses. E.g., lose $5, lose $10, lose $15, lose $25, win $40, win $25 -- net +$10. Like all such elaborated and convoluted "martingale"-type systems, you will *probably* win any given start, but when the inevitable catostrophic loss occurs it will more than wipe out the total of all previous wins.
Again, I would be more than happy to test your progression and give you the results. If it has any merit, I would like to learn about it; and if I think it can be improved, I can make some recommendations.
Which progression?
If it looks promising, I can evaluate it for you.
I would need the progression, and the win and loss limits.
If you don't feel like posting it here, feel free to email me.
i have had continued success by using variations of the D'Alembert progression...
right now i use a 200unit BR....when a double or split is made, drop back 2 units instead of of one....if its after bet 6, then i drop back 2 on a BJ as well....if its after bet 10, i will normally drop back 2 units on any win
the problem with this is the way i drop back 2 units after bet 10....if it keep hovering around this point then i will slowly lose money...if i cut my losses and drop to a lower bet then ive done just that, cut my losses....this is all in favor of making sure i dont reach the higher teens bets...(or the lower teens for the matter)
normally in a long session, i wont go past bet 10-12....i have normally made a buffer before this sort of thing happens anyway...
basically i was wondering how things would turn out on average say for about 5000 hands...(not that i go anywhere near that, just want to see longterm losses)
so no stop for a win, but stop point for loss is 200 units....
i dont expect it to produce a positive result, im just interested in seeing how it does...
ive played alot over the last year (enough to justify longterm), but am still very far ahead...ive survived many bad runs, and have only had a few good runs (i can remember every single one of them)...
my losses arent much larger than my wins (often smaller), and like i said, ive made alot of money of it...
im into bonus hustling too, and having been counting since the beginning, so its not that im not an advantage player, ive just found this has worked so far...
again, i dont expect it to produce an edge, i know it wont, but im interested in finding out how it goes....
I start enough already...
"2. Players will be awarded the 100% match bonus on their first purchase, by purchasing and wagering the purchased amount (up to the maximum bonus amount) once in full."
By this I assume you must wager your deposit (once) before they will credit you with the bonus....(ie to get the $100 bonus, wager your $100 in full, then you'll receive). I assume this is in part to make up for the low wagering requirement (5 times dep, then 5 times bonus before cashout).
Just like to be sure, so I don't cheat myself out of money...
you only give them the once thru, because any extra action MAY not count towards your total play requirements.
Don't expect anyone to run a SIM for you. It can be rather difficult to create a sim for a specific set of conditions in the first place. Normally you will get "the inevitable will happen" which is absolutely the truth, without the SIM results to go with it.
What I suggest you do is compare flat betting the minimum to the outcome using your system with all of your own perameters in place and find your own results. It would be much nicer to press a button and have it done for you a million hands per minute, that is if you tell the program the proper process in the first place or you will get GIGO.
I know it's been proven a progression won't produce an edge in favour of the player.
HOWEVER, I have clocked in a massive amount of time using this system. I started playing BJ for $$ a bit over a year ago, and have been using this system for a year. I have had a high win rate, and my average loss has been about the same as my average win (which bascially means I've made a lot).
I haven't been particuarly lucky, as I've had more terrible runs (the runs that can kill my BR for the night) than those 'invincible' runs (well they have the edge).
I have played about 4 times a week every week for nearly a year now, and have done countless tests (10's of thousands of hands) on BJ sims/games.
I have made more in my first year than what most counters will make in their peak. That said, I do know how to count. It just required a BR I didn't have at the time, and have since found my progression works better for me.
I also bonus hustle, so I am an advantage player. I just happen to use a progression because it's working at the moment, not because I believe it gives an edge.
Why does it work? That's what I'd like to know. Is it because I set modest targets that I will reach sometime most sessions? Is it the fact that it's exremely hard for me to get into the last third of the bet ramp (never been there, bets 14-20)? It's not because I've been lucky. Quitting on a high? Well we know it's just one big session.
When you say it's one big session you assume it's impossible to dodge the statistics and your living on borrowed time. You all say that qutting ahead won't help. Your saying my 'luck' will balance out. If I keep winning I'll eventually have a large loss to more than make up for it. Not so. I have both a win rate and small losses (same size as an average win).
Under average conditions (no bad or extreme luck), a negative progression is extremely productive for obvious reasons. As long as that progression can cope with anything but a lengthy bad run then it should come out ahead.
Basically what I'm saying is that you can state that it won't produce an edge, and I would agree with you. However you can't state that I will eventually lose it all or that it won't make me money. If I do you can say 'I told you so'. Until then, I'll be happy making an unlimited amount of money enjoying my expensive taste in cars.
BTW the 'you'is not directed at Rob, but to the BJ community as a whole. Just because progressions are mathematically flawed, and haven't worked for most people doesn't mean that they don't or won't work for ALL. Hey just look at the Porsche 911. Fundamentally a very flawed design (takes getting used to), but look at it's success both in racing, sales and popularity (well at speed and looks and anything is popular, but it still has a great racing history).
I guess I'm the young porsche of blackjack. I use a mathematically flawed way to make money (among other ways), and have ironically been more successful in my first year than all but the top counters.
and those 5*dep& bonus requirements are worth it
"However you can't state that I will eventually lose it all or that it won't make me money."
Yes I can, and yes, I can.
"If I do you can say 'I told you so'."
I will.
"Until then, I'll be happy making an unlimited amount of money enjoying my expensive taste in cars."
No, you won't.
"Just because progressions are mathematically flawed, and haven't worked for most people doesn't mean that they don't or won't work for ALL."
Let me clarify...you will face certain ruin.
"Hey just look at the Porsche 911. Fundamentally a very flawed design (takes getting used to), but look at it's success both in racing, sales and popularity (well at speed and looks and anything is popular, but it still has a great racing history)."
Please explain this "fundamentally flawed design" of the 911, that can be "gotten used to".
"I guess I'm the young porsche of blackjack."
I'd say more like a scooter, or a moped maybe.
"I use a mathematically flawed way to make money (among other ways), and have ironically been more successful in my first year than all but the top counters."
J? J? wake up, Its mommy, time for school!
I've tried to be helpful in the past without criticising your play, but...
I know it's been proven a progression won't produce an edge in favour of the player.
That's right - it won't effect the edge at all, which is in the house's favour. You say you "know" this but it obviously hasn't sunk in. You have contradicted this fact several times in your post.
HOWEVER, I have clocked in a massive amount of time using this system... I have played about 4 times a week every week for nearly a year now, and have done countless tests (10's of thousands of hands) on BJ sims/games.
10,000 hands is but a drop in the ocean. The house edge your system would need to overcome is less than 1%. Therefor you need to determine the effectiveness of your system to within a fraction of 1%. AT LEAST 100,000,000 hands would be necessary to do this. You will not live to play this many hands manually. You have played only 0.01% of this amount.
I haven't been particuarly lucky
You have been on the right side of the bell curve. This bell curve is centered about a point less than zero. As you play more and more this bell curve will narrow, and it will be less and less likely for you to stay above zero.
I have made more in my first year than what most counters will make in their peak... Just because progressions are mathematically flawed, and haven't worked for most people doesn't mean that they don't or won't work for ALL.
You are but a drop in the ocean (a wealthy drop, admittedly). If 10,000 people played your system correctly for a year and 10,000 people played a counting system correctly for a year, which group do you think would have more winning players? Which group would you rather be in?
I also bonus hustle, so I am an advantage player.
I sing in the shower, does that make me Frank Sinatra? You won't be an advantage player till you get over this progression thing.
Is it because I set modest targets that I will reach sometime most sessions? No.
Is it the fact that it's exremely hard for me to get into the last third of the bet ramp (never been there, bets 14-20)? No.
It's not because I've been lucky. Beg to differ.
you can't state that I will eventually lose it all or that it won't make me money.
It won't make you money in the long run and you will eventually lose it all. :-P Eventually and the long run can be a very long time, perhaps longer than you're playing career if you're very lucky.
Now you are preparing to write a heated response saying you have the money to prove your system works. Please try to include at least some sound arguments - I spent a fair amount of time writing this in an honest attempt to help you.
Best of luck.
"Mathematical Proof that Progressions cannot overcome Expectation"
This also means:
"Mathematical Proof that Progressions Will Not Change the Expectation (+/-) of a Game of Chance"
Longer, but more readable, and from the man himself Dr. Edward O. Thorp.
I am on the nicer side of the bell curve, and there is a possibility I will continue to stay on it (however remote).
This tells me that what you say is probable, but not absolute.
This progression fettish I have is probally only due to early success with it, but as long as I keep making money I will use it. When or if I start to lose steadily I will leave it alone.
911 as a flawed design? You'ver got to be kidding me, right?
Having the engine hung out over the rear axle, and having such a short wheelbase is not optimal. If most drivers were asked to push the 911 to their limits they would crash. Asked to do so in a vette, I'm sure they would manage a bit better.
The 911 takes a long time to MASTER, due to its sensitivity. The only advantage the 911 has is its qucker turn (short wheelbase) and the way it brakes better (better weight transfer). However these advantages are also present in a car like the 360, which will also be more forgiving, and will handle better.
I love porsches but I can't ignore the layout of the 911. Porsche can, for reasons I won't go into here.
By saying I'm the 'young Porsche of BJ', I was refering to my success and survival using a technique which is flawed. As long as success continues, I won't stop my use of progressions.
