Another does this work thread?

[bpastermack]

Best comment I've ever seen about InPlay. Just ignore the guy.

Speaking of which... is there a "foe" feature on these boards? (Not to hijack the thread or anything...)

I do find it kind of funny that a guy that just wants everybody to shut up and play is on a forum that is designed to discus blackjack theory??

 

[shadroch]

I believe his point is that nothing anyone says will convince you this won't work. The only way you'll be convinced is to blow your own money trying it.
Or you can accept the fact that after several hundred years of people trying to find a progression that works,the chance that you've manged to do what no trained mathalete has ever been able to do is nil.

 

[InPlay]

I believe his point is that nothing anyone says will convince you this won't work. The only way you'll be convinced is to blow your own money trying it.
Or you can accept the fact that after several hundred years of people trying to find a progression that works,the chance that you've manged to do what no trained mathalete has ever been able to do is nil.

Bravo! Nothing like getting hit in the head with a sledge hammer to prove a point, one way or the other! Reality is real and sometimes Painful!

 

[bpastermack]

Bravo! Nothing like getting hit in the head with a sledge hammer to prove a point, one way or the other! Reality is real and sometimes Painful!

I understand how some really close minded people have probably walked in here and pissed you off in some way, but at what point out of anything I have written did you get the idea that I wasn't very open to the possibility that I just happened to get really lucky?

I'm a math guy. I understand how probability works, but I'm new to blackjack and I do know that there are a lot of nuances to the game. I came here more to disprove my theory than to prove it.

I have plenty of balls to do it, but the brains to go with it. I'm not a gambler. I usually play poker because I know the odds are with me, being a game of skill.

 

[daddybo]

I understand how some really close minded people have probably walked in here and pissed you off in some way, but at what point out of anything I have written did you get the idea that I wasn't very open to the possibility that I just happened to get really lucky?

I'm a math guy. I understand how probability works, but I'm new to blackjack and I do know that there are a lot of nuances to the game. I came here more to disprove my theory than to prove it.

I have plenty of balls to do it, but the brains to go with it. I'm not a gambler. I usually play poker because I know the odds are with me, being a game of skill.

I don't know InPlay, but here's how I see it ...

I think everybody takes InPlay the wrong way... He has knack for dry humor and is totally opposed to tact. I believe he would probably call a Spade a Spade. :laugh:

 

[duanedibley]

I'm a math guy. I understand how probability works

Every bet you make has a negative expected value due to the house edge. Since you are not counting every bet is also essentially independent. Therefore any series of bets you make will have a negative expected value by the linearity of expectation.

By doubling your bets you are giving yourself a >50% chance to win a small amount over a small number of hands, but not enough to offset the <50% chance of a large loss if you lose (in terms of your expectation).

Essentially what you are trying to do is artificially create a winning trend by pushing your inevitable loss over the horizon, and in the process you are making that inevitable loss very large. If you only play a small number of hands then you might win a small amount, and this might be justifiable if you never play again (with this strategy) in your life.

If you play more than a small number of hands, then you will most likely lose, since you will lose consecutively until you can't double your bet anymore. This will almost always happen before the small bets you previously won add up to more that the max bet you just lost, and this trend will continue if you start again from a min bet.

And unfortunately it will not be long before you lose the required N hands in a row to exceed the max-bet were you to double again. For example if the min bet is 10 and the max is 1000, then you are expected to lose 7 in a row (ignoring any pushes in between) in less than 100 hands.

 

[Sonny]

I understand how probability works, but I'm new to blackjack and I do know that there are a lot of nuances to the game.

The first nuance you have to learn is called the gambler's fallacy. Once you understand that, you will be able to see the weakness in these types of systems.

-Sonny-

 

[InPlay]

If you don't think that you can lose a lot of hands in a row you don't know jack. I just came from the MGM casino. I reallly just took the wife for dinner no serious gambling. After dinner, I decide to see if I can pick up a hundred or so, chump change. I sit down at a $10 dollar table played for about 1/2 hour and at one point I lost 8 straight hands. So that was just in a 1/2 hour of play. Like I said before go for it and tell us what you do.

 

[Kasi]

I can fully see why the Martingale system would not work since if you ever lost 12 hands in a row like I have several time, then you would lose everything and then some. With my system though, you would have lost $1,000. Then you you go on a string of not losing 5 in a row again after that, then you make $1,500 and you're ahead.

I tend to agree that losing one hand has no indication on the next hand, although from what i have read losing streaks actually have a higher frequency than winning streaks, which makes sense.

I've read some threads that seem to indicate that there is no predictability to BJ, which is a joke, because nothing is more predictable in the long run. I can see why this betting system would have no value in any game I have found except in BJ. I think this has something to do with the BJ and double down opportunities that will present themselves while more chips are on the table. Does anyone know a good BJ simulator that I can actually place all of those criteria in and see what it comes up with, just out of curiosities sake? It would probably save me a lot of time of continuing to play this game.

Well, I'm in your side lol.

If you had a simulator, you would have to define such things as when you stop betting. Like BJ with Martingdale, if your goal is 1 unit, and you only are a half-unit ahead, do you continue the betting system? Or, worse, you split to 4 hands and double each, and lose 8 units in 1 loss in a row.

If you have $15 left, do you continue to bet when you will not be able to diouble if the hand calls for it or will you play the hand without doubling it becasue you have insufficient funds?

Along with starting roll.

In a "system", all rules must be defined exactly, ideally anyway in my world lol. If you had a sim all it's going to know is the rules you tell it. In a sim, you can't add to your roll when you feel like it kindathing. You either achieve the goal of the "system" or you lose your roll.

If you started with $450 and had 30 units betting $15, great. Winning $150 and increasing unit to $20 would maybe make no difference since that 30 unit roll would have the same chances of success as the $450 30 unit roll.

I'm assuming that you do understand, if you live long enough, you will eventually lose all but are more interested in the question, how many times will I achieve "defined goal" of system vs losing all initial roll and how long will each take? However, it does seem possible you actually may not understand that eventually you will lose all, even in BJ.lol.

Personally, I hate Martingdale applied to BJ as it is not an even-money pay-off and difficult to define goal and "rules of system".

But, no, I don't know of any sims that can do this. Maybe CVDATA using some "cover betting" system? Or write your own computer code to deal with it. I'm sure it would be easier applying to roulette or craps.

Now, while any betting "system" would assume a fixed unit roll, I always go back to the most basic voodoo system of all, say it's true if I have $12K roll and I know I will win at least one unit in the next 50 or 60 hands with a 12 unit roll 90% of the time while flat-betting in a -EV game. The Wise One has the particulars on this somewhere.

I flat-bet $1k and have a 90% chance of winning $1k in the next 50 hands.
Say I do and I can switch back to a $1 unit. Essentially, I now have a 10% chance of losing all my money over the next 3MM hands or so since I will either lose it all in the first 50 or 60 hands or I won't.

I will still lose it all eventually but, if I am willing to take a 10% chance of losing $12K in the next 50 hands vs the 10% chance of playing the next 3MM hands before losing all, then that's the risk I choose to accept to my roll over the next 3MM hands.

My lifetime ROR is still 100% but I have a 90% chance of never losing all, assuming I will actually live that long to realize it.

Here, in voodoo world, all you will ever get from anyone is the mindless mantra your "lifetime ROR" (assume infinite lifetime) "is 100%". Get used to it. You will never, ever, be told how long you must live to have an x% chance of it actually happening.

They will never take into consideration that, as mere mortals, our "BJ lifetimes" are limited in duration by the number of rounds we will play. They will never consider our unit roll, that we may choose to switch $unit size, how old we are and how long we will be playing or the goal we choose to achieve in that time with whatever risk.

Forgive them, as Someone's last words were, for they know not what they do.

 

[sagefr0g]

...
Now, while any betting "system" would assume a fixed unit roll, I always go back to the most basic voodoo system of all, say it's true if I have $12K roll and I know I will win at least one unit in the next 50 or 60 hands with a 12 unit roll 90% of the time while flat-betting in a -EV game. The Wise One has the particulars on this somewhere.

....

yes, i ran a sim with sageblackjack software (no relation, lol). had to use a macro and save these crazy text files to get the data. then used QFIT's calculators for the same game to figure probabilities of success for gaining a unit.
what ever here is a graph of part of it.
the other data is around here somewhere as well. the data shows essentially to make that unit you have success like a high percentage of the time (circa 85%-90% of the time) then that 15% or so of the time that you don't make the unit you lose some huge amount, like your entire roll, sort of thing.

interesting to me how the probability of making a unit starts out relatively low for one hand, then rises rather dramatically as the number of hands increases up to about 30 hands or so, then tends to flatten out after 50 hands or so.

to me it's the rate of increase that's interesting.

i'm not sure what to make of it.:whip:

 

[kewljason]

Or, worse, you split to 4 hands and double each, and lose 8 units in 1 loss in a row.

If you were to play the martingdale at its truest form, you would not split or double these hands, which means you would be giving up even more ev. This would push you even further away from the 50-50 senario that martingdale was designed for, making it an even worse choice for blackjack.

 

[Kasi]

yes, i ran a sim with sageblackjack software (no relation, lol). had to use a macro and save these crazy text files to get the data. then used QFIT's calculators for the same game to figure probabilities of success for gaining a unit.
what ever here is a graph of part of it.
the other data is around here somewhere as well. the data shows essentially to make that unit you have success like a high percentage of the time (circa 85%-90% of the time) then that 15% or so of the time that you don't make the unit you lose some huge amount, like your entire roll, sort of thing.

interesting to me how the probability of making a unit starts out relatively low for one hand, then rises rather dramatically as the number of hands increases up to about 30 hands or so, then tends to flatten out after 50 hands or so.

to me it's the rate of increase that's interesting.

i'm not sure what to make of it.:whip:

That is the graph I was thinking of. Didn't realize it came from sageblackjack. Or sageblackjack with QFIT's stuff.

I don't know EXACTLY what to make of it either but I assume it probably may mean making "at least one unit or more" if starting with 12 units. Like if you lose the first hand and are down 1 unit and then get a BJ so you are now up 0.5 units at that point, you keep playing.since your orig 12 unit roll has not become 13 units yet.

Don't understand your vagueness of 85-90% of the time etc. It clearly shows the % of achieving 13 units with a 12 unit orig roll over some fixed number of hands. The chances of achieving a 1 unit win before losing 12 unit roll do go up the more willing one is to play more hands to achieve that 1 unit goal.

It says little about what to do after x hands when I have neither won that unit or lost my orig roll.

 

[Kasi]

If you were to play the martingdale at its truest form, you would not split or double these hands, which means you would be giving up even more ev. This would push you even further away from the 50-50 senario that martingdale was designed for, making it an even worse choice for blackjack.

Exactly.

I suppose it may mean "Martingdale in it's purest form" i.e. keep the rules of Martingdale the same, would also imply refusing the extra 0.5 unit with a BJ on your first hand.

Or, it would mean changing the rules of what traditionally defines a "Martingdale system" in which case it is not a "Martingdale" system to me anymore. It's some other betting system.

Either way, it's why I said "Personally, I hate Martingdale applied to BJ as it is not an even-money pay-off and difficult to define goal and "rules of system".

 

[bpastermack]

Exactly.

I suppose it may mean "Martingdale in it's purest form" i.e. keep the rules of Martingdale the same, would also imply refusing the extra 0.5 unit with a BJ on your first hand.

Or, it would mean changing the rules of what traditionally defines a "Martingdale system" in which case it is not a "Martingdale" system to me anymore. It's some other betting system.

Either way, it's why I said "Personally, I hate Martingdale applied to BJ as it is not an even-money pay-off and difficult to define goal and "rules of system".

With my system I assume getting blackjack or double down wins or losses are bonuses. If I double down and lose I just increase my bet as if I lost one unit instead of two. Then if I double down and win I start over and pocket the extra units. In the long run i just assume the double down and splits will put me way ahead despite any short term set backs. If I lose four in a row and then double down win on my fifth hand, then I just won 17 units total. I wonder what would happen if you increased your bets a little every time until you get a BJ, split, or double that wins if you would do well with that. It would have to be regardless of wins and losses, only looking for the payoffs of the odds being in your favor. But I'm pretty sure you couldn't double the bets each time, cause it could easily take 15 hands some times. Anyone know the odds on getting one of those three on each hand?

 

[bpastermack]

So I'm trying to read that chart. Is it saying that if I play out 100 hands, then the chances of me being up 1 unit is like 92%? According to the odds even money is at about 91.77777% of the time.

 

[sagefr0g]

That is the graph I was thinking of. Didn't realize it came from sageblackjack. Or sageblackjack with QFIT's stuff.

well, ok the graph came from running QFIT's calculator 'probability of reaching a goal given a time constraint' for the game and method of playing titled on the graph.
then i ran a sim of that game using sageblackjack for a player just trying to make one unit. so the sim was ran in such a way so as once the player made one unit the data was collected and the sim was restarted. did this about for about 3,000 trials, which was about 71,317 rounds.
then i put the data in excel as depicted in the image below. calculated the number and percentage of successful trials making one unit and the number of unsuccessful trials trying to make one unit.
like for those sims it was like you had 25 units and was trying to get to 26 units. different number of units than 12 to 13 units, but i think the principle is the same.

I don't know EXACTLY what to make of it either but I assume it probably may mean making "at least one unit or more" if starting with 12 units. Like if you lose the first hand and are down 1 unit and then get a BJ so you are now up 0.5 units at that point, you keep playing.since your orig 12 unit roll has not become 13 units yet.

right, it's just to me interesting how the slope of that graph changes as you approach thirty hands or so and then it flattens out after about fifty hands or so. that's some sort of rate change relative to probability of success vs hands played. that rate of probability of success seems to increase at a dramatic rate for the first thirty hands or so.
i dunno for some reason that catch's my attention.:whip:
kind of the voodoo way i interpret that, is well if your trying to make just one unit flat betting with just basic strategy and you don't succeed by fifty hands or so, well maybe you might want to try something else as far as methodology.:whip:

Don't understand your vagueness of 85-90% of the time etc. It clearly shows the % of achieving 13 units with a 12 unit orig roll over some fixed number of hands. The chances of achieving a 1 unit win before losing 12 unit roll do go up the more willing one is to play more hands to achieve that 1 unit goal.

lol, probably i'm confused more than vague.
just, those trials i ran in the image below, as i remember from looking at the data, well, ok first off, yeah circa 95% of the time you succeed in make one unit, but circa 5% of the time you never made one unit and in fact lost your starting bankroll. so if 159 times out of 3000 you lost your starting roll you lost $159,000 but only gained circa $40 2,841 times or $113,640. so overall you end up losing circa $45,360 . (note: i'm here, just ignoring the fact that at times trying to make just one unit, one might have made two or more units)
i mean, well, those numbers are probably off by a lot, because i'm just sort of estimating how it goes from the image below, but it's pretty much what the sim and trials showed in general.
what ever, i also think i tried to look at that data and see what would happen if you just limited each session attempt for making a unit a thirty hand attempt and then quit, then try again, sort of thing. i don't think i simmed that sort of thing specifically but just tried to 'fool' with the data and guesstimate what would happen, lol.
but anyway, the thing is, yeah the more your willing to play the probability of making the unit increases, but to me so does the chance of hitting those big losing streaks. :whip:

It says little about what to do after x hands when I have neither won that unit or lost my orig roll.

that's where i wonder about that slope of the graph and the apparent rate of change of increase in probability of achieving one unit around the thirty hand point. i mean i wonder if that has any significance as far as what the wise number of x hands might be if one is trying to make just one unit, sort of thing.
i mean, the way it looks to me, for a session where your trying to make just one unit, that ok, the first thirty hands or so should represent going through a phase where your probability of success rises dramatically as you continue to play hands, but then after about fifty hands or so your in a phase where yeah, your probability of success is relatively high but it doesn't increase as dramatically the more you play. :whip:

 

[Kasi]

well, ok the graph came from running QFIT's calculator 'probability of reaching a goal given a time constraint' for the game and method of playing titled on the graph.
then i ran a sim of that game using sageblackjack for a player just trying to make one unit. so the sim was ran in such a way so as once the player made one unit the data was collected and the sim was restarted. did this about for about 3,000 trials, which was about 71,317 rounds.
then i put the data in excel as depicted in the image below. calculated the number and percentage of successful trials making one unit and the number of unsuccessful trials trying to make one unit.
like for those sims it was like you had 25 units and was trying to get to 26 units. different number of units than 12 to 13 units, but i think the principle is the same.

right, it's just to me interesting how the slope of that graph changes as you approach thirty hands or so and then it flattens out after about fifty hands or so. that's some sort of rate change relative to probability of success vs hands played. that rate of probability of success seems to increase at a dramatic rate for the first thirty hands or so.
i dunno for some reason that catch's my attention.:whip:
kind of the voodoo way i interpret that, is well if your trying to make just one unit flat betting with just basic strategy and you don't succeed by fifty hands or so, well maybe you might want to try something else as far as methodology.:whip:

lol, probably i'm confused more than vague.
just, those trials i ran in the image below, as i remember from looking at the data, well, ok first off, yeah circa 95% of the time you succeed in make one unit, but circa 5% of the time you never made one unit and in fact lost your starting bankroll. so if 159 times out of 3000 you lost your starting roll you lost $159,000 but only gained circa $40 2,841 times or $113,640. so overall you end up losing circa $45,360 . (note: i'm here, just ignoring the fact that at times trying to make just one unit, one might have made two or more units)
i mean, well, those numbers are probably off by a lot, because i'm just sort of estimating how it goes from the image below, but it's pretty much what the sim and trials showed in general.
what ever, i also think i tried to look at that data and see what would happen if you just limited each session attempt for making a unit a thirty hand attempt and then quit, then try again, sort of thing. i don't think i simmed that sort of thing specifically but just tried to 'fool' with the data and guesstimate what would happen, lol.
but anyway, the thing is, yeah the more your willing to play the probability of making the unit increases, but to me so does the chance of hitting those big losing streaks. :whip:

that's where i wonder about that slope of the graph and the apparent rate of change of increase in probability of achieving one unit around the thirty hand point. i mean i wonder if that has any significance as far as what the wise number of x hands might be if one is trying to make just one unit, sort of thing.
i mean, the way it looks to me, for a session where your trying to make just one unit, that ok, the first thirty hands or so should represent going through a phase where your probability of success rises dramatically as you continue to play hands, but then after about fifty hands or so your in a phase where yeah, your probability of success is relatively high but it doesn't increase as dramatically the more you play. :whip:

See, there u go. I knew I'd (I think) learn something thanks to you.

Now, it looks like the graph is the chances of winning a unit but says nothing about the chances of losing all lol. Like I think you were trying to tell me but I couldn't get thru my thick head lol.

In other words, say at 30 hands, you have a 88% (whatever it was) chance of winning 1 unit. Now I think it means that the other 12% of it only means you will have failed to win that 1 unit at that point, NOT that the other 12% of the time one would lose all 12 units as I was thinking. You may still be even, down 5 or 10 units or even maybe have actually lost all 12.

Guess it's a voodoo variation of "if they don't get you early, they may not get you at all" lol.

 

[sagefr0g]

See, there u go. I knew I'd (I think) learn something thanks to you.

Now, it looks like the graph is the chances of winning a unit but says nothing about the chances of losing all lol. Like I think you were trying to tell me but I couldn't get thru my thick head lol.

In other words, say at 30 hands, you have a 88% (whatever it was) chance of winning 1 unit. Now I think it means that the other 12% of it only means you will have failed to win that 1 unit at that point, NOT that the other 12% of the time one would lose all 12 units as I was thinking. You may still be even, down 5 or 10 units or even maybe have actually lost all 12.

Guess it's a voodoo variation of "if they don't get you early, they may not get you at all" lol.

some where i thought i had some images of qfit's calculator stuff on ror for this stuff.
can't find it.
but here is from your sheet for that game for circa 100 hands.

edit: but it would be interesting to work out using qfit's calculator, ror by number of hands and graph, similar to that other graph for probability of making a unit by hand, sort of thing.
another edit: ok below is an image for a similar eight deck game risk of ruin going for a one unit goal, $1000 trip roll, flat betting $40 trying to make one unit goal.

 

[Kasi]

some where i thought i had some images of qfit's calculator stuff on ror for this stuff.
can't find it.
but here is from your sheet for that game for circa 100 hands.

edit: but it would be interesting to work out using qfit's calculator, ror by number of hands and graph, similar to that other graph for probability of making a unit by hand, sort of thing.

Well, my sheet is for reaching a goal given no time limit - one plays forever until either goal achieved or goes bust.

You specified reaching a goal given a time constraint (I think) - one plays a fixed amount of time and either achieves goal or doesn't in that amount of time.

I think my sheet will agree with QFIT's calculator if you use the correct calculator.

I have never been able to do the "goal with time constraint thing" thing like it appears you may have used, basically the "double-barrier" formulae on pp 136 and 137. Summing an infinite series is stuff I don't know alot about. Unfortunately Don deemed "it is beyond the scope of his book to go into the math of that". So there was nothing for me to mimic. Reallistically speaking, I'm OK with adding, subtracting, multiplying and dividing and that's about it. No, that IS it. Period.

I struggle with this cr*p as much as anybody and am not ashamed to admit it.

I think my trip ROR stuff is OK and agrees with a QFIT calculator though.

 

[sagefr0g]

Well, my sheet is for reaching a goal given no time limit - one plays forever until either goal achieved or goes bust.

You specified reaching a goal given a time constraint (I think) - one plays a fixed amount of time and either achieves goal or doesn't in that amount of time.

I think my sheet will agree with QFIT's calculator if you use the correct calculator.

I have never been able to do the "goal with time constraint thing" thing like it appears you may have used, basically the "double-barrier" formulae on pp 136 and 137. Summing an infinite series is stuff I don't know alot about. Unfortunately Don deemed "it is beyond the scope of his book to go into the math of that". So there was nothing for me to mimic. Reallistically speaking, I'm OK with adding, subtracting, multiplying and dividing and that's about it. No, that IS it. Period.

I struggle with this cr*p as much as anybody and am not ashamed to admit it.

I think my trip ROR stuff is OK and agrees with a QFIT calculator though.

right, right your sheet and QFIT's calculator agree given the correct calculator is used. like you say i was using a different calculator for a different purpose. (also with some slightly different numbers as well, lol )
but yeah given the right numbers and right calculator there is agreement as shown below...