Roulette Martingale alternative (Carsch)

@thereddiamanthe this might help you to practically compute your probabilties and waiting times for a complex betting strategy:

https://www.cantorsparadise.com/magic-tricks-with-markov-chains-96a4465c20a

That article is using the R language which you can type in and play with using https://replit.com/languages/rlang

You don't need to understand the language but just use the snippets provided and plug in your own numbers.

If you want to go further then I would suggest look at the free Google Colab notebooks which stores everything in your Google drive.

 

I notice that you, Median Joe and Thereddimanthe are looking into the possibiliy of a betting strategy based on pattern wait times. The articles you post back and forth in regards to all of this I do not understand the math behind it but I do understand the theory somewhat. I am working with a person who is a researcher and tester of systems and methods. He has been coding my basic strategy with a few of his ideas thrown in for 100,000 spins at a time. He tests only with flat betting and the results so far are very promising. I am telling you this because I have a suggestion to make that maybe you should include in your research also. My basic strategy is that outcomes repeat readily and XXX, OXO and OXXO will always form in a given length of a series. How often and when is the question that needs to be answered. I have been lucky enough to be successful with this approach and I am sure with the correct math applications it could be much improved. I would bet you REAL money that no matter what you come up with, it will not be successful if you do not use a negative betting progression. The reason for that, all of the math probabilities of what is expected will not hold true in many instances. A negative progression will help with the discrepancies. And without it, it will never work. Of course, all of this is based on my limited observations.

 

Find a way to win and then all this MM controversy becomes a bit of a moot point.

If you believe the following...

''In the long run, any progression is simply a set of many different levels of flat bets.

So, if flat betting is long term negative, then progressions must be, too.''

Then the only logical answer is to come up with a bet selection that is positive.

As others have already said, if you have no positive bet selection, then you really are better just flat betting.

@David Gregory, you alluded to the fact that you may be already using these patterns with lower waiting times and so maybe that's why what you are doing is working well. I don't know but maybe you are using them in a way what works.

 

Progressions get killed by losing streaks. That alone should tell you about looking for a better strategy. But even with applied targeting you will still encounter a sequence of death from aggregated unique starting points. With a mathematical approach with a three or four step marti you might win enough to over power the rare sequence of death. That is where RR and Martingale together has yet to be tested or bragged about. I know for a fact that on rare occasions I lose an EC battle to a swarm of first try bets that are 7 in a row even though I waited for new starting points for each bet. In the case of a Marti that would be each step. Targeting when for each step could work. Also only three or four steps of the progression might be the smartest move. Test that if you want to show you can learn.

 

Yes, that could be true. It may be that I stumbled on this just through lots of trial and observation. It's not perfect but I think there is room for improvement with mathematical probabilies applied.

 

David, have you tried using any positive progressions like the reverse Martingale or reverse Labby? If your bet selection tends to result in wins clumping together it might be worth a try. You could start off with a reverse marty of 3 steps and see how that goes.

 

Yes I did try that many times but just could not make it work. It seems everytime you double on a win once or twice you lose the next outcome and are back to square one. I would give anything if I could make a possitive progression work. The best reason for that is I wouldn't be constantly beat to death by those who loathe negative progressions.

 

The problem with the long term view, is that it
is not what you actually get, it is what the casino gets across all players.

The variance we face on our particular random walk in a session will cut deeper than the house edge.

But comparing flat vs progressions should not be done niavely. Yes betting more pays more house edge but so does betting longer.

With flat betting you need to make more bets than with a progression. 10 losses needs 10 wins to recover, and that will typically take far more than 10 bets to dig the hole and more than 10 to get out of it , perhaps 15 or even 30 bets on each side so 30-60 bets.

e.g. lose 2, win 1, lose 1, win 2, lose 2, win 1, lose 4, win 1 etc which is a total of 15 bets to get -5 unit loss and still more bets to go..

A fixed marty made 3 bets with 7u of total risk to make +1 unit profit.

House edge is paid per unit bet, so who's paying more house edge, the one who risked 7 units in two bets who's done for the day or the guy risking 30-60 units.

One method goes vertical and compresses time and one goes horizontal and stretches out time.

The house cares less if you bet 4 units stacked on top of each other on a single bet or 1 unit over 4 bets. House edge is the same.

The flat bet is more forgiving, a slow bleed with a very low unit profit per hour. You have to win more times than you lose to make profit which is tough when the probability to lose is greater.

The martingale is good until it breaks, the grand martingale profit is even better at +1 unit per bet made. A beautiful thing until that inevitable kaboom.

 

That's where a tiered positive progression approach or the reverse labby can accommodate some losses but still grow.

Eg 1-2-3, 2-4-6, 3-6-9, 4-8-12, 5-10-15 etc work you way up using a 3 step neg progression at each level.

Or a reverse labby 1, 2, 3, 4 or stretched 1, 1, 2, 2, 3, 3, 4, 4

 

Hello friend @TwoUp , if you don't mind, I need a percentage out of curiosity.
I've seen you're good at math.
What is the chance of me losing 3 consecutive times a bet that I have a 70% chance of.

I'm grateful if you can answer.

 

Agree that the losing streaks kill the negative progressions.

That's why Median Joe and I were discussing if one can reduce the dispersion in the W/L registry which basically means reducing the streak lengths and without any probability edge advantage all of a sudden a negative progression becomes viable and in fact a grail.

Reducing dispersion is the key.

 

If your probability is 70% then losing is simply 1-0.7 = 0.3 and losing 3 times is 0.3×0.3×0.3 = 0.027 = 2.7%.

That 2.7% is conspicuously the same as the house edge on single zero roulette.

 

There used to be a website way back ran by someone called Mr Oops and he had his own version of a reverse labby which looked interesting but I have not done any testing with it because it looks a little complicated. It may be difficult to run both a complex bet selection and MM like his reverse labby at the same time.

Reversed Labouchere with Stop Loss (888casino.com)

 

Thanks for the answer!!!

 

Oh by the way,

But you have to consider that inevitable kaboom may not be that bad if you are ahead quite a few units and your win goal per session is equal to the number of Marty steps that can bust. I.e. If a 3 step Marty is used, a complete bust will be 7 units. Therefore every session has to have a win goal of 7 units. This is a very important part of the whole thing. Of course the more Marty steps used the better chance of a hit within. But the higher the loss also. What has to be accomplished here is to have a BS that consistantly hits within the number of Marty steps used. If a million outcomes dictate that a 9 step Marty wil be enough to cover losses, then so be it. If you can ever make a possitive progression consistantly work, I will consider you a genius. Or at least a lot smater than me which isn't saying much.

Marcos has been testing this strategy for quite a while now and has yielded some very good results. He basicly uses my strategy as defined, but has found that waiting for certain triggers is a better approach than mine. He actually shows me screen shots of him playing for real money. It doesn't matter to me who comes up with the best solution, just as long as something is found that yields the best results. Come on mathboyz, let's find out exactly what that it is with all of us working together towards same goal.

 

This will tick off those that like the RR method and flat betting with virtual bets and then the big bet when the only bets are placed. In other words interrupted flat betting. But what would happen If I mixed EC's with my hedging, insurance betting and a Marti? I've never tried that. I think I will play a bunch of sessions on my R-Sim for Giz2. I really don't care if Turbo thinks I chickened out or not. This might put a new light on the Marti. I like the idea of a stop loss after four steps in a row loses.

This will be fun even if it does not work. Perhaps I can get some of that excitement DG is having. So I think I like EC's with a half priced hedge bet and increasing to double down on the EC and double down on the hedged bet.

So $50 + $20 on the hedge bet will cover 28 numbers and break even on the 10 numbers hedging and +$30 on the EC win for the first step. From there it is double down for four steps.

$70 , $140, $280, and $560 = $1050.

So I would need 35 wins to break even.

Now I apply RR to the bet selection process. I know some sessions are easy and some are difficult. The trick is not to lean of the progression but to avoid the difficult sections. So that would be a virtual bet until it eases up. Very simple.

If this works then it will be a MathZombie & Frankenstein hybrid. Too funny.

I think I might like this. Now watch both sides blow up. Heck with flat betting. LOL

 

TwoUp, below is a copy of some of the correspondence I've been having with the gentleman who is independently testing my strategy. See if it offers anything to the mix.

Hi Dave, actually I have been doing a lot of testing with your basic Doubles, Alternating, FTL.

I have found that the weakest link in the 3 patterns is the FTL. I have found that FTL, DBL, OTL produce pretty much the same overall results.

In my research, I have found the Static Pattern of RBRBRB produces a more stable result, one that might be more susceptible to progressions.

I have tested your basic Doubles, Alternating, FTL and it produced a profit of 309 units on 100,000 spins of No Zero.

When I substituted the FTL with the Static RBRB only when in profit and used FTL when not, I got a profit of 400 units on the same 100,000 No Zero spins.

I think the patterns you are using, Alternating and Doubles have some basis in Math so that may be why they might be doing better than expected.

You could argue that these patterns are nothing more than the AP in the VDW theorem. These would be the AP's 123 and the AP 1-3-5, 2-4-6,3-5-7,4-6-8, and 5-7-9. If you are not familiar with the VDW theorem, I have attached a PDF summary.

While this looks good, it is only a start and more testing needed but at least it gives us a direction.

 

Have you come over to the dark side?

 

That's what they all say before the kaboom..

 

What if bet placement is good enough to always recoup the kaboom and continue to get ahead?