Reducing Volatility @ TwoCatSam

[bloomone2002]

Ok, I'm still thinking here. Lucky Strike, I think your Cut Point Methodology can be modified to benefit this method's functionality. I just did a quick read of the concept, I didnt understand it all, but my intuition says it will be beneficial when combined with the L0W registry.
Bloom

[lucky_strike]

Well i will keep it up and give it some thoughts i will test some things but i have to say that i agree with Two Cat Sam, but at the same time i find it fun and interesting to play that way, i will continue this but every thing takes time because this is a new concept for me.

I will be back and maybe we can come up with a strategy that does work. Now i have to make some effort to the baccarat section.

Cheers LS

[lucky_strike]

bloomone2002 should we make something out of this or should we skip it.

[bloomone2002]

Well, i think there is something here, but if no one adds to what is already here, then we will have put on it hold for now. I am willing to entertainment any continued progress on this topic. It is unfortunate the thread author abandoned the thread

[TwoCatSam]

Gents

I put in a lot of hours on that little bugger.  Many hours of thought while building a carport.  I had that idea years ago and wanted to re-visit the thought.  

Here is my opinion:

What he calls a "reduction in volatility" is the same as just betting every tenth spin of the wheel.  You can say, "Hey, I will only face the chance of loosing one spin out of ten."  You have also reduced you chances of winning to one spin out of ten.

I did find it interesting that he actually increased the house edge.  I didn't think that was possible.

Sam

[bloomone2002]

I think in part that is true. i personally think, if you review my last 6 post or so on this topic it is more about adding variables to the same.
i.e. you take an apple. it will always be an apple you can bet on it. However, if you factor in other variables about the apple like, shape, color, height, width, depth, texture, smell, etc. now you you are armed with getter distinction but the top level (apple) doesnt change. That's another reason, i created the L0W registry, to evaluate a more multi-dimensional change. The more dimensions you add to something the less subtle the changes are within itself. it is the interworking of all multi-dimensional subtle changes that causes the big top level change, like an apple gone bad in texture, shape, smell. i dont mean to get too metaphysical here but it does have its root in this application.
Also, for greater emphasis here, the work that Victor and Lanky have done on the LW methodology is fundamental here, however, it is only linear to this concept, its needs to be applied in a more dynamic or multi-dimensional fashion, again that is why i modified that concept with the L0W registry. its like tracking the LW for several interconnected games to give you one best case result of the overall game, because what manners most is the overall game. I hope I am making sense here. I feel I'm starting to write that the author of the thread. Scary!

Bloom

[TwoCatSam]

OK, bloom

I reviewed your LOW idea.  Let's get specific.  You look for three Os?  What about three Ls?  Combination?

As I studied the Lw Methodology, we first had to determine what pattern the wheel was producing.  

Are you saying something like this:  When the wheel has produced three non-Ws in a row, we bet?  Example: LOO, LLO, OLO, LLL, LOL.

I'll study with you on this as it's a new twist.

Sam

[bloomone2002]

Yes, Sam and there maybe other acceptable patterns to capitalize on. The key is applying the fundamentals of the LW to a multidimensional structure (ie. b/r & e/o or b/r & h/l or more add one more dimension b/r & e/o & h/l), essentially we have to establish an out of balance pattern, that would otherwise be subtle to top level game.  So, we would have to look at more spins and maybe we should start a new thread in the testing zone to iron this out. These out of balance patterns to me are stronger than the typical 2 dimensional patterns we have been used to. If you agree, i will start a new thread and maybe LS and others will join in at some point. In reply #73, my 2 independent test, i up 800 u in 17 spins, using the L0W registry, now i dont know if this will hold up, but this is what i call exploitation from those results, even if short term. i could walk out the door with those results.
Bloom

[TwoCatSam]

bloom

Sure, start a thread.  

Could you post exactly how you are doing the test?  

I'm thinking I'll do this:

Look for the EC + EC then wait for three Ls or Os, then bet the opposite of that EC

Example:  Red/Odd is the first thing I see.  I am seeking Black/Even and wait for three spins with no Black/Even winner the bet for B/E to occur.

I assume zero is always a loss.

Is that acceptable?

Sam

[bloomone2002]

yes, that is an acceptable start

[reddwarf]

Hai all,

After reading a lot of statistics books and after studying a lot of the systems/strategies that can be found on the web, the remarks of monte carlo really puzzle me (sorry, I'm not a native speaker).

Has any of you heard of this person again? I'm really interested in what he means with "inverse betting"!


greets rd

[reddwarf]

Because when I do the math, I see indeed a tiny decrease in variance and an increase in the house edge, but I am not able to keep a reduced variance and standard house edge by "taking the square root of the original matrix".  Need some input here!

greetings rd

[reddwarf]

OK,

It took a while, but I finally cracked it: the statements of monte carlo are, unfortunately, not true.

By combining two EC bets, say high/low and black/red we do decrease the "volatility" when we defined this by win/loss per number of total bets, but we actually INCREASE the probability of a long loosing streak!

I will not bother you with mathematics, but a simple example will show what I mean:

When we combine bets there are many ways to loose two times in a row:
break even= b/e
loose =l


l - l
b/e - l - l
b/e - b/e - l - l
etc. . .

l - l
l - b/e - l
l - b/e - b/e - l
l - b/e - b/e - b/e - l
etc. . .

l - b/e - l
b/e - l - b/e - l
b/e - b/e - l - b/e - l
etc. . .

if we do the math, the probability of loosing two times "in a row": 0. 2770.  When this with only black/red:
The probability of loosing 2 times in a row with black/red is equal to (19/37)*(19/37)=0. 2637 for European roulette.  In other words, the probability of loosing two times in a row increased!! Well, we can actually proove that this is always the case with even money bets.

This basically means that all other conclusions drawn by monte carlo are incorrect as they depend on reduction of loosing streaks.

[Noble Savage]

Good observations reddwarf.

[reddwarf]

Unfortunately.  I would, like many others, rather have it otherwise.  I tested all kind of methods for almost a year and these are my conclusions:

1.  spins are truly independent: online casino's do not need to cheat, they just have to invest in good RND generators
2.  in the end the house edge will get you
3.  progressions do not work, in the end
4.  systems where you have to bet on single numbers have a better "survival rate".  no wonder systems based on single number bets take a long time to prove unsuccessful, but in the end the house edge will get you
5.  things like Allison mixtures, Parrondo Paradox, Volatility pumping, Kelly criteria all do not work because no betting combination favors the bettor

There is one more system that I'm designing: a system that is using unlikely events to come out positive: all systems I have seen thus far focus on regular winnings.  money management is used to try to survive rough times. 

After that, I'm done with gambling: I lost (luckily) only a few hundred dollars.  And many, many hours simulating million of spins for almost all systems that I could find on internet (+ a couple of my own design that were also used to study Parrondo, Allison mixtures and Volatility pumping, Optimal strategies).  None of them were winners, although, again, systems using single numbers might give you the illusion that you have an holy grail somewhat longer than systems where you bet on outside bets.  And I found 1 system that gave you winnings 99. 7% of the time, but it was devastating 0. 3% of the time.

[reddwarf]

-