Author Topic: Playing with Quotes and Concepts  (Read 2978 times)

0 Members and 1 Guest are viewing this topic.

vitorwally

Playing with Quotes and Concepts
« on: February 04, 2019, 09:31:52 PM »
Just to let you know, the system do exist.

DrTalos starts his journey in this message board with this sentence.
For many, these words and future ones are already judged. "There's no such thing as a winning system", "Have you ever heard of House Edge?", "HG straight up on the topic title? What a disgrace", ...
For the remaining, posts of encouragement and intellectual challenge on this game that brings us here.

But what if someone really tries to live up to the illusion of an approach providing a good bankroll growth rate?

The only way to achieve that (other than connect all clues and hints of this long thread) is start with the idea that not all that there is to know is already known. I suggest an empirical approach to find a backdoor to crack the game. If you start with the same point of view of others, you will follow paths already fully explored.
Pure math says the HE cannot be beaten? Well, maybe there's another way, a creative one to bypass that rule we don't like.

I've been reading DrTalos posts, posts from other users claiming to have the Holy Grail (sorry, my bad), posts from experienced users that hadn't such claims and posts from people "giving a fight" to previous mentions.

I never closed any door, if we talk about what I read, but it's indeed important to close some doors, even if for brief moments, when we're thinking about our options. There's the only way to INSURE that you go all out in the doors that are left open. It's a pathological fault of mankind, especially men. Attention should be given to a close spectrum of hypothesis. The wider it is, more time it will take you to fulfill your purpose.

Therefore, I enrolled in my own testing and I kept reading. I watched videos, I heard opinions, I eyeballed behaviors.

I've been developing a way of playing for over a year, and lately the breakdown of my results is shifting from a "Hmm, still not good enough" to "I need to double-check these". However, before the results, let me emphasize the main aspects of my playing:
- an hit ratio; @DrTalos
- insurance of my moves in the table; @Rinad
- a money management tool;
- static bet selection and variable bet selection;
- variable sized bets and flat betting;
- having a stop point;
- wagering every spin.

Here arrived, let's proceed with the results. I have two versions derived from the way of playing I ended with. One is more mild, the other is more aggressive. Basically what differs is the number of losing MICRO-sessions I get in a MACRO-session. I had my simulations running for thousands of iterations (before you ask, it did cross the million mark). Results:
- Mild version: average 0.8 units per spin in a long run, less than 10 micro-sessions per 100k iterations, all ending with profit.
- Aggressive version: more or less 1.25 units per spin, over 90% win rate (winning micro-sessions Vs losing micro-sessions). The average profit per spin here is more volatile due to the increase in risk. I can have it reach per example 1.4 units per spin. It depends of the nature of the MICRO-sessions individually and the MACRO-session as a whole.

IMPORTANT: I never lost the starting bankroll units in the more mild approach, and in the more aggressive one it happened ONCE (with the MACRO-session ending in LARGE profit). The cases where I have to stop simulations because they're getting to long establish the norm, e.g., leaving a simulation overnight and get it still running in the morning. To conclude this "important" note, as long as it's insured that the ways and means you use protect you from a rare loss event, what's the deal of the loss event at first place?

Before I end, let me say I'm gonna upload as attachments some graphs. They're named in a way that eases the identification of the conditions. That leads me to the major motivation of creating this topic:
What do you guys have to say about them?
I'm not here to brag on or start a brawl. Although I infrequently post, because I have my duties, I think some of you know I'm not the type of person that promotes that type of environment online. I'm just curious to know what goes trough the mind of users with other involvements after considering my results.
Recalling the first post I started with:

Good work to all, and thank you for keeping this forum so amusing.
 
The following users thanked this post: juice, MickyP, rimsky, Third

vitorwally

Re: Playing with Quotes and Concepts
« Reply #1 on: February 04, 2019, 09:33:50 PM »
Just a couple more attachments.
« Last Edit: February 04, 2019, 09:38:13 PM by vitorwally »
 
The following users thanked this post: juice, MickyP, Third

juice

Re: Playing with Quotes and Concepts
« Reply #2 on: February 05, 2019, 01:47:05 AM »
I don't know jack sh*t about graphs! But they all seem to be going in the correct direction.
Congratulations, and thanks for a great post.
All the best, juice
 
The following users thanked this post: Third

MickyP

Re: Playing with Quotes and Concepts
« Reply #3 on: February 05, 2019, 03:04:33 AM »
I like the way you think Victorwally.

Is is playable on a B/M casino table or did you develop it for fast spin online roulette?

« Last Edit: February 05, 2019, 03:07:58 AM by MickyP »
 
The following users thanked this post: Third

MrPerfect.

Re: Playing with Quotes and Concepts
« Reply #4 on: February 05, 2019, 07:02:21 AM »
Looks like you killed your excel or RX prng on the paper testing. Real life results is more rough then your tests show. Somewhere used to be 1m life roulette spins l posted ( do not know where exactly, sry), test on these to compare if you find them.
   Do not see big difference between "agressive and mild ".. maybe your agressive is not motivated. ..
   Normally agressive betting is used to curve fit for best conditions in play... like your average expectation is 2-3% positive, but sometimes it's 20-30%, so you use agressive to jump to another high in bank while game is good ( natural camo as well).
 
The following users thanked this post: Third

vitorwally

Re: Playing with Quotes and Concepts
« Reply #5 on: February 05, 2019, 01:10:38 PM »
@juice
Thanks for the kind reply!

@MickyP
I want to show recognition for your words of appreciation.
Regarding the way of playing I ended with, I don't want to disclose much information about it for now. Why? Because then the results would not be judged by their unadorned features, but by a compound of that with the opinions of each one in relation to roulette playing. I may give some hints in the future.

@MrPerfect.
I recognize worth in your suggestion, testing with spin data from real wheels. Naturally a large sample. However I cannot find the file you are referencing neither another I read about posted by Bayes. I know where to find outcomes generated by real wheels, but not a large sample like 1M records. Maybe I'll need to compile my own data. Perhaps I'll do data fetching from random[dot]org as well.
I understand your view respecting aggressive playing. It's veritable a good technique to approach the waves of variance.
 
The following users thanked this post: MickyP

vitorwally

Re: Playing with Quotes and Concepts
« Reply #6 on: February 10, 2019, 11:30:01 PM »
Experience is the process of learning. You don't get experience just growing old.
You need to fall, to fail, to get to the ground, to hit the rock bottom, and the way you raise yourself up, that's your experience, that is very limited to that situation that brought you down.
...
The process of building, testing, failing, tweaking, correcting, and finally apply with success is the full method of "experience gaining".

Considering MrPerfect. suggestion, I searched for B&M wheel spins, or in his words, "real life results". I searched, downloaded and fetched different samples of spins, as I said I would do in my previous post on this thread.

Consequently, I tested the spins. But all the practice of preparing the conditions for the simulation helped me discover a "hidden" computer glitch. It was causing improper performance, and, therefore, the extremely positive results. E.g., it was maybe the cause of the simulation that got itself running overnight. It was causing more errors like control structures not functioning properly when dealing with big arrays. What was the matter? Memory management. I guess I learned my lesson regarding memory management. This just shows how critical lack of system awareness may be.

At the moment I got perception of the malfunction I thought it was my moment "to hit the rock bottom". I thought I published online a misleading post. Thus, I find this a good occasion to assume a mea culpa. Dear readers, take this post as my errata to the opening one.

Before advancing to the aftermath, the results I'm getting are now from a single approach, because, again, I followed MrPerfect. assertion. There was not much divergence between both approaches, so I'm going on with the one that suits me the most.

Repercussions?
The issue I explained corrupted my way to profits ranging from 0.8 to 1.4 units per spin. Now it's around 0.5 units. More timid earnings per spin but way less losing micro-sessions. It was, at best, less than 10 per 100k spins, now it's less than 10 per 1M spins. Slower profits, but more protection from harms.

I've recoded, almost "from the ground", the base of the simulation, making it as clean as possible, and was concerned about memory management, using all the resources I could with my knowledge of the framework. By this order of ideas, I have the confidence to say the results shouldn't (I'm going to repeat it, shouldn't) change. I ran multiple tests and in consonance with the opening post I'll leave some attachments. Some tests were defined by the length of the data sample, others, per example the RNG ones, by how long I wanted the simulation to go.

In short, I learned a lesson, never dismiss good programming practices. Plus ending with the accurate view of what I had in mind. Sorry for being ambiguous and confusing, even if without purpose.
 
The following users thanked this post: Third

vitorwally

Re: Playing with Quotes and Concepts
« Reply #7 on: February 10, 2019, 11:33:28 PM »
More attachments.
 

vitorwally

Re: Playing with Quotes and Concepts
« Reply #8 on: February 10, 2019, 11:34:53 PM »
Final attachments.
 
The following users thanked this post: Third

Third

Re: Playing with Quotes and Concepts
« Reply #9 on: February 11, 2019, 05:30:15 AM »
Wow incredible results!  On the 100K spins it looks like the scaling is a bit off on the X axis but I guess there is nothing much we can do with so many spins and the important point is that it just goes up anyway.
 

vitorwally

Re: Playing with Quotes and Concepts
« Reply #10 on: February 11, 2019, 03:52:42 PM »
@Third
Thanks a lot for your feedback! I agree with you, those aesthetics details mean nothing when side-by-side with the actual results.
 

vitorwally

Re: Playing with Quotes and Concepts
« Reply #11 on: February 14, 2019, 12:42:05 AM »
Some stats after 7770221 RNG spins, when reaching a bankroll of 4000000 units. Base chip 1 unit, not cent chips or more than the unitary value. 25 micro-session interruptions. All ended in profit with a growth rate on average of 0.51 units per spin. Here's the points of interruption. The spin counter represents the number of spins since debt started. I'm posting the stats as a .png cropped from a screenshot of Notepad++ in order to be possible to see the results aligned. It's easier for the eye than the preview I was getting here. 
 

Third

Re: Playing with Quotes and Concepts
« Reply #12 on: February 14, 2019, 12:55:33 AM »
Wow that is amazing!  What happens when you get an interruption?
 

vitorwally

Re: Playing with Quotes and Concepts
« Reply #13 on: February 14, 2019, 01:12:56 AM »
@Third
I start a new micro-session, as if I had just sat down at the table. Have in mind, and using as support the results I just posted, that on average, I got less than 4 micro-session interruptions per million spins.
 
The following users thanked this post: Third

Third

Re: Playing with Quotes and Concepts
« Reply #14 on: February 14, 2019, 05:12:43 AM »
Wow that IS amazing.  Do you start over as if you have no debt in the next micro session after an interruption?