Diminishing weight for 37-number cycles and their key usage

Started by VLS, Dec 30, 2022, 06:38 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

VLS

The weight for numbers as cycles go by resembles a chart of diminishing utility within my betting framework:

marginal-utility.png

Each cycle within this diminishing weight approach has a key usage.



First cycle: 1-37 spins = active

Only numbers from the last 37 spins / last cycle should be considered for real betting.

The closer to the start of the cycle, the greater the weight for the affected events affected by the more recent spun numbers, with no betting eligibility for numbers past the 37th spin (they remain only for stats).

This means no sleepers are being bet. Ever.

Second cycle: 38-74 spins: guide

The second cycle is used as a guiding mechanism for collecting eligible patterns for tracking continuations.

If they become steady across several sub-cycles then the better.

Third cycle / 75-111 spins: untie

The last cycle has minimum weight hence its functionality is basically to be used for untying events with equal weight.

Even with minimum weight, it's still a valid cycle and it's part of the immediate timeline for the current/realized full cycle set of three.



In the grand scheme of Roulette, 3 cycles is a very short term... yet when used in a rolling window basis, they are enough to "ride the ever-moving wave" in the timeline of events for recent randomness.
Email/Paypal: betselection@gmail.com

-- Victor

HardMan

That's a great point!

My question is do you affect this in coding, as I have two such intervals -- foremost dedicated to qualifying, & second to fine-tuning the grouping, in short, positioning.

VLS

Quote from: HardMan on Dec 30, 2022, 09:42 PMThat's a great point!

My question is do you affect this in coding, as I have two such intervals -- foremost dedicated to qualifying, & second to fine-tuning the grouping, in short, positioning.

Yes, indeed.

Variations on the "weight" can be programmed/calculated as % of the total spin-window according to position.
Email/Paypal: betselection@gmail.com

-- Victor

TwoUp

I have similar views as per my table on appearances of numbers and repeaters.

I was thinking a ranking of all numbers based on the past 155 spins or so using a simple formula based on a cumulative sum of each occurance being the square root of the number of spins since it occured.

Here are some curves on what that looks like 1/√x seems intuitively about right as a decay function for ranking purposes.

Once you have a ranking then decide how many and which numbers you want to bet based on their relative strengths, hit count etc.

Screenshot_20230106_081744.jpg