Random Thoughts

[RMore]

Good approach Drazen - I do indeed wish you well!

Amazing work Falkor! A question - what do you think the rules are for the Quad selection and betting? Not sure I quite follow that. It seems a bit random.

cheers
Rog

[falkor2k15]

OK, I sussed it!

The stats only tell us about the next cycle event - not individual spins as far as I can tell yet!

So in the Priyanka tradition of stitching bets we have the following options:

Cycle Length 1

Bet last Quad (3 Streets @ 1 chip each) = 11/3

Cycle Length 2

1. Virtual spin
2. Bet last 2 Quads (6 Streets @ 1 chip each) = 11/6

OR

1. Bet the other 3 Quads (9 Streets @ 1 chip each) = 11/9
2. Bet last 2 Quads (6 Streets @ 3 chips each) = 35/9 parlayed

Cycle Length 3

Part virtual

OR

1. Bet the other 3 Quads (9 Streets @ 1 chip each) = 11/9
2. Bet opposite of last 2 Quads (6 Streets @ 3 chips each) = 35/9 parlayed
3. Bet last 3 Quads (9 Streets @ 6 chips each) = 67/9 parlayed

With 67/9 payout odds for Cycle Length 3 this explains clearly why we want increased constants for CL3 followed by CL2! And that's why CL1 is ignored - because it increases our chances of CL3 and CL2? Before, CL1 was more prevalent, albeit with lesser profit potential.

The parlay bet has been hinted at several times - but must have deliberately been left out of the video.

[RMore]

When you refer to cycle length, are you referring to the VdW on an EC, or are you talking about the repeating dozens and going for the one that defined the previous dozen?

And while I am on the dozen thing, do I recall correctly that a repeat of the defining dozen from the previous group is LESS than 33% likely?

[falkor2k15]

In fact, the stats don't seem to increase CL2+3 per se - but they increase them specifically for when we expect them to be the same as the previous cycle - bringing them in line with the bias created out of the defining quad. That's why Priyanka has different betting rules following a cycle ending in CL2 or CL3: she is going for a repeat of the same cycle length!

When you refer to cycle length, are you referring to the VdW on an EC, or are you talking about the repeating dozens and going for the one that defined the previous dozen?

I've not got to the stage of using VdW yet - I am just trying to understand how Priyanka came up with her betting rules and selections based on stitching bets for increased odds and exploiting the cycle length 1. So right now this is more about going for the Quad repeats on CL2 and CL3 - in line with the defined bias - but not yet with the specific defining Quad in mind.

And while I am on the dozen thing, do I recall correctly that a repeat of the defining dozen from the previous group is LESS than 33% likely?

That was for Dozens - but here for Quads I believe we need to go for the same as per my first paragraph above.

[falkor2k15]

OK, I sussed it!

The stats only tell us about the next cycle event - not individual spins as far as I can tell yet!

So in the Priyanka tradition of stitching bets we have the following options:

Cycle Length 1

Bet last Quad (3 Streets @ 1 chip each) = 11/3

Cycle Length 2

1. Virtual spin
2. Bet last 2 Quads (6 Streets @ 1 chip each) = 11/6

OR

1. Bet the other 3 Quads (9 Streets @ 1 chip each) = 11/9
2. Bet last 2 Quads (6 Streets @ 3 chips each) = 35/9 parlayed

Cycle Length 3

Part virtual

OR

1. Bet the other 3 Quads (9 Streets @ 1 chip each) = 11/9
2. Bet opposite of last 2 Quads (6 Streets @ 3 chips each) = 35/9 parlayed
3. Bet last 3 Quads (9 Streets @ 6 chips each) = 67/9 parlayed

With 67/9 payout odds for Cycle Length 3 this explains clearly why we want increased constants for CL3 followed by CL2! And that's why CL1 is ignored - because it increases our chances of CL3 and CL2? Before, CL1 was more prevalent, albeit with lesser profit potential.

The parlay bet has been hinted at several times - but must have deliberately been left out of the video.

Correction: the last one should be:
3. Bet last 3 Quads (9 Streets @ 5 chips each) = 55/10 parlayed

[falkor2k15]

I'm not entirely sure my theory holds water - it's possibly contradicted by these stats for same/different:
Defining Quad      55.48%   44.52%
Cycle Length      28.83%   71.17%
CL1 Ignored      30.37%   69.63%

To get the same cycle length as opposed to the defining quad (even though that's the catalyst for the bias in the first place), is only 28.83% - but it is increased slightly to 30.37% with CL1 ignored. It might mean something along the lines of: we will lose less sets of 9/10 chips in trying to obtain our 55 chips. I might go further and say we are looking at a 1.5% edge - or is this wishful thinking? 

[falkor2k15]

Defined by 1 to X            54.50%   15.52%   15.12%   14.86%
Defined by 2 to X            14.54%   55.84%   14.50%   15.11%
Defined by 3 to X            11.53%   33.39%   43.82%   11.26%
Defined by 4 to X            14.76%   15.63%   14.19%   55.42%

Why is defined by 3 a special case: 2 and 3 both become dominant!? Could it be a bug in my simulator?

[falkor2k15]

Defined by 1 to X            54.50%   15.52%   15.12%   14.86%
Defined by 2 to X            14.54%   55.84%   14.50%   15.11%
Defined by 3 to X            11.53%   33.39%   43.82%   11.26%
Defined by 4 to X            14.76%   15.63%   14.19%   55.42%

Why is defined by 3 a special case: 2 and 3 both become dominant!? Could it be a bug in my simulator?

Yeah there was a bug - now corrected - so expect it to be 14/15% like the rest. This test is therefore too granular - same/different are the only meaningful variables here.

[falkor2k15]

In Priyanka's last 2 videos - unlocking magic - spin 1 is missed out and fewer bets are placed. Let's look at the stats for that:
CL1 Ignored   1st spin ignored   40.47%   59.53%      49.62%   37.68%   12.71%
CL1 Ignored (same)   1st spin ignored            60.91%   35.10%   3.99%
CL1 Ignored (different)   1st spin ignored            41.94%   39.43%   18.63%

Same has now increased to 40%! CL2 (betting only the 2nd spin) becomes the most dominant for both same and different.

[falkor2k15]

Great effort Falkor

In the meantime I decided for another approach.

I traveled 2500 kilometers from my home and I moved to the island next to Priyanka. From here I plan to hunt him down and force to tell me the answer. 

Wish me well

Best

Drazen

He's probably with the mafia in Italy!

[falkor2k15]

I suspect that in a losing session the stats will go pear shaped - prompting us to switch tactic to try to bet on the running totals coming back in line.

[falkor2k15]

It's not clear what Priyanka does after Cycle Length 4 or why she does it:

1	34211	7	1		End of cycle: Bet all the other quads
		28	4	W
		27	3	VL
3	23143	22	3		End of cycle: Bet all the other quads
		18	2	W
		35	4		Bet last 3 quads
1	21341	4	1		End of cycle: Bet quads 1,2,4
		24	3	L
		30	4		Bet quads 1,2,3

In the first instance she supposedly would have played the 2nd bet and sustained a (virtual) loss - but the other examples suggests she doesn't play the 2nd bet in that situation.

The last example is from video 3 where Priyanka changed tactic - probably to do with aligning her bets to the stabilising of ratios. I am going to enter those numbers from video 3 into my simulator and see if anything stands out in terms of the ratios.

[NextYear]

I traveled 2500 kilometers from my home

Great move, Drazen!

Sretno!

[falkor2k15]

The only stats that are affected by the removal of Cycle Length 1 seem to be based on the cycle length in combination with the defining dozen, so I need to devise a new test.

Cycle Length (same)				44.98%	33.41%	17.19%	4.42%
Cycle Length (different)				0.00%	42.01%	42.09%	15.90%
CL1 Ignored (same)				0.00%	60.91%	35.10%	3.99%
CL1 Ignored (different)				35.84%	26.91%	25.30%	11.96%

We know that after Cycle Length 4 we expect a "different" cycle length (91%!) - but it doesn't appear the current tests show which cycle lengths are more likely for different.

CL234 (CL1 Ignored) to different

62.86%

71.48%

90.55%

[Tomla021]

hell of a lot of good work falkor!