Need expert advice with my DS system !

[TierPlayer]

Hi !
I'm new here and looking for advice.  I designed a system based on Dealer Signature and put it to the test for 700,000 real spins from the Hamburg casino (well, not by hand, I wrote a program to do that).
The program takes into account dealer shifts and cc/ccw spins.  It also can be set to bet either "loosely" or "strictly" (ie.   only bet when ideal conditions apply).

So the program predicts a number for *some* spins, not all.
My questions is : do the results show anything about the system being any good or crappy ?!

Here are the results : (in about 700,000 spins)
When betting 1 single number  : made 587 bets    -14. 1%
When betting 3 numbers           : made 587 bets    +4. 3%
When betting 5 numbers           : made 587 bets    +10. 4%
When betting 7 numbers           : made 587 bets    -0. 1%
When betting 9 numbers           : made 587 bets    -0. 5%
When betting 11 numbers         : made 587 bets    -0. 8%
When betting 13 numbers         : made 587 bets    -2. 8%
When betting 15 numbers         : made 587 bets    +2. 6%
When betting 17 numbers         : made 587 bets    -0. 4%
When betting 13 numbers         : made 587 bets    -2. 5%

My guess is that 587 spins are not too many. . .   

[lucky_strike]

What does it capture ??

LS

[TierPlayer]

I'm looking for the same distance (supposedly travelled by the ball) in the last spins.  If I see the same distance at least 4 times in the last spins, I bet that it will be repeated in the next spin too.
For example (all spins clockwise for simplicity) :
Spin #  Number    Distance from release point
1          8               -
2        10               +2
3        24               +2
4        16               +1
5        27               -10             
6        11               +3
7         0                -14
8        19               +3
9         4                +1
10      25               +3

The program will see that close distances appear very frequently and will advise that we bet on number 34 (last spun number was 25 and the average distance is +2, I think, so we get 34).  Well this number is the prediction, you normally would prefer to place ± 1, 2 or 3 numbers ie a total of 3, 5, or 7 numbers.

I have been playing this system in the local casino and I think it works.  Unlike the program I can use some more info  like wheel speed, rotor speed, ball scatter etc - in general I prefer to play when these factors are favorable (=not making the game overly random).  I have not been keeping financial data but I think I'm up.

[lucky_strike]

I know exactly what you talk about, here is a short samel of bias using distance, but I do it different.

2
11       
22             
0            
222               
44               
33            
44444               
333         
2222         
33333         
111         
00      
1111         
222222         
333333         
444444         
111111         
22222222   
000      
33333333   
1111111      
333333333   
111111111   
3333333333   
222222222    
33333333333   
2222222222    
1111111111
44444444  
11111111111
333333333333
111111111111
444444444444
1111111111111

Maybe you should try to explore local attractors, i do.

LS

[Marven]

The average is one bet per 1192 spins?

Unrealistic I'm afraid. Plus there's nothing too significant in your results anyway.

My experience is no DS method works unless you include some VB elements that take the randomness away (dominant wheel drop, rotor speed..).

You might wanna try something like this:

- Find a wheel with a dominant drop zone (I.e. a diamond that receives the majority of the hits, this somewhat takes the randomness away from the wheel for visual ballistics/dealer signature players).

- Track the travel distance (TD) between the start number (the number under the dealer's hand as he/she spins the ball) and the final outcome. Collect data, for both spin directions separately (clockwise, and counter-clockwise) or stick to one spin direction.

- In your tracking chart, add a column where you note the difference between the current TD and the most recent one. Example: TD for last spin is +6, TD for current spin is +10, the difference would be +4.

- Record only the spins where the ball hits the dominant diamond, and ignore those (less frequent) spins where it hits the other diamonds. If you have 2 dominant diamonds, track the results of both separately (but then you'd need to predict what diamond will hit when betting and base your prediction on that diamond's chart).

- Look at your data (mainly the difference column) for developing patterns/trends and ride/exploit any.

- I wouldn't recommend sticking to the mechanical trigger based mentality. Learn how to adapt to your wheel/data.

- You might wanna keep data profiles for each dealer/wheel you play for future reference.


Marv

[TierPlayer]

@Lucky Strike : I am afraid I do not understand your chart, could you please explain?
@Marven : Yeah it seems no real edge is achievable this way.  As we move to wider sector plays the win/loss % converges to -2. 7% I believe.  However, other than the spun numbers, the program has no clue if the DS can possibly exist, because it can't see the dealer spinning the ball.  As you said, the bets are less than 1:1000, so it gets impractical.  I did it like this just in case something big emerged under ideal conditions.  Perhaps incorporating some VB might improve the whole thing.  Another thing is that in my local casino they recently replaced their wheels.  I can't say this system currently works as it used to do : in the past I would see the same patterns over and over again, now only seldom and possibly this is random and unpredictable. 
Trouble with VB is mostly timing.  Perhaps a system could be crafted so that one would only glance at the wheel twice and then have a clue about the landing sector.  That would be totally acceptable - even if a tiny edge could be achieved.
I had a book in German which I didn't understand - perhaps the automatic translation was little crappy.

[lucky_strike]

Lets assume you divide the wheel into six quads.

Then every quad will have six pockets, wish number is not so important.

Now with every quad there is six pockets, 123456 and you can play that one will repeat 12133, like that, you can also do the same with distance 252, distance from 2 to 5 is 3 and from 5 to 2 is 3.

So you have a quad with six pockets but you only play 2 numbers and that give you 18 placed bets.

You have to know how to capture the quads with an high hit ratio.

Left column pockets hit ratio.
Right column distance hit ratio.


5
1      2
4 L   3
2 L   4
5 L   3
5 W   0
6 L   1
2 L   2
4 W   2
1 L   3
6 L   1
6 W   0
6 W   0
2 L   2
1 L   5
5 L   4
4 L   5
4 W   0
1 L   3
4 W   3
6 L   2
3 L   3
3 W   0  
4 L   1
1 L   3
4 W   3
1 W   3
6 L   5
3 L   3
1 L   4
5 W   4
4 L   5
4 W   0
2 L   4
4 L   2
3 L   5
1 L   4

Or you could play four numbers and play against repeats of pockets and distance.

Left column lines.
Right column distance.


5
1      2
4 W  3
2 W  4
5 W  3
5 L   0
6 W  1
2 W  2
4 L   2
1 W 3
6 W  1
6 L  0
6 L  0
2 W  2
1 W  5
5 W 4
4 W 5
4 L   0
1 W 3
4 L   3
6 W  2
3 W  3
3 L   0  
4 W  1
1 W  3
4 L   3
1 L   3
6 W  5
3 W  3
1 W  4
5 L   4
4 W  5
4 L   0
2 W  4
4 W  2
3 W  5
1 W  4

Local attractors - chaos theori.



LS

[TierPlayer]

Oh I see now.  Interesting, it simplifies some things.  As for DS, it appears that many people believe there is such a thing, others doubt it.  I have asked some dealers their opinion and basically they don't think DS really exists. . . but who knows? Supposedly they are in the "zone" when they function like that so they wouldn't realize it anyway. . . 

[Marven]

I make a distinction between dealer signature and sector shooting.

The latter is an existing skill but not all dealers can do it. You can learn it by purchasing a wheel, tilting so it forms a strong dominant drop, then practicing spinning over and over again, releasing the ball from the zero, keeping constant rotor speeds.

Assuming the wheel doesn't have unmanageable scatter/ball bounce, soon you'll notice a certain distance from the zero that is getting more hits than it should. You can then use that to your advantage and adjust the release points to "aim" at the number/sector of choice, giving it a higher probability.

This is why (rare) experienced sector shooters will tell you that they can't do it on any wheel, but their "favorite" wheel (one that has a dominant drop and minimum scatter).

As for dealer signature, it's a phenomenon that could happen with any dealer, but not necessarily intentionally, mostly not. It happens rather because of the wheel (having a dominant diamond and low scatter). You'll see the signature/patterns developing when the dealer spins the rotor at constant speeds, merely due to laziness/carelessness. The casino instructs the dealers to vary wheel speeds, but dealers are not machines and often get lazy/monotonous.

I'm sure they'd have no clue if you ask them though. Dealers and casino personal are generally not as informed about the physics of the game as advantage players are.

[TierPlayer]

Marven, very enlightening and helpful analysis !

[lucky_strike]

Well this is what happens when you observe release from the dealer and it is not random.
Lets assume you deivide the wheel into 12 sectors of three and observe release.

The dealer release from 8 or less with some kind of bias.

0 XXX
1 XXXXX
2 XXXX
3
4
5 X
6 XX
7
8 X
9
10 X
11 XX

12 times out of 20 from 9 numbers or three sectors.

The outcomes ended up at... with tilted wheel...

9 XXX
10 XXXX
11 XXXX

Then I also want to mention that local attractors can use an scale of -4 -3 -2 -1 0 +1 +2 +3 +4 where you pick two or four.

-2 and +2 or if you aim for four and get this distribution 122121132332423 you would get WWWWWLWWLWL

There is many things to explore, this aint all.

There also exist bias when it comes to the dealers spin behavior, soft, average and hard and it correlates.
The intressting thing here is that if you categorize one rotor speed around any value you can explore bi-modal effect or bomernang shape effect, as long the ball ends up in the same half of the wheel or direct oppisite.

Look at this 17 18 17 18 19 15 22 20 21 20 22 21 18 13 14 20 21 20

Here you have

17 18 17 18 19
20 21 20
22 21
20 21 20

With an tilt and constarnt rotor speed you will get two peaks where the ball end up.

With rotor speeds you can measuring them with 0.10 accuracy and only use categories of them.

3.1 3.0 2.9
3.4 3.5 3.6
3.9 4.0 4.1

With tilt this will become an true signature if you can make an correlation of does parameters above.

LS

[TierPlayer]

Lucky Strike, is it possible that shifting the numbers in the sectors would yield significantly different results ? Eg you have defined sectors 0 to 11 like this : sector 0-->32,15,19 sector 1-->4,21,2 etc.  If you had defined them starting from 15 ie.  sector 0-->15,19,4 etc, would this segmentation give totally different results? I'm trying to make a representation where a number is replaced by an area centered at this number, to use in spins analysis. 

[Twocando]

Hi mate

Dealer can only land the ball in two places on the wheel. Take the last number draw a line across the wheel. Now there are 2 sides - Left and right from the number. The ball will land in 1 and there will be a trend to repeat the side from the last. Even with the change of direction with the spins. The dealer will repeat the side from the last.

This is a quick way to see DS.

Now if you use this last number and places it in the middle of the half of the wheel. Check if the next is in the same half or across.

If you use both at the same time you will leave 8 to 9 numbers open and play the reset. You will see that there are dealers that have a trend to move in an clockwise movement from the last. That's more to the left from the last and not going more than 18 numbers away from the last.


Twocando 

[TierPlayer]

Twocando, sounds interesting.  I am prepared to write a program to test this theory through my database of 700,000 real spins, if you can explain the method with an example.  Then I will post the findings (or maybe I won't  )

[Mr. Investable]

Twocando - I see DS rotating in clockwise or counter clockwise directions on my favourite wheel.  I have studied around 1000 live spins on a dominant drop, constant 5. 5 sec per rotation wheel, always on a counter clockwise rotation with 30 minute dealer changes and I see a sector pattern.  Different patterns for different dealers but all visually identifiable.  I'm a noob (playin for 5 years) and can only (relatively consistently) predict hits in 12 pocket sectors (13 pockets under certain conditions) using a 2:1 progression to cover my misses BUT tonight I saw a guy (DassaB) place £110 on #26 and it came in! 26 was the VERY middle number of my 13 pocket sector!?! Coincidence? Who puts £110 on one number? I was watching his style through my session and I'm sure he wasn't using a simple Martingale on sleepers.  His patient approach, big wins and our coincidental sector bets led me to believe he was stalking the DS and I was (much less efficiently) following the same sectors.  I walked away in profit using a DS sector tracking system I'm working on (developed from an old system I found in a dead VLS)
I'd love to share the system with you (it's exciting) and you could run it through your 1000 spins.  If it's a winner of that many I'm visiting the cashier!
Ps: the guy who posted the original system claims to make +£2000 per month (based on his units converted to GBP) playing every other day! Hmm not millions but enough to quit the day job

Peace   

[Mr. Investable]

-