Coding request in exchange of 1 month unlimited use of the tool!

[Richard]

Hi Richard,

It's tough to deduce a general rule from one example, help me out here! 

Does it mean you DO see a section through to 'completion' when you're interrupted by a zero?

22
36
0
---

0 completes the section, so it's recorded as a loss in the above example and its treated
similar to an n/b, that is again in the example 12 col B had 1D*^ and the same column
is chosen after the 0 with 17 getting the correct result 2D*^.

[Twisteruk]

I think you guys would be better off talking on the phone !

Or use our chat room ?


Just a thought 

[Richard]

I think you guys would be better off talking on the phone !

Or use our chat room ?


Just a thought 

I think at this stage we have just about sorted out the problems.

Regards
Richard

[Twisteruk]

Cool 

[Bayes]

I think at this stage we have just about sorted out the problems.

Yep. 

[Richard]

Hi Bayes,

Just to let you know I am working my way through your spins, so far 357 select spins
(placed bets not including n/b) and just one 10 with reference to the number of intervals
between hits

[esoito]

@ Twisteruk

Those unfamiliar with Richard's system would find this thread of great value as Bayes works his way through the steps.

It serves a very useful educational function. And that, surely, is what this forum is about...

So, no. Using the phone or chat would not have been a good idea!!

[Twisteruk]

@ Twisteruk

Those unfamiliar with Richard's system would find this thread of great value as Bayes works his way through the steps.

It serves a very useful educational function. And that, surely, is what this forum is about...

So, no. Using the phone or chat would not have been a good idea!!

Yes, I see your point of view. I had not thought of that. 

[Bayes]

@ esoito,

Good point, If Richard's system has merit in the long term, this thread could be useful reference, since the system is not that simple to understand.

@ Richard,

I've more or less finished the coding, but I need to do some testing to make sure the program behaves correctly in the no-bet cases. All being well, I'll upload some output tomorrow (or maybe later today) so that you can check against your results.

[Richard]

Bayes,

Great news, look forward to your output.

I've managed to get up to 487 select spins and just two intervals between hits 10 and
above, 10 and 13

[Bayes]

Hi Richard,

Ok, everything seems ok and I've uploaded the results from the first couple of thousand spins in my file. All being well (ie; no errors/bugs) I'll get some more spins from wiesbaden for the big test.  

In the big test I'll add some code to collect statistics (frequencies of the 'gaps').

Note: I've shown all the spins so that you can see each complete section, but NOT the zeros in cases where it has hit before the actual bet is placed. This was just to make the programming easier and in any case a zero contributes nothing to a section so can effectively be ignored. A win is shown as "+0". Also you will notice that sometime in n/b situations a Dozen can be marked with a ^ or *. Again, this isn't an error but just a consequence of the way I coded (working sequentially and from left to right). Just ignore it, it doesn't mean that the column in the next section will be chosen by referring to this n/b situation.

The maximum length 'gap' was 13 in this sample.

P.S. I've just run the whole file (11,241 spins) and the longest gap was 15. So assuming the code is ok this seems to back up your previous testing. Note that this isn't 11,241 placed bets, which I'm guessing is about 3000 - 4000, but I'll add code to count them later.

[Richard]

Many thanks, Bayes I'm most grateful for coding it for me

So it's edged up from 14 to 15 It really is though a great result. Combining all the
results so far, my tests and your spins, I'm sure no one could replicate those results
and therefore we are in new territory

The big test will be the first really important hurdle and hope for the best.

I have got some further questions regarding the big test, but will wait for the results first.

I'll download and check the results this evening.

[Richard]

Bayes,

I have checked the results of the program and just two errors relating to zero.

7
5
0
---

13
17
0
---

They are both showing as a loss when it should be n/b.

Col 1 expects 1D and also col 2 1D and therefore n/b

[Bayes]

Ok, that was easily fixed. I've uploaded the new results file.  

I've been trying to download spins from wiesbaden but there's something weird going on there. Sometimes I get a file without any spins, or spins are duplicated. I just don't think it's a very reliable source. Any other ideas? I assume you'd rather not use random.org spins for the 'big' test.

[Richard]

I would prefer actuals for the first test, then depending on the results perhaps consider
random.org.

They have spins available at VLSroulette:

link:://vlsroulette.com/actuals-permanences/wiesbaden-t3-spins-01-04-2003-31-12-2008/