+ Register

[sqzbox]

I agree completely. In fact I said as much way back in reply #6 and even offered Vic a chance to suggest a better forum but I don't think he got my point as his reply was about functionality, which wasn't what I was asking at all.


Gizmo - let's call it quits on this branch of discussion at this time, and let me just say in closing that I am really looking forward to reports on your progress - perhaps in a more apropos thread.  Thanks for your detailed and interesting info on your work.

[topcat888]

Thanks Bryan, indeed if I were a mod I would assign you a place of your own, I believe we have a lot to learn from you. 

[Bayes]

Hi sqzbox,
Welcome to the forum, you've written some interesting posts, and nice to have another programmer here. 

BTW, regarding PCs, there are alternatives to Windoze; have you tried Linux? it's a very nice platform for developers with lots of free tools for those who like to get "under the hood". And of course there's the Mac, which is also Unix based.

Regarding testing, I rarely write full-blooded simulations of systems any more, at least, not to find out if they "work", but only to test concepts and find probabilities which are too complex to be be calculated manually. Like you, I prefer a scientific approach, I focus on trying to find ways to reduce the variance - if you can do that, then you have the game beat, and it's not necessary to find a flat-bet winner in my experience.

[Gizmotron]

I noticed that the topic had split too. I didn't want to suggest it right away. These were Bryan's  first posts. I didn't want it to be tossed around right off the bat. I would like it split off in my software section. That way finding it will be easy.

Confession time. Byran, I'm only using the 12 outside bets and the colissions of about half of them to offer as a proof of concept demo. I believe it's enough to manipulate, conjure, or extrapolate the effectiveness track. In the end, the primary task is to win by allowing myself to win. Proving it will be this year's task.

[esoito]

I would like it split off in my software section. That way finding it will be easy.

Done.

I had to provide a title for the splitting to work.

Let me know if you want it changed.

[Gizmotron]

It's a better name than I was thinking. It's what it is too. Good name.

[esoito]

It's a better name than I was thinking. It's what it is too. Good name.

Thanks, mate. I appreciate that. 

[sqzbox]

Hi Bayes - and thanks for the welcome.  Yes I've thought about Linux but, I don't know if you are familiar with VMS or not, but after working with this most elegant OS for so many years I have to say I find Linux unnecessarily cryptic and difficult compared.  I've stuck with Windoze (love it!) because in my consultancy work I have to stay with an industry standard - and even, dammit!, have to recommend Windows when interoperability is a consideration, which is most of the time in business.


What you say about testing concepts rather than systems is exactly where I am at also.  However I still refer to what I do as Monte Carlo simulations because, while I am not applying funds in any way, I am running simulations of strategies to expose the true probability resulting - this is what a MC sim actually is.  I am not "betting" but rather comparing the actual hits to expected hits and determining the resulting probability compared to theoretical probability so as to determine any advantage (or reverse which is just as good).


Gizmo - "proof of concept" is all it takes.  This is the scientific method.  I'm particularly interested in the concept, as you say, of abstracting the bet selection from the effectiveness track and seeing where that leads.


All the best
Bryan

[Gizmotron]

The trick is to first understand the problem. I think Einstein had a nice quote regarding that. I came to the conclusion that I was allowing myself to win. But it's a little different than that. I'm isolating synchronization. This is a new field of research emerging.

http://plus.maths.org/content/sync-emerging-science-spontaneous-order

"And here we get the first taste of Strogatz's excellent skill in guiding the reader through a mathematical proof, explaining the first tractable model of group synchronisation developed by a Japanese physicist, Yoshiki Kuramoto. Reading this section feels like touring a beautiful architectural structure, appreciating its ingenuity and creativity, its beauty and its strength, giving a real feel for the proof without any of the mathematical details that might put some people off. That isn't to say Strogatz shies away from mathematical concepts, there is much discussion on linear and nonlinear dynamics, differential equations, differences between particular mathematical models, but these are always explained in a physical, tangible way. "

Imagine that. A scientific study of the concept of spontaneous order. And they might have missed Roulette of late. HAR HAR HAR.