This is an excellent way to measuring even money bets.

[lucky_strike]

If you find an method on this forum and want to know if its any good or not you can measuring it and find out the truth by your self.

Flat betting.

The number of won units is to be divided by the root from the placed bets.

0.5 Keep looking
1.0 Keep looking
1.5 Hint
2.0 Interesting
2.5 Good /
3.0 Very good
3.5 Super
4.0 Dream on
4.5 You wish
5.0 Forget it
5.5 Never in an life time
6.0 The end of roulette as we know it

1024 placed bets would be the absolute minimum to get an hint about the method you test.

Take care with affection.

LS

[Jish]

so if you made 3000 bets and made $1000

The number of won units is to be divided by the root from the placed bets.

=$/ √Bet
=1000/ √3000
=1000/ 54.7
=18.2

I definately have this wrong Lucky

[lucky_strike]

The calculation of the statistical Ecart for flat betting not valid for progression.

G = total gain in units
b = number of placed bets

nolinks://img121.imageshack.us/img121/839/estat.gif

LS

[Mr Chips]

Hi LS,

On the Signum thread you asked me about placed bets and I said I don't record them.

Tangram has also requested placed bets and it is likely now, that I will record them either
here or on the website. Just thought I'd let you know.

Regards

Richard

[Tangram]

Note that the simple formula:

[tex]Estat = \frac{G}sqrt{b}[/tex]

assumes that the probability of the EC is 0.5 (no house edge). Using it to get the SD for the signum results (G = 49, b = 254) gives 3.07 instead of 3.506 which I got using the true value for the EC of 0.4865.

This shows how much "harder" the system has to "work" to overcome the house edge.

@ Victor,

Thanks for fixing the latex, but there is text under the formula which I didn't write (the title of the thread - "This is an excellent way to measuring even money bets"). There is a bug lurking I think.

[Mr Chips]

Started putting placed bets on the Signum website.

[lucky_strike]

Great Mr Chips.

LS

[Tangram]

Another limitation of this simple formula is that it doesn't give you any idea of how good your method is when the gain is zero. If you break even after 50,000 placed bets, the SD is around 6.0, but the formula tells you it's zero!  

I will be adding the full formula to the reference section over the next couple of days. This can apply to all bets on the table - even money, dozens, single numbers etc.

[Jish]

The calculation of the statistical Ecart for flat betting not valid for progression.
LS

Not valid for progression, most systems are progression?

[Homeito]

Hello All,

Maybe I'm misunderstanding Tangram's intentions but I read it as you're going to post the formula for Standard Deviation...

The SD formula is already in the reference area:
nolinks://vlsroulette.com/reference-area/roulette-probability-made-easier-t2193/15/
(Chapters 14 - 16)

SD is not measuring your method's effectiveness money-wise - it takes only placed and won bets (not units).
It isn't good for progressions, that's true, but that's only because the progression itself is never considered.

You can still measure the method (bet-selection) that's using the progression.

From the above link:
"Breaking the 1 SD barrier is said to be equal to a 16% probability that the result is random
Breaking the 2 SD barrier is said to be equal to a 5% probability that the result is random
Breaking the 3 SD barrier is said to be equal to a 1% probability that the result is random
Breaking the 4 SD barrier and beyond is said to be a very small probability that the result is random"


Best regards,
Homeito Bemek


PS. No, I'm not here to stay; I was just checking if there were replies to my posts and thought I'd see what you're discussing nowadays...

By some of the posts around here, it seems to me that some very good posts in the Reference area are sadly forgotten - the above included.
I would recommend reading the following posts in addition to the above:

nolinks://vlsroulette.com/reference-area/it-will-not-happen-in-your-life-time!/
...which shows that what's happening extremely rarely, may never happen. Or may happen tonight! You don't know.

nolinks://vlsroulette.com/reference-area/12-million-rng-spin-stats-shared-by-poit/
nolinks://vlsroulette.com/reference-area/multi-million-spins-tests/
... that both show the same things but Poit's is targeted at 12M spins whereas the second shows the math values for many different number of spins, including 12M. It shows very clearly that (RNG) spins absolutely conform to math as all Poit's values are also found using a calculator instead of (RNG) spins.

All these posts are absolutely worth a close study.

Bye!

[Tangram]

Hi Homeito,

You are right that the info is already in Kon-Fu-Sed's tutorial, but as you point out, it's not that easily found and I think it deserves a thread on its own to (a) expand and clarify the concept a little more, and (b) give a step by step procedure (with some varied examples) so that anyone can do the calculations (and maybe create a spreadsheet). Also there are other possibilities which Kon-Fu-Sed didn't cover, like how to do the calculation if you're betting different portions of the wheel in a series of bets.

Also, it may seem like nit-picking but everyone refers to this measurement as the "SD", but the correct name is "z-score". The standard deviation is used in the calculation, but the final result is the "z-score" which you could think of as a figure of merit which expresses the effectiveness of the system/method in terms of the number of standard deviations from the mean (SD).

[Homeito]

Hello Tangram,

I don't mind if you post about SD... Not at all. Of course not!
I agree that Kon-Fu-Sed's explanations are not complete and somewhat simplistic, so: Good! Do it! 

My post was more of a reflection that it seems that good posts in the Reference area, showing very basic things that anyone should know, are rarely read.
Or not believed if they are read, it seems, in many cases.
Unfortunately.

By posting those links I hope that at least some will read and learn...


And not to be nit-picking  but SD and z-score are two different things, really:
The z-score is calculated using the SD (that is already calculated using the posted formula) to get the relative deviation from the mean, expressed in SDs...
This is what I learned, but in Swedish so I hope I translated it correct...
But this is another formula, that's using the SD value... I have it somewhere... No, can't find it...
(Of course the SD formula can be included in the z-score formula as it is the result that is used)

Maybe we say the same thing
(wouldn't surprise me)


My very best to you, Tangram.
And to all!

Homeito Bemek

[lucky_strike]

Hi Homeito.

Nice to see that you are from Sweden.
I from Stockholm and visit Casino Cosmopol.
If you are from the same city it would be cool to meat you and thank you for you cool site.

Take care LS

[lucky_strike]

-