Is there anywhere to find, or a way to figure out specific parameters in a set of 100 spins?
For example is there any way to determine that over a 100 spin session on the even chances there will be some kind of probability where we can say that there will be:
a Chop X% of the time,
doubles X% of the time,
triples X% of the time,
quads X% of the time, etc.
Or is a 100 spin window to small of a window for this?
Can this answered by math or do a large number of tests need to be run to figure this out?
Nobody has any input on this? :o
In my opinion 100 spins is far too small a sample, within 100 spins you could have massive streaks over 10, a streak of 20 reds is already 20% of your sample. I have no idea what the best sample size would be either as I have run similar tests over different sample sizes and none have provided information that I find useful, or could be used to bet against/for.
Thanks for the reply superman....
Yes I also agree that there can be massive streaks, but with the method I have been brainstorming the streaks will not hurt you, you will be able to catch them. I am more concerned about the percentage of times you might encounter CHOPS, DOUBLES, and TRIPS...Anything over Trips would be winning with this method.
If you have saved the data from your tests I would love to see them. Even if its a 1million spin sample, I think it would help get a better idea of the probability of certain events happening. And that's all I am really looking for.
dear BigEz
three hundred spins is a satisfactiry sample for even chances and inside bets. 20 samples of 300 spins could help you to determine your heuristic approach. i have tested even chances for certain parameters which gave me identical (approx.) data.
thanks
catalyst
Read this link link:://saliu.com/bbs/messages/588.html (link:://saliu.com/bbs/messages/588.html).
Maybe you will find interesting things about 100 spins.
Thanks for that info Catalyst.
Did you happen to save any of your test data?
@donik,
I will check that out as well...thanks
QuoteIf you have saved the data from your tests I would love to see them
oops, I never do, sorry
Hi Big EZ,
As superman has pointed out, 100 spins isn't really enough, but having said that, it won't affect the AVERAGE which you can calculate like this:
0.5 × 100 = 50 singles
0.5 × 0.5 × 100 = 25 doubles
0.5 × 0.5 × 0.5 × 100 = 13 trebles (rounded up)
0.5 × 0.5 × 0.5 x 0.5 × 100 = 6 quadruples (rounded down)
etc...
ie, if the number of spins is X, to find the number of singles, doubles, trebles etc, multiply X by 0.5, 0.52 , 0.53 which correspond to singles, doubles trebles, etc.
To describe any particular distribution or bet/system, you need TWO things: the average and the standard deviation, which tells you how spread out your wins and losses will be. (ie; the dispersion).
This is what superman is referring to, so although the average doesn't change, in 100 spins the deviation from the average is likely to be much greater than if you were taking 1000 or 10,000 spins.
Note that the above is for the even chances only! if you wanted to find the number of singles, double etc in 100 dozen bets you would use the probability of a single dozen, and similarly for double streets, streets etc.
Bayes,
Thanks :thumbsup:
Now on to method design.... :twisted: