what is the square root of ??
Thread Rating:
I believe as an example the square root of 2000 is about 44.
Thus 1 divided by 44 is about 2 %.
The 2% indicates an expected standard deviation for the next 2000 plays.
In Other words if you won 49% of the events in the first 2000 events in the next
2000 events you should win anywhere from 47 to 51% of the time.
Am I right ? am I close to right ? Am i Totally mistaken ?
Thanks
Quote: RomesIrregardless of your math, you're dipping in to the Gamblers Fallacy. Even if the math dictates you're expected to win 50% of the events, just because you won 25% the first 2000 plays does NOT mean you should expect to win 75% of the next 2000 plays to "balance out." Especially if it's something like roulette or craps in which each event is an independent trial.
I think it is You That might be mistaken ? I think you are confusing an A OR B out come with an out come that has more variables.
Example you toss a coin it can only be A head or B tails. In Craps if you are looking for specfic numbers like ONLY 4 OR 10 then the
math would get far more complex And Each throw of the Dice would have Many Variables.
If you toss 2000 coin tosses and you get 75% heads, that DOES NOT mean that in the next 2000 coin tosses you can expect 75% tails, to "balance" out. This is literally the Gamblers Fallacy and I'll implore you to do a bit more research on that.
Quote: gary55I think it is You That might be mistaken ? I think you are confusing an A OR B out come with an out come that has more variables.
Example you toss a coin it can only be A head or B tails. In Craps if you are looking for specfic numbers like ONLY 4 OR 10 then the
math would get far more complex And Each throw of the Dice would have Many Variables.
What would those extra variables be? there's still only 3 possible outcomes: 4, 10, other. And, the dice still don't remember the previous rolls. So the future odds are not impacted by the past results.
Anyway the other out comes could be 5 6 7 8 9 etc etc....even if it were only 3 outcomes if moves
a 50/50 situation to a 1/3 or 2/3 depending on what you are saying which would be a Huge Diff
Quote: gary55The 2% indicates an expected standard deviation for the next 2000 plays.
In Other words if you won 49% of the events in the first 2000 events in the next
2000 events you should win anywhere from 47 to 51% of the time.
Standard deviation is from the expected mean - not from the previous results.
Quote: gary55I think we are getting away from the orginal question.
Anyway the other out comes could be 5 6 7 8 9 etc etc....even if it were only 3 outcomes if moves
a 50/50 situation to a 1/3 or 2/3 depending on what you are saying which would be a Huge Diff
No, no it doesn't move the probability to 1/3 or 2/3 just because there are only 3 possible outcomes. That would only make sense if the outcomes were equally likely. Either I'll win the powerball tomorrow or I won't. I don't have a 50% chance of winning the lottery. Out of 36 possible combinations on 2 6 sided dice, there are 3 ways to make a 4 and 3 ways to make a 10. There are 30 ways to make "other". So the probabilities in our example are 3/36, 3/36, and 30/36...not 12/36, 12/36, 12/36...
Romes good math guy, this I admit.Quote: RomesThen let's take your example, for example =)...
If you toss 2000 coin tosses and you get 75% heads, that DOES NOT mean that in the next 2000 coin tosses you can expect 75% tails, to "balance" out. This is literally the Gamblers Fallacy and I'll implore you to do a bit more research on that.
But, eventually the numbers do reach where they should be, and this is Rome's basis for most discussion here.
2000 is no where near 'eventually'. Need lots more zeroes..... ;-) Just 2F
Here's the truth:Quote: TwoFeathersATLBut, eventually the numbers do reach where they should be, and this is Rome's basis for most discussion here.
a) Eventually the percentages converge on where they should be. This is the law of large numbers.
b) The actual numbers, in absolute terms, actually run away from "where they should be." Believing the opposite is "the Gambler's Fallacy."
It's a subtle difference, but an important one. It's not a contradiction to realize that after 10 coin flips you might be at 4 heads and 6 tails (40% heads and a difference of 2 flips or 20%), while after 10,000,000 coin flips you might be at 4,998,973 heads and 5,001,027 tails (49.99% heads and a difference of 2054 flips or 0.02%). 0.02% is far smaller than 20%, but 2054 is far larger than 2. See what's going on there? The Gambler's Fallacy says that the actual difference (2054) should trend toward zero. That's false.
Quote: MathExtremistHere's the truth:
a) Eventually the percentages converge on where they should be. This is the law of large numbers.
b) The actual numbers, in absolute terms, actually run away from "where they should be." Believing the opposite is "the Gambler's Fallacy."
It's a subtle difference, but an important one. It's not a contradiction to realize that after 10 coin flips you might be at 4 heads and 6 tails (40% heads and a difference of 2 flips or 20%), while after 10,000,000 coin flips you might be at 4,998,973 heads and 5,001,027 tails (49.99% heads and a difference of 2054 flips or 0.02%). 0.02% is far smaller than 20%, but 2054 is far larger than 2. See what's going on there? The Gambler's Fallacy says that the actual difference (2054) should trend toward zero. That's false.
I'm about to step over the line, I see red in my future...
From my perspective, ME, you just agreed with me, which is funny in and of itself;-)
I understand the difference between 20% and 0.02%, when described in the context you laid forth.
Now I'm laughing.
I will prolly not come back here, this thread, I've seen you rip people to shreds, you're very good at it, my compliments, very good at it. Just 2F
Honestly, who cares?Quote: TwoFeathersATLI'm about to step over the line, I see red in my future...
Prolly no one but me you seriously evil looking dark sinister character carrying a wheat/grain scythe in your hand, Prolly no one. ( just referring to your Avatar, I should add).Quote: TheGrimReaper13Honestly, who cares?
I think we met before, where was that Casino anyway, I forget ;-)
The Casino Cornflakes, of course. Lol.Quote: TwoFeathersATLI think we met before, where was that Casino anyway...
I do agree with you -- it's just a question of being precise. You said "eventually the numbers will reach where they should be" and that's true for percentages but not for absolute trials. Understanding that difference is the key to not falling prey to the Gambler's Fallacy.Quote: TwoFeathersATLFrom my perspective, ME, you just agreed with me, which is funny in and of itself;-)
I understand the difference between 20% and 0.02%, when described in the context you laid forth.
almost all in the whole universe says this "the NUMBERS"Quote: MathExtremistYou said "eventually the numbers will reach where they should be"
and they mean the actual numbers, just like losing the EV the more you play, actual $$$ lost, NOT percentages.
only I, ME, BruceZ and AlanS know the truth and speak up
the others that due know the truth say nothing (like my husband)
and let all believe that you are
more likely
to have equal NUMBERS of Heads and Tails after more and more flips.
they should read more about what William Feller wrote about
He even mentioned the math folks also struggled with simple coin flipping concepts
funny
a
coin flip
you are in a small minorityQuote: MathExtremistand that's true for percentages but not for absolute trials. Understanding that difference is the key to not falling prey to the Gambler's Fallacy.
an example of the law of large numbers is my coin flip pic
the 1st 1000 flips produced 800 Heads and 300 more than what SHOULD have been.
now afte 1 million total flips the 300 excessive HEADS NEVER equaled out, NEVER!!
but the % of Heads sure looks pretty (Sally pretty)
close to 50%
Sally pretty
is this a trick-trick question?Quote: gary55Does anyone know what the square root of variance is ?
looks like it is from what follows
the answer is the standard deviation
(called that here in the USA)
http://www.mathsisfun.com/data/standard-deviation.html
so we use 45Quote: gary55I believe as an example the square root of 2000 is about 44.
noQuote: gary55In Other words if you won 49% of the events in the first 2000 events in the next
2000 events you should win anywhere from 47 to 51% of the time.
Am I right ?
noQuote: gary55am I close to right ?
here againQuote: gary55Am i Totally mistaken ?
Thanks
http://www.mathsisfun.com/data/standard-deviation.html
some like to use the normal distribution (but no no how)
some dun't
(cuz they no understand it)
fun question
good luck!
Quote: MathExtremistSee what's going on there?
I see what you did there!*
*I just realized the young people are using that expression and I felt a need to use it too, to show I am still hip (an expression I doubt they use)
