10 Heads In A Row

they are trying to assert that [...] if there have been 10 heads, then the next in the sequence will more likely be a tail because statistics says it will balance out in the end

There'south merely a "balancing out" in a very particular sense.

If information technology'south a fair coin, and so it's still fifty-l at every toss. The coin cannot know its past. Information technology cannot know at that place was an excess of heads. It cannot compensate for its past. Ever. it only goes on randomly existence heads or tails with constant chance of a caput.

If $n_H$ is the number of heads in $n=n_H+n_T$ tosses ($n_T$ is the number of tails), for a fair coin, $n_H/n_T$ will tend to 1, every bit $n_H+n_T$ goes to infinity .... but $|n_H-n_T|$ doesn't go to 0. In fact, it also goes to infinity!

That is, goose egg acts to make them more even. The counts don't tend toward "balancing out". On boilerplate, imbalance between the count of heads and tails actually grows!

Here'due south the effect of 100 sets of 1000 tosses, with the grey traces showing the difference in number of head minus number of tails at every step.

The grey traces (representing $n_H-n_T$) are a Bernoulli random walk. If yous think of a particle moving up or downwardly the y-axis by a unit stride (randomly with equal probability) at each fourth dimension-step, and then the distribution of the position of the particle will 'diffuse' away from 0 over fourth dimension. It still has 0 expected value, but its expected distance from 0 grows as the square root of the number of time steps. [Annotation for anyone thinking "is he talking well-nigh expected absolute difference or the RMS deviation" -- really either: for large $north$ the get-go is $\sqrt{2/\pi}\approx$ 80% of the 2d.]

The blue curve to a higher place is at $\pm \sqrt{n}$ and the green bend is at $\pm 2\sqrt{northward}$. As you lot see, the typical distance between total heads and total tails grows. If at that place was anything acting to 'restore to equality' - to 'make upward for' deviations from equality - they wouldn't tend to typically grow farther autonomously like that. (It's not hard to testify this algebraically, just I doubt that would convince your friend. The critical office is that the variance of a sum of independent random variables is the sum of the variances $<$run into the stop of the linked department$>$ -- every time you add some other money flip, y'all add a constant corporeality onto the variance of the sum... so variance must grow proportionally with $n$. Consequently the standard difference increases with $\sqrt{northward}$. The constant that gets added to variance at each step in this case happens to be i, but that's not crucial to the argument.)

Equivalently, $\frac{|n_H-n_T|}{n_H+n_T}$ does go to $0$ as the total tosses goes to infinity, simply merely considering $n_H+n_T$ goes to infinity a lot faster than $|n_H-n_T|$ does.

That ways if nosotros carve up that cumulative count by $n$ at each pace, information technology curves in -- the typical absolute difference in count is of the order of $\sqrt{n}$, only the typical absolute deviation in proportion must then exist of the order of $one/\sqrt{n}$.

That's all that's going on. The increasingly-large* random deviations from equality are only "washed out" by the even bigger denominator.

* increasing in typical absolute size

Run across the footling blitheness in the margin, here

If your friend is unconvinced, toss some coins. Every time you lot get say 3 heads in a row, get him or her to nominate a probability for a head on the next toss (that'south less than 50%) that he thinks must exist fair by his reasoning. Ask for them to requite you the corresponding odds (that is, he or she must be willing to pay a bit more than one:1 if you bet on heads, since they insist that tails is more likely). It'south best if it'southward set equally a lot of bets each for a pocket-size corporeality of coin. (Don't be surprised if in that location's some excuse as to why they can't take up their half of the bet -- but it does at to the lowest degree seem to dramatically reduce the vehemence with which the position is held.)

[Withal, all this discussion is predicated on the coin existence fair. If the coin wasn't fair (fifty-50), and so a different version of the word - based around deviations from the expected proportion-difference would be required. Having 10 heads in 10 tosses might make you suspicious of the supposition of p=0.5. A well tossed money should be close to off-white - weighted or not - but in fact nonetheless exhibit small but exploitable bias, especially if the person exploiting it is someone like Persi Diaconis. Spun coins on the other hand, may be quite susceptible to bias due to more weight on one face.]

10 Heads In A Row,

Source: https://stats.stackexchange.com/questions/136870/does-10-heads-in-a-row-increase-the-chance-of-the-next-toss-being-a-tail

Posted by: ruthclowboulat.blogspot.com

Share this post