neither - even odds.

Ironically they would lose their shirt.

At 100 flips you could start knowing the odds at points of whether or not a next flip might be heads or tails. At 896 tosses the game is utterly bankrupting them

That is because of pascals triangle.

1 - 1
1 - 2 - 1
1  - 3 - 3 - 1
1 - 4 - 6 - 4 -1
1 - 5 - 10 - 10 - 5 - 1
1 - 6 - 15 - 20 - 15 -  6 - 1
1 - 7 - 21 - 35 - 35 - 21 - 7 - 1

The final row represents 128*7 tosses statistically. There is of course room for deviation, nothing perfectly fits at this level, but for gambling purposes this shall do.

In short there is one chance of heads seven times, 7 chances of heads six times and tails once, twenty one chances of heads five times and tails twice, 35 chances each for heads four/tails three and heads three/tails four, twenty one for heads twice, tails five, seven for heads once and six tails, and once for seven tails.

All youneed do is understand odds of reoccuring however. And then wait for enough stretches of heads before the current tails. Ideally you would watch for 4h 2t in a given 6 flips. This is 29 chances out of 128 that a heads will come up, or a near 1 out of 5 chance.

I have considered using this on Roulette but my memory is poor and I would have to show counting of odd/even results and red/black results as well as my efforts to identify first half and second half. Bet real low on the marginal times and medium on the good, double the bet when you lose because the odds keep getting higher that your result will happen.  The day a memory chip is implantable in the head.... I will be a millionaire (and be banned from all casinos) the following week.

That is three different variables that can skew in one direction or another... a Rainman type math guy could of course do far better... for him it would take about 1/10th the time to win.

no. you have to take into account random selection, the fact that the coin wont be tossed exactly the sam each time and that the casino will be aware of the odds you've just mentioned and do everything in their power to avoid the odds being in the punters favour -_-.

lol let me educate you
pascals triangle is a correct mathematical equation based on a perfect situation. it doesnt take into account any variables. so what your saying is correct on paper but not a definitive answer .

Of course it is not perfect. I already showed in my own example there is a 22% (irc) of the next flip being heads. But in 100 such encounters of the set up if there is 29 flips of heads and 71 results in tails you still walk away a winner.

However as the numberof flips increases then Pascals triangle becomes FAR MORE exacting. When I do 100 bit strings chances for occuring... I consider the rounding at the end of the very long decimal to be the highest possible deviation. At such points I would bet 99.999% of my cash with ease.

So your saying someone manages to get heads 100 times, his next toss is 50/50 for heads? Same for the next 5 tosses?

Shannon, Fano, Huffman, Einstein, Fibbonci, Pascal, and other mathematicians are laughing or rolling in their graves

you are misusing pascal's triangle.
pascal's shows the distribution over a population.
in other words, if I had 100 people toss a coin 7 times then
the number of people that got a certain number of heads would resemble pascal's triangle; however, for the individual  every toss is a 50/50 chance.

you are confusing population distrubution with probability.

It seems to me if you limit the number of players at a time to 5 then the House has a 50% chance of wiping out the whole table each throw.  But if it loses then it has to pay 5 players at once.

In a worst case scenario, 5 players plunk down $5 ante, let it ride 6 times and each get $32.  The House would be down $160.
But if you guys are right this would happen less than 3% of the time - because apart from the 3% odds of losing 6x in a row there is also the assumption that the same 5 players sit in.

More than 97% of the time the House should get $5.

If so I guess the next question is, why can't we make an app for this

http://www.mathsisfun.com/pascals-triangle.html

Check the coin toss section

http://www.asylum.com/2009/12/10/how-to … -up-cheat/

However it is possible to rig the game and it does seem to be a bit rigged as noted above.

The alarming display of ignorance to statistics, odds, and probabilities in this community alarms me.

Einstein:

the link you posted proves what I said.
Pascals triangle shows the distribution over a population.

Your understanding of probability is skewed.

Firefox:

Due to the heads being heavier, yes it is actually 51/49.

Idk I just posted the link.

it should be possible to flip it to impact before it stops flexing from the prior impact