Probability Paradox!?! | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Parrondo's Paradox: A Study 9 replies

MTS Vs Random Entry - A Paradox 6 replies

Elliott wave paradox? 3 replies

Question : is this a paradox? 28 replies

Newcomb’s paradox revisited? (Trading context) 2 replies

Attachments: Probability Paradox!?!
Exit Attachments

Probability Paradox!?!

  • #21
  • Jan 15, 2012 2:28pm Jan 15, 2012 2:28pm
Quoting Noloqy
Ok, if this is a genuine question, the answer is that it doesn't matter. Your chance of success will always be 50%. Sticking with heads, sticking with tails, switching every round, switching every other round - it's all the same.
O OK. I really thought it would change the probability.
  • #22
  • Jan 15, 2012 2:37pm Jan 15, 2012 2:37pm
Quoting Agro
No 'selection method' will improve the odds. It is a 50/50 chance on a coin flip no matter what. No but's about it. If you are talking about a casino game (such as roulette) then the house have a built in edge (the 0). It is a mugs game and you will always lose in the long run.
Oops, let me explain that a bit better, I meant the bet selection ON 50/50 chances, as roulette you suggested, not that this bet would improve odds.

Now did you fully understand the method? By betting on the second previous result, you will winning every time there is a streak, after the first 2 results, every bet made will be a winner.

So in the aforementioned streak of 14, you would win 12 times.

This is about bet selection, not changing the odds, follow the trend, not pick the bottom as most seem inclined to do by human nature.
It takes a lot for me to IGNORE you, but you can try.
  • #23
  • Jan 15, 2012 2:58pm Jan 15, 2012 2:58pm
Quoting Erebus
Oops, let me explain that a bit better, I meant the bet selection ON 50/50 chances, as roulette you suggested, not that this bet would improve odds.

Now did you fully understand the method? By betting on the second previous result, you will winning every time there is a streak, after the first 2 results, every bet made will be a winner.

So in the aforementioned streak of 14, you would win 12 times.

This is about bet selection, not changing the odds, follow the trend, not pick the bottom as most seem inclined to do by human nature....
Random is Random

a Random streak of tails still has a probability of 50% being heads next toss.
there are no trends in Random data, all the "trends" are RANDOM, and only hindsight can you see the streak which says nothing as the continuation thereof as far as i understood.
AVT INVENIAM VIAM AVT FACIAM
  • #24
  • Jan 15, 2012 3:17pm Jan 15, 2012 3:17pm
Quoting the redlion
Random is Random

a Random streak of tails still has a probability of 50% being heads next toss.
there are no trends in Random data, all the "trends" are RANDOM, and only hindsight can you see the streak which says nothing as the continuation thereof as far as i understood.
No one has dispute this, your reply has no bearing on my statement and certainly no relevance that I can see
It takes a lot for me to IGNORE you, but you can try.
  • #25
  • Jan 15, 2012 3:25pm Jan 15, 2012 3:25pm
Quoting Erebus
No one has dispute this, your reply has no bearing on my statement and certainly no relevance that I can see
same goes with your roulette strategy if in fact it's random, all strategies are bound to fail as the next individual event will respect no streak.

random streaks are random

a streak by definition is not.

hopefully i make sense
AVT INVENIAM VIAM AVT FACIAM
  • #26
  • Jan 15, 2012 4:28pm Jan 15, 2012 4:28pm
Trouble is a coin flip is random with 50/50 chance. But is what you are betting against random too? Is forex random? Is a trend a trend or a series of random events that just happen to be the same?

Not looking for an answer. It's a hypothetical question
  • #27
  • Jan 15, 2012 4:39pm Jan 15, 2012 4:39pm
I just quickly put together a spreadsheet that will create random candlestick stock charts. Well, it's a little more cumbersome than that thanks to excel. Basically on the master input sheet you can change the randbetween function values. Just change the value of the first cell for example (=randbetween(-25,25). Then delete the existing cells down the column and re-autofill the column with the new =randbetween formula.

To make a new chart, select the the cell I1 and then start the auto filter and filter by non blanks. Copy the OHLC data starting in column I and then paste it into a new worksheet. Once in the new worksheet you can select the data and play around with weighting the market in different ways. I'd suggest using value > 100 so that the weights you apply are more subtle. For instance, -150,152, this will give a more gentle bias than something like -3, 7.

The existing chart I have going in chart 6 is pretty convincing...

In Adal's recent posts you can find where he talks about the crux of this random business once we begin to look at real price data. I'm not motivated enough at the moment to post links because my daughter is watching the classic "Three Ninjas" and I just heard a loud "I-YA!" come from the living room...Never a good sign...
Attached File(s)
File Type: xls Forex Factory Random.xls   2.4 MB | 500 downloads
  • #28
  • Jan 15, 2012 4:41pm Jan 15, 2012 4:41pm
and then newyear made it rain fire for seven days and seven nights...

I guess it was a good question since i'm not the only one who gets confused.. basically no matter what you do the coin still has two sides and your next flip is random.. thats what I've always settled on...

I got confused by streaks which states in a series of 10 flips.. there is an 86ish% chance of it landing heads two times in a row during that 10 flips

when we step into market logics it really doesn't matter tho because the market is continues so to start questioning streaks in a series is intense.. the market is continues and relatively infinite (aka not a process of just 10 times)

forex quants way of thinking makes more since for an infinitely continuous non stoping cycle which if given 50% odds you'd have 0.5x0.5 = 25% chance of that outcome

makes more sense to me atleast...
  • #29
  • Jan 15, 2012 4:55pm Jan 15, 2012 4:55pm
Alright, nothing in the living room is broken. By weighting the random data we can achieve a "net-flow" much like in the real market. One simple principle that undercuts the random market idea is the fact that it's a normal market procedure to split large orders into little chunks over a day or whatever time frame is required. This gives us a baseline confidence in pervasive/persistent moves in price (trends). Again the mechanism is net flow, and it may be random in the sense that the world is random and the net flow happens to be negative the Euro today or vice verse. What I'm getting at is that there are certainly times that one can to some extent predict what price will continue to do based on what it is doing. That or wait for times where the market becomes fast and employ some simple momentum strategies. If the market is prone to sharp moves and you know that you are operating in one of those periods, enter with stop orders, employ tight stops along with a trailing stop. The market is random as you can easily read in the academic literature, but at times it is definitely predictable. Yep, the paradox of price.
  • #30
  • Jan 15, 2012 5:00pm Jan 15, 2012 5:00pm
One more passing thought that I've come to appreciate:

It's take's a loaded freight train a while to stop.
  • #31
  • Jan 15, 2012 7:12pm Jan 15, 2012 7:12pm
Quoting newyear498
and then newyear made it rain fire for seven days and seven nights...

I guess it was a good question since i'm not the only one who gets confused.. basically no matter what you do the coin still has two sides and your next flip is random.. thats what I've always settled on...

I got confused by streaks which states in a series of 10 flips.. there is an 86ish% chance of it landing heads two times in a row during that 10 flips

when we step into market logics it really doesn't matter tho because the market is continues so to start questioning...
Actually I had a similar thought like you before because I tried to look at such a simple problem in different angle. I was damn confuse too.

People may ask why you complicated such a simple problem. Everyone knows that it is 50% of chances. In fact we just tried to think and interpret the problem in different way, a "what if" way. I believe such an idiot process is actually good because it improves our understanding and train us to think outside the box.

Well I guess I have leave the room for you guys, war is gonna to begin soon. I better run.
  • #32
  • Jan 15, 2012 8:42pm Jan 15, 2012 8:42pm
Quoting Agro
If you are talking about a casino game (such as roulette) then the house have a built in edge (the 0). It is a mugs game and you will always lose in the long run.
The '0' is not the house's edge and a myth. The ball can only land in one slot, so that is 1 in 36 (using UK roulette), which means a probability of you winning is only 3%.

Worded differently, you have a 97% chance of losing, which means the house has a 97% chance of winning. Saying the house has an edge gives the impression they only have a small advantage, when in reality they have a certainty.

In regards to the OP's opening post, it's coincidence that you get two in a row as the probability remains at 50/50 each time. Do enough coin flips and count the times you get two in a row and you'll realise it's insignificant.
Good judgement comes from experience. Experience comes from bad judgement.
  • #33
  • Jan 15, 2012 8:54pm Jan 15, 2012 8:54pm
Quoting Reactor
The '0' is not the house's edge and a myth. The ball can only land in one slot, so that is 1 in 36 (using UK roulette), which means a probability of you winning is only 3%.

Worded differently, you have a 97% chance of losing, which means the house has a 97% chance of winning. Saying the house has an edge gives the impression they only have a small advantage, when in reality they have a certainty.

In regards to the OP's opening post, it's coincidence that you get two in a row as the probability remains at 50/50 each time. Do enough coin flips...
I think you mean 1 in 37 ???
Attached Image (click to enlarge)
Click to Enlarge Name: roulette01.png Size: 434 KB
It takes a lot for me to IGNORE you, but you can try.
  • #34
  • Jan 15, 2012 9:09pm Jan 15, 2012 9:09pm
Quoting Erebus
I think you mean 1 in 37 ???
I stand corrected, 1 in 37 it is!
Good judgement comes from experience. Experience comes from bad judgement.
  • #35
  • Jan 15, 2012 10:58pm Jan 15, 2012 10:58pm
Quoting Reactor
The '0' is not the house's edge and a myth. The ball can only land in one slot, so that is 1 in 36 (using UK roulette), which means a probability of you winning is only 3%.

Worded differently, you have a 97% chance of losing, which means the house has a 97% chance of winning. Saying the house has an edge gives the impression they only have a small advantage, when in reality they have a certainty.

In regards to the OP's opening post, it's coincidence that you get two in a row as the probability remains at 50/50 each time. Do enough coin flips...
Nope, 0 is the edge. You are forgetting that the casino pays 36 times your bet if you win. If there were only 36 numbers then there would be no edge, but since there are 37 they have a slight edge (about 2.7%). If you bet on red or black then you have 18 red and 18 black slots and the 0. Total slots are 37 so you have an 18 in 37 chance winning which is about a 48.6% chance. The casino knows it only needs a very tiny edge to win so long as they are sufficiently capitalised and cap bets according to their capitalisation in case of a losing streak.

Incidentally, roulette is one of the larger edges as far as the casino is concerned. If one was to play proper blackjack strategy, depending on the house rules, the edge that the casino holds can be taken down to around 1%. Of course most people do not play proper strategy so their practical edge is in fact much larger.
  • #36
  • Edited 1:31am Jan 16, 2012 12:33am | Edited 1:31am
Quoting Scotty B
Alright, nothing in the living room is broken. By weighting the random data we can achieve a "net-flow" much like in the real market. One simple principle that undercuts the random market idea is the fact that it's a normal market procedure to split large orders into little chunks over a day or whatever time frame is required. This gives us a baseline confidence in pervasive/persistent moves in price (trends). Again the mechanism is net flow, and it may be random in the sense that the world is random and the net flow happens to be negative the...
Fine post Scotty, great to see you posting here. Happy New Year, my friend, and all the best best for 2012.

Scotty's post outlines important distinctions and implications. He talks about weighting random data. However, unweighted randomly generated charts (i.e. where there is always exactly 50% probability of an uptick, and 50% downtick) may look very similar to price charts, but there are subtle but significant differences. I've read comments from folk who believe that the randomly generated charts are beatable by using different kinds of strategies, including "cut losses; let profits run". However, probability theory suggests otherwise. Assuming that the RNG is generating a perfectly random sequence, every possible strategy must necessarily have an expectancy of zero. Just as it would be if you were to play roulette without the zero (see Agro's post for roulette info).

However, as Scotty says, there are times when "price is definitely predictable", at least on balance of probability, making it possible to devise a system of entries and exits that tips the balance in the trader's favor. For example, to whatever extent "trends" (and by that, I mean trends that are "weighted", i.e. qualitatively different from those produced by the RNG) exist, a "cut losses; let profits run" strategy becomes viable. Of course --- just like the 0 on the roulette wheel --- the trader's strategy must overcome the forex equivalent of the "house edge", namely transaction costs (spreads, commissions, swap differential, etc).

A word about MM. Just as you can't use a staking system to beat roulette, you can't use a position sizing system in itself to beat the market. Having a system that's founded in price inefficiencies is prerequisite to having any kind of an 'edge'.

Finally (and getting off-topic, sorry), threads like this one disturb me. Not that there's anything wrong with the material that's being presented --- it goes without saying that any edge is eroded to the extent that the trader is unable to implement it --- but the heavy emphasis on psychology can create the impression that even random entries and exits can work, provided that the trader gets his head on straight. I tend to side with those whose focus is aimed at the prerequisite 10,000 (or whatever) hours of chart study. Perfect discipline is meaningless if the trader doesn't already have a profitable plan to remain disciplined to.

David
  • #37
  • Jan 16, 2012 1:18am Jan 16, 2012 1:18am
Quoting hanover
Having a system that's founded in price inefficiencies is prerequisite to having any kind of an 'edge'.
  • #38
  • Jan 16, 2012 6:30am Jan 16, 2012 6:30am
Quoting Agro
Nope, 0 is the edge. You are forgetting that the casino pays 36 times your bet if you win. If there were only 36 numbers then there would be no edge, but since there are 37 they have a slight edge (about 2.7%).
You haven't done the maths correctly. If it was only 36 slots (1/36 = 2.78% probability of winning) the probability is still 97% that you will lose. Having an extra slot does not increase the casino's edge by 2.7% either (1/37 = 2.70% probability of winning, not 2.78%-2.70% = 0.08%). Let's be clear here, it's the 97% that is the real reason for the casino's advantage, not the 3%. There is no gimmick with the 0, it's just obvious maths.

Don't worry, your behaviour reminds me when they first introduced the National Lottery in the UK. Everyone thought it was simple to pick 6 numbers from the 49 and they would be millionaires. They didn't understand what 1 in 14 million actually meant. But just in case you or anyone else is not aware, it means the chances of you losing are 99.99999%. Of course someone can win, but it's unlikely to be you.
Good judgement comes from experience. Experience comes from bad judgement.
  • #39
  • Jan 16, 2012 7:12am Jan 16, 2012 7:12am
Quoting Reactor
You haven't done the maths correctly. If it was only 36 slots (1/36 = 2.78% probability of winning) the probability is still 97% that you will lose. Having an extra slot does not increase the casino's edge by 2.7% either (1/37 = 2.70% probability of winning, not 2.78%-2.70% = 0.08%). Let's be clear here, it's the 97% that is the real reason for the casino's advantage, not the 3%. There is no gimmick with the 0, it's just obvious maths.

Don't worry, your behaviour reminds me when they first introduced the National Lottery in the UK. Everyone thought...
I am sorry but you are just wrong. Simple maths bores me and this topic does not interest me enough so I am not going to get too involved, but the 0 is the casino's edge and if it were not there then roulette would be an even money game. As I say, you are ignoring the fact that if you win picking a number you are paid 36 times your bet. A casino's edge is very small on all the games they offer, they don't need a 97% edge when there is so much money going through the place every day and there sample sizes are large enough to get past statistical anomolies
  • #40
  • Jan 16, 2012 7:13am Jan 16, 2012 7:13am
Quoting Reactor
You haven't done the maths correctly. If it was only 36 slots (1/36 = 2.78% probability of winning) the probability is still 97% that you will lose. Having an extra slot does not increase the casino's edge by 2.7% either (1/37 = 2.70% probability of winning, not 2.78%-2.70% = 0.08%). Let's be clear here, it's the 97% that is the real reason for the casino's advantage, not the 3%. There is no gimmick with the 0, it's just obvious maths.

Don't worry, your behaviour reminds me when they first introduced the National Lottery in the UK. Everyone thought...
i don't know about the specific maths, but this doesn't sound right at all....

the easy case. betting on red or black is a 50/50 and will get no one anywhere over time. the zero/s are an automatic win for the house. edge. period. it's no longer a 50/50 for the player, it's sub 50 / sub 50.

the harder case. betting on individual numbers. just because someone is almost guaranteed to lose doesn't matter. the winners get paid handsomely for being lucky. this would create, again, a neutral payout situation over time. again, the zero/s tip the odds towards the house having an automatic win. edge.

fuck the math.... if you have an even payout game then add one extra element that favors one side then that is an edge and where the money goes.
Congratulations ScalaFX for coming out of the closet!
0 traders viewing now