coin toss entry question | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Scripts for Limit entry, Stop entry, Market entry 47 replies

TEB: Random Entry - Coin Toss Great with R.R.R. 105 replies

Coin-Toss Trading Strategy 152 replies

Toss a coin? 9 replies

coin toss entry question

  • #1
  • First Post: Jan 17, 2009 11:24pm Jan 17, 2009 11:24pm
Guys I have a question about coin toss entries.


I've read many times a coin toss will give you a 50% win rate (or very close to it) . Well this would definitely be true if you toss a coin with heads a winner and tails a loser. The theory behind this is its considered a random event with a 50% probability.

But are currency prices random events? would a coin toss entry with heads long with a tp of 1 pip sl 1 pip and the same for tails result in a 50% or near about win rate.

Would this be true?

Or is it possible if I use my coin toss entry although the toss would certainly result in a 50% or near about heads to tails ratio prices however happen to go the other way most of the time so on my 50 heads tosses when i go long prices retrace and my sl is hit and vice versa when I go short which means although my coint toss entry results in a near about 50% hit rate prices do not and my winning to loss ratio considereing a 1:1 target price to stop loss ratio actually loses most of the time.

This would mean when people say "well a coint toss entry will give you a 50% win to loss ratio" they are in fact wrong as the result of the coin toss would be near about 50% but the result of a trade entry based on it could be much much below that number.

The reason I ask is because if a coin toss entry does result in a winning trade 50% of the time all you have to do is ensure your winners are greater than your losers and you would be sure to make money.

Why bother with any form of analysis and waste time and energy as in the long run most people (not all) actually lose out and following this simple strategy would ensure not only you do not lose but in fact actually make money.
  • #2
  • Jan 18, 2009 12:02am Jan 18, 2009 12:02am
What you describe in your analysis Fugly is a perfect world where spread is 0.

Unfortunately, given the nature of spread, a coin toss does not have a 0 edge. It is slightly negative. Your application of the coin toss I feel seems to contradict itself.

You seem to be of the opinion that the market is not relatively random and does in fact have some sense of order. If that were the case, then the use of a coin toss is irrelevant, because if there is order, there are consistent patterns with 100% win rates.

Let's assume that the market has a 0 spread, is in chaos, and there is no slippage.

If I were to toss a coin (heads-long, and tails-short), then I have a 50% chance of getting either a long, or a short. Now, assume the trade has 10 pip TP and 10 pip SL. If the market was in chaos, then yes the trade has a 50% chance of winning and a 50% chance of losing. If the market had some order, then my chances of winning and losing change.

If I were to toss a coin, and trade with that, in the aforementioned scenario, there are only four possibilities:

  1. Long: Win
  2. Long: Lose
  3. Short: Win
  4. Short: Lose

I still have a 50% chance of winning, but a 25% chance of that trade being a winning long trade, and a 25% chance of that trade being a losing long trade, and a 25% chance of that trade being a winning short trade, and a 25% chance of a losing short trade.

Also, the statement "a coin toss has a 50% chance of winning" applies only if the Risk:Reward ratio is 1. Your logic assumes that the Reward / Risk is greater than 1, and that is why it is not easy (although still possible) to make money on a coin toss.

  • #3
  • Jan 18, 2009 12:16am Jan 18, 2009 12:16am
Quote
But are currency prices random events? would a coin toss entry with heads long with a tp of 1 pip sl 1 pip and the same for tails result in a 50% or near about win rate.
First, no they're not random events. There is some memory to price.

Also remember there's a spread to consider. Which means you have to run w/ TPs greater than SLs in order to make money. That means you can't run a coin toss and profit, since the odds would constantly stop you out.

Quote
The reason I ask is because if a coin toss entry does result in a winning trade 50% of the time all you have to do is ensure your winners are greater than your losers and you would be sure to make money.
A random flip market, where the ticks are sums of 50/50 flips, has an equal chance of going up 10 pips as it does down 10 pips. Think about it for a second then, how can you get bigger winners than losers if the odds of winning and losing are the same? If you set an SL of 10 pips and a TP of 20 pips, then you're odds are no longer 50/50, but 25/75. With 3/4ths of your trades losing. Add in spreads and commissions... you're going to lose money.

This means you need to enter a market w/ an edge. Where the odds of one thing happening are greater than 50%. It is the only way you can make money over time. A coin toss becomes a negative sum game.

The problem you're going to have is that the markets don't always offer this potential. Sometimes the next X (roughly a month's ATR) ticks has an equal chance of going either direction. So you not only need to know when, but how, to take trades.
  • #4
  • Edited 2:06am Jan 18, 2009 1:54am | Edited 2:06am
Quoting Ronald Raygun
You seem to be of the opinion that the market is not relatively random and does in fact have some sense of order. If that were the case, then the use of a coin toss is irrelevant, because if there is order, there are consistent patterns with 100% win rates.

I'm not of the opinion markets are or aren't random thats why I'm asking. Lots of very prominent traders, hedge fund managers, economists have different opinions and I havent found a definitive answer. I know its a very hard question to answer but you do realize there can only be one answer yes or no.

From your post I can't quite gather what you think. Do you or do you not believe markets are random.

Quote
Let's assume that the market has a 0 spread, is in chaos, and there is no slippage.
Quote
If I were to toss a coin (heads-long, and tails-short), then I have a 50% chance of getting either a long, or a short. Now, assume the trade has 10 pip TP and 10 pip SL. If the market was in chaos, then yes the trade has a 50% chance of winning and a 50% chance of losing. If the market had some order, then my chances of winning and losing change.

If I were to toss a coin, and trade with that, in the aforementioned scenario, there are only four possibilities:

  1. Long: Win
  2. Long: Lose
  3. Short: Win
  4. Short: Lose

I still have a 50%...

Thanks you're quite correct I had completely overlooked that. Lets forget the spread for a second .... just for arguments sake.


Lets also assume I only went long with a tp the same as the sl over many trades. Statistically you need a sample size of 30 before analysing for a probable outcome buts lets say i did this for many trades. Now there are only two possible outcomes right? Long win or long lose. Would I be near to a 50% win to loss ratio.

Is price action a random event?



Quote
Also, the statement "a coin toss has a 50% chance of winning" applies only if the Risk:Reward ratio is 1. Your logic assumes that the Reward / Risk is greater than 1, and that is why it is not easy (although still possible) to make money on a coin toss.
I couldnt understand this part ronald could you please elaborate a little bit.

Actually I've found someone who not only easily but effortlessly made money on a completely random entry with the sl twice the tp even after considering the spread. He generated 1311 positive and 599 negative trades over a month. traded with a 3 pip spread which he ignored in his tp and stop (meaning he started with a disadvantage). His net gain was (1311- 599*2) 112 times the limit. He set my limits at 10, loss at 20, so he earned 1120 pips very easily.

I dont know if its luck as you can see I'm still studying.
  • #5
  • Jan 18, 2009 2:06am Jan 18, 2009 2:06am
Quoting Rabid
First, no they're not random events. There is some memory to price.
Then all we have to do is look for the few consistently repeatable pattern and if we find it we'd be right 100% of the time as Ronald says.

Quote
Also remember there's a spread to consider. Which means you have to run w/ TPs greater than SLs in order to make money. That means you can't run a coin toss and profit, since the odds would constantly stop you out.

I agree with what you're saying but as I said above I've met someone who makes money on coin tosses.

Quote
A random flip market, where the ticks are sums of 50/50 flips, has an equal chance of going up 10 pips as it does down 10 pips. Think about it for a second then, how can you get bigger winners than losers if the odds of winning and losing are the same? If you set an SL of 10 pips and a TP of 20 pips, then you're odds are no longer 50/50, but 25/75. With 3/4ths of your trades losing. Add in spreads and commissions... you're going to lose money.

Rabid what if you only went long with a tp twice your sl ignoring spreads and/or commisions what would your trade win:trade loss ratio be.


I appreciate yours and ronalds help
  • #6
  • Jan 18, 2009 2:10am Jan 18, 2009 2:10am
I'm sorry I seem to have posted in the wrong section, i'd request the moderator to shift this to the rookie section
  • #7
  • Jan 18, 2009 2:19am Jan 18, 2009 2:19am
I personally don't believe the market is completely random. There are plenty of repeated patterns which I notice every so often. It's those patterns I take advantage of. In my trading, I look for regions of probability. Basically, I ask myself: "What is the likelihood of the price moving to x by y o'clock?"

Price in itself is not a random event. Remember, what causes the price to move? Simple supply and demand. The vast majority of the forex market is made up of multinational corporations who need to convert currencies. Example:

Apple makes iPods in America. Apple sells iPods in Europe. The ipods sold in Europe are paid for in Euros, right? Well, Apple can't pay their American employees in Euros, so they convert to dollars. That's what makes up the vast majority of the forex volume.

Also, with your example:

Quote

Lets also assume I only went long with a tp the same as the sl over many trades. Statistically you need a sample size of 30 before analysing for a probable outcome buts lets say i did this for many trades. Now there are only two possible outcomes right? Long win or long lose. Would I be near to a 50% win to loss ratio.

Where does the coin toss apply?
If it doesn't apply, and you're talking about straight long trades, it's not necessarily 50%. Depends on the overall trend and market direction. If you were to get a sample size of 10,000 trades, you'd be close to 50%.

The guy you're talking about as I see it probably wasn't using a coin toss as his entry =D
  • #8
  • Jan 18, 2009 2:41am Jan 18, 2009 2:41am
Quote
Then all we have to do is look for the few consistently repeatable pattern and if we find it we'd be right 100% of the time as Ronald says.
Uhm, no. There's a big difference between a coin flip and an edged pattern w/ a less than 100% expectancy.

I know for a fact that markets are not random. You can even prove it by looking at the distribution of data over time. The market does have memory... some times. I'm not really interested in theories or beliefs, I know for a fact that price is non-random some times. The ol' "floor noise" effect in action.

Let me give you an analogy... a game I call "9 or better" ok? Here's how it works. I roll a single 6 sided dice 2 times. I roll it once, then I roll it again. You can bet either before the first roll, or before the second roll, but not after the second roll. Ok? You win if you've bet and the sum of the 2 dice is 9 or greater. You lose if it's less than 9. Payout is 2:1 (matched), so if you bet $100, you get $200 back (your $100 bet plus the $100 win).

Can you, seeing this problem, now create a strategy to beat the bank? Yes, you can. Look at the distribution of probabilities. If you bet before the first roll then odds are against you. In fact if you bet whenever anything other than a 5 or a 6 rolls, the odds are against you.

If, however, you wait for a 5... then the odds are even money. If you wait for a 6, however, then you have the edge. If, then, you divide your money into tiny even amounts and bet a single one of these amounts every single time a 6 is rolled... over time you will win. The law of large numbers is in your favor.

This is how the markets work. It's a basic problem of conditional probability. Sometimes the market is showing a 1 and there's no chance of getting 9 or better. Sometimes it's showing a 6 and you need to bet. The question then is whether or not ppl can read the dice.

Quote
I agree with what you're saying but as I said above I've met someone who makes money on coin tosses.
Well you could just weigh the coin, lol. I believe a study of actual coins (quarters I think) found that because of the shape of the coin one side is slightly more likely than another. If you're a fan of football stats take a look at the coin toss there, I think heads is more likely.

Actually I believe you could do it using the central limit theorem assuming a normal distribution and a fixed mean. I think I even talked about that in a previous post, but the markets are way too complex for that to work when applied directly. Altho mean regression can be a viable edged trading method if you know how.

Quote
Rabid what if you only went long with a tp twice your sl ignoring spreads and/or commisions what would your trade win:trade loss ratio be.
A coin toss system means that every tick has an equal probability of going up or going down. That means there is identical probability of a 10 pip move up or down. A 20 pip move is no different, there is a 50/50 chance of a 20 pip move in both directions. A 50 pip move is no different, assuming a coin toss. This means that if you set a TP that's twice your SL, your SL will by definition get hit twice as often as it would've if it were twice it's size (your TP).

With a random data set if you double your TP then you cut your chances of winning in half. You're no longer working w/ a 50% expectancy at that point. A 2:1 payout with a 66/33 chance of winning is still even money, minus transaction costs... a negative sum game.
  • #9
  • Jan 18, 2009 8:34am Jan 18, 2009 8:34am
In-line

Quoting fugly
Then all we have to do is look for the few consistently repeatable pattern and if we find it we'd be right 100% of the time as Ronald says.

I never said that.


I agree with what you're saying but as I said above I've met someone who makes money on coin tosses.

[color=Red]Either he's awfully lucky or he's doing something more than a straight coin toss. Because as rabid tries to explain:

With a reward of 1 and a risk of 2, it would take two wins to equal one loss. That means when trading, we need to have a win rate...
  • #10
  • Jan 18, 2009 8:34am Jan 18, 2009 8:34am
Then all we have to do is look for the few consistently repeatable pattern and if we find it we'd be right 100% of the time as Ronald says.

I never said that.


I agree with what you're saying but as I said above I've met someone who makes money on coin tosses.

Either he's awfully lucky or he's doing something more than a straight coin toss. Because as rabid tries to explain:

With a reward of 1 and a risk of 2, it would take two wins to equal one loss. That means when trading, we need to have a win rate of 66%. Tossing a coin results in a 50% win/loss rate.

Rabid what if you only went long with a tp twice your sl ignoring spreads and/or commisions what would your trade win:trade loss ratio be.

If you increase the TP, it decreases your breakeven rate to 33%. That means only 1/3rd of your trades need to be in profit BUT, there's a problem. What is the likelihood of a trade moving 20 pips in a profitable direction vs. 10 pips in a negative direction?

Assuming the market is random, then I think you can agree that it is more likely to hit the 10 pip stoploss than the 20 pip takeprofit because the 10 pip stoploss is closer.

I appreciate yours and ronalds help
  • #11
  • Jan 18, 2009 8:45am Jan 18, 2009 8:45am
While there are two sides to a coin, i think that the third alternative hasn't been taken into account. The sideways market, entering too soon could result in a stop out or entering in completely the wrong direction would be risky.

In some ways i would consider it as gambling.
"The trend is your friend, until the end."
  • #12
  • Jan 18, 2009 8:54am Jan 18, 2009 8:54am
Quote
Either he's awfully lucky or he's doing something more than a straight coin toss. Because as rabid tries to explain...
An edge can be a funny, and very easy, thing. I've been playing around w/ using Nikkei advancer/decliners as a predictor for the Euro open, and strangely... lately, it works very well. I don't really need it, but I enjoy this business so it's fun to play around with ideas.

This is what I refer to as a "micro correlation" because current trends in the market mean that people buy dollars when the market performs poorly, and convert to euros so they can buy stocks there when the markets are doing well. I don't expect that trend to continue forever, but if the Nikkei is looking strong around the EU open... the EURUSD seems to end up with an up bias on days it decides to move, and vice-versa.

If you look at the way the DOW ended on Friday and how it's movements correlated in the EURUSD during the US session. Notice anything? LOL.

I guess what I'm saying is this. An edge can be a very simple thing. It need not be a magic bar pattern, or some tricky clever set of indicators, or even some lines drawn to represent S&R. At the end of the day the best edges are simple, and all they do is make you aware of the market's sentiment. Which is where skill comes in. Good traders pick up on the market's sentiment very quickly, poor traders do not. You can toss coins till you're blue in the face, it doesn't fix the core issue.

Quote
While there are two sides to a coin, i think that the third alternative hasn't been taken into account. The sideways market, entering too soon could result in a stop out or entering in completely the wrong direction would be risky.

In many ways a sideways market is an up, followed by a down, repeat ad nasuem. So it could be fit within a coin model, but yea... in real trading a sideways market is something to be aware of, for sure.
  • #13
  • Jan 18, 2009 9:46am Jan 18, 2009 9:46am
Quoting Rabid
If you look at the way the DOW ended on Friday and how it's movements correlated in the EURUSD during the US session. Notice anything? LOL.
Hi Rabid,

But how do you know where the DOW is going?

Or is there a time shift between the move of the DOW and the EURUSD? If yes, how much?
  • #14
  • Jan 18, 2009 10:33am Jan 18, 2009 10:33am
Quote
But how do you know where the DOW is going?
Well I'm saying if the DOW is moving up, you might not want to hunt for EURUSD shorts... lol. At least as far as this goes. I wouldn't make it a system, just find it a curious and somewhat useful correlation in the short term.

My concern mostly is improving my feel for the opening so I know what to watch for, and for that it seems for now you can watch the predecessor indexes. Ie: If the Nikkei and FTSE are/were in strong trend days perhaps it's not a good idea to range trade around the US open .
  • #15
  • Jan 20, 2009 10:06am Jan 20, 2009 10:06am
thanks for your replies, correct again he would need 66% win rate to come out on top so i suppose he hasnt used a coin toss entry he just used a random entry just went long or short without any thought on the spur of the moment.


Well however I've found someone who has made money on a coin toss entry with a 2:1 ratio live in a chat room ....... Mr. Ken Wood ...... he's published the statistics but I cant access it as for some reason I cant register on his forum (you can download the statisitcs only if ur registered)

Is it probablly luck again .... i dont know? He didnt manage the trade he set his tp twice his sl and enter long on a heads short on a tail ..... He did move his stop to BE once the trade went 8 points in his favour.

I'm basically trying to find out whether a coin fliiping entry combimed with a fixed reward:risk ratio can be profitable in the long run. you might wonder why, The reason is I find with the passage of time no matter how hard I try my win:loss ratio keeps falling and that too after a lot of effort using technical indicators and fundamental considerations and what not.

This led me to think about whether a coin flip strategy based on a fixed reward to risk ratio would actually be profitable in the market. At first i thought no its not possible but what I didnt consider is the amount risked on each trade and this makes a big difference.


I've found my coin flip strategy would result in a $50 profit (50 wins * $2 each - 50 losses * $1 each) , now if I started with say a $100 and followed the same strategy (coin flip assuming 50 heads and 50 tails - 100 tosses of the coin) but this time I risked 10% on each flip (if I lose i lose 10% and if i win I win 20% - since my risk to reward ratio is 2:1) Id end up with $4700!

If I risked 25% on each flip I'd end up with $36100! so on and so forth. I've found for this particular fixed win ratio ie 50 heads and 50 tails with a reward twice the risk , 25% is the optimum value to risk on each trade. If I increased my risk on each trade to 40% my returns after 100 tosses would drop to $4700.

So by varying the risk to reward and the amount of money risked per trade I can come out with a profit. My question is, -- can i practically implement
such a strategy?

I can vary my risk:reward and amount of money risked per trade
but will a 100 coin tosses give me a certain amount of WINS (not heads and tails - I know it will give me 50 heads and 50 tails) even if its less than 50 its alright i ust need to know its not totally unpredicatble, and I can expect say x number of wins.

Just by varying the amount risked per trade the returns shoot from $4700 to $36100!

There are three elements in this equation

1)the number of wins
2)Risk to reward
3)Amount risked on each trade

Number one is a function of number 2 and number 3.


Is it possible to come up with a combination such that it maybe profitable in the market
  • #16
  • Jan 20, 2009 10:29am Jan 20, 2009 10:29am
Quote
Is it probablly luck again .... i dont know? He didnt manage the trade he set his tp twice his sl and enter long on a heads short on a tail ..... He did move his stop to BE once the trade went 8 points in his favour.

Luck doesn't really exist. Luck is a temporary deviation from the norm, the law of large numbers cancels it out. If he's setting his TP twice his SL then his entries can't be based on a coin toss method, unless his data is for something small like 50 trades. At that point I wouldn't trust a word of what he said.

If you measure the average rate of movement for the time frame and move your SL to BE after a statistically significant movement, then that will prevent a winning trade from becoming a loser. It, however, will not turn a loser into a winner.

Quote
I'm basically trying to find out whether a coin fliiping entry combimed with a fixed reward:risk ratio can be profitable in the long run. you might wonder why, The reason is I find with the passage of time no matter how hard I try my win:loss ratio keeps falling and that too after a lot of effort using technical indicators and fundamental considerations and what not.

You can't. Just sounds like you don't have an edge. This is what happens when you don't have an edge. If for every 2 trades you win .99 and lose 1.01, you're win/loss ratio will constantly decline over time.

Fundamentals are squat in this market. This is a trader's market, volatility is too high and changes happen to fast. In good bull market you can trade on value propositions, you can't in a bear market... especially this bear market. Technical indicators are crap, they only read the market after it's happened. By then it's old news, we don't make money trading the past.

The secret is insanely simple. Edges are based on time. Some times the market wants one thing, other times it wants another. Read the context. You have to learn to read what the market wants, then wait for the market to tip it's hand... then, and only then, will you have an edge. It's the 9 or better game, you've got to wait for a 6.

Quote
If I risked 25% on each flip I'd end up with $36100! so on and so forth. I've found for this particular fixed win ratio ie 50 heads and 50 tails with a reward twice the risk , 25% is the optimum value to risk on each trade. If I increased my risk on each trade to 40% my returns after 100 tosses would drop to $4700.

Right, but that would constitute an edge.

*IF* you could know that the market was going to move twice in one direction for every once in the other direction... well that's the definition of a trend. You're saying that IF you could guess a trend in advance then you'd make money. Well yea, of course you could. You've discovered the magic of a trend breakout. Congrats .

But that's not a coin flip, either. A coin flip is 50/50 for equal chances at both directions. If you double the distance in one direction then you need to double the distance in the other, if you don't, then your odds are no longer 50/50.

In other words for your scenario to play out you would need an edge that gives you a higher than 50/50 chance of something happening. You'd need a 70% chance, basically, just to get a 50/50 chance of a 2:1 return.

Quote
Number one is a function of number 2 and number 3.

That's where sooo many people get hosed in this business. The whole "but I can change my trading volume." Well yes, you can, but unless you have an edge that tells you when to increase it profitably... over time it will only average out to random.
  • #17
  • Jan 20, 2009 11:24am Jan 20, 2009 11:24am
Rabid, I don't think fugly understands our explanations.

Fugly, go and get yourself a demo account, and TRY IT. Clearly you don't seem to be willing to listen to Rabid or myself when we keep telling you that the odds of a coin toss is 50%, and the chances of a win/loss depends on your risk:reward ratio. Let me make it clear.

Your win/loss ratio in a coin toss is only 50% if your risk:reward ratio is 1.
  • #18
  • Jul 21, 2009 7:44pm Jul 21, 2009 7:44pm
Quoting Ronald Raygun
Rabid, I don't think fugly understands our explanations.

Fugly, go and get yourself a demo account, and TRY IT. Clearly you don't seem to be willing to listen to Rabid or myself when we keep telling you that the odds of a coin toss is 50%, and the chances of a win/loss depends on your risk:reward ratio. Let me make it clear.

Your win/loss ratio in a coin toss is only 50% if your risk:reward ratio is 1.
You are mixing apples, oranges and pears.

Coin toss win/loss ratio is 50/50 for coin tosses.

Coin toss win/loss ratio is not 50/50 for trade entries if you enter long for heads and short for tails. If you enter long for heads and DO NOT ENTER for tails, now you will be closer to 50/50 because when you get heads you will either win or lose. When you also enter short on tails, the possibility exist that you always lose.

Win/Loss ratio does not have Risk/Reward as part of the equation. Expectancy does. Dp not confuse them. They are not the same.
  • #19
  • Jul 21, 2009 8:37pm Jul 21, 2009 8:37pm
Quote
Win/Loss ratio does not have Risk/Reward as part of the equation. Expectancy does. Dp not confuse them. They are not the same.
Generally yes, but in this specific case, no. It's not "go long on heads" and "go short on tails" but rather this...

Consider a person that flips a coin every second. Heads, it's an up tick. Tails it's a down tick. In that case you could generate an entire day's price chart with 86,400 flips. A week with 604,800 flips, a month with approx 2.6m flips. That would create a bar, line or candle chart that would look exactly like a market chart, but would be completely random.

In this situation, the odds of price moving any one direction X ticks is the same as price moving the other direction X ticks. Ie: the odds of price moving 50 pips up is exactly the same as the odds of price moving 50 pips down. In a situation like that, equal distances have equal odds, inverse win/loss and risk/reward ratios. If you then try to get a 2:1 return rate, you cut your win/loss in half. Why? Because if you now use a 100 tp and a 50 sl, instead of a 50 tp and a 50 sl, and the odds of equal moves are the same, then the odds of you hitting a 100 tp is half as likely as the odds of you hitting a 50 sl.

This means that for every increase in RR, you decrease your WL. If you decrease your RR, you will increase your WL, but it will only offset the decrease in RR.

The trick to a random market is knowing there's a fixed mean. If you were to go over each tick and sum up the number of heads and tails, then wait for one to be 2 deviations from the norm, you could trade that as a "regression to the mean" since you know that it will eventually regress back to equal numbers of heads and tails. In a situation like this you would not want a stop loss, since you know with statistical certainty that it will regress, but you do not know what the variance will be... ie: price might move a billion pips between now and then, but it will eventually get there.

The problem with that in the real market, tho, is obvious. Nobody has infinite leverage and an infinite time to wait for the regression. To make matters worse, the mean is not fixed and the distribution is not a clean bell curve. It's skewed with a "long tail" altho the actual distribution varies over time (the distribution itself has an average, but there's a lot of variance there too, making traditional stochastic risk models very inaccurate over time). This actually proves that the market is not random. Basically you have a kind of cumulative chaos factor, a kind of "sudden, pseudo-random butterfly effect" that occurs because of the way people trade. That's why I subscribe to the "auction market theory" as it explains this effect perfectly in human terms.

Back to my point tho, in a random market R/R is an inverse of W/L. This is why you need "an edge" to make money in the markets. Edges are very simple creatures tho, and even the best edge in the world can be ruined by bad MM (or psychology).
  • #20
  • Jul 25, 2009 5:43pm Jul 25, 2009 5:43pm
Your reply was very well put. We are not using risk/reward the same way. I understand what you are saying.
0 traders viewing now