Long term Profits within Negative Expectancy | Page 11 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

What is Expectancy? 13 replies

System Expectancy - Monte Carlo Simulation 5 replies

Short Term / Medium Term / Long Term? 3 replies

Random Market vs Set Expectancy 8 replies

Expectancy is always zero 41 replies

Attachments: Long term Profits within Negative Expectancy
Exit Attachments

Long term Profits within Negative Expectancy

  • #201
  • Aug 23, 2009 12:13am Aug 23, 2009 12:13am
Quoting Chicky
Neither I. I was just trying to say that some of our trading tools seem to work in pure gambling situations.

For example, I would have played the following two trendline breaks with 1:1 risk /reward .......... and interestingly both of these "trades" would have been a success

Next time, I will try this in a Casino

Come on Chicky, thats it. I was hoping for something more. I'm going to bed.

Good luck bud, trade between the lines.
  • #202
  • Aug 23, 2009 12:15am Aug 23, 2009 12:15am
Quoting knuckledrag
Come on Chicky, thats it. I was hoping for something more. I'm going to bed.

Good luck bud, trade between the lines.
You too. Later.
The Thief of Wall Street
  • #203
  • Aug 23, 2009 12:43am Aug 23, 2009 12:43am
Quoting Chicky
Neither I. I was just trying to say that some of our trading tools seem to work in pure gambling situations.

For example, I would have played the following two trendline breaks with 1:1 risk /reward .......... and interestingly both of these "trades" would have been a success

Next time, I will try this in a Casino
Chicky,

Thanks for explaining what you meant. I am an extremely sarcastic person by nature and I sensed you were only being sarcastic in your responses. Could you apply simple charting to gambling? Sure you could. I can't imagine how in the world you'd record data and do analysis in such a setting though. Unless you had one heck of a memory and could chart in the mind.

If you wanted to take it even further you could start applying things like point and figure type analysis that could possibly gain you even more of an edge (in terms of filtering out noise).

As far as work towards positive expectancy in trading, I think one needs to start looking for both entries and exits in raw S/R areas. Price trades all day long to and from simple S/R. Take advantage of it, theres no need to re-create the wheel. Most traders see S/R work, but because of it's sheer simplicity they reject it and think that success in trading is something that needs to be extremely difficult and so go off on the futile search for a holy grail only to come back to the basics after they tire of failiure.
  • #204
  • Aug 23, 2009 1:00am Aug 23, 2009 1:00am
daytrading,

thank you very much for your explanation and I will look into those 3 areas you mentioned which are profitable.
at the moment I`m spending a lot of time backtesting the first 4H of the week BO on G/Y.

Petra you might want to look at this system by Ace as well, `cos its definately profitable. I do have a feeling that your way of testing systems is too rigid, because volatility changes over the years and you will have to adapt your parameters.
  • #205
  • Edited 1:57am Aug 23, 2009 1:40am | Edited 1:57am
Quoting Scotty B
Chicky,

Thanks for explaining what you meant. I am an extremely sarcastic person by nature and I sensed you were only being sarcastic in your responses. Could you apply simple charting to gambling? Sure you could. I can't imagine how in the world you'd record data and do analysis in such a setting though. Unless you had one heck of a memory and could chart in the mind.

If you wanted to take it even further you could start applying things like point and figure type analysis that could possibly gain you even more of an edge (in terms of filtering...
No brother, I was not sarcastic. This topic may not attract many, but very interesting for me because I extensively researched on gambling (my family business). I was twice thrown out of casinos, both times in .......(very famous place). This does not surprise me, the surprising part was on earning as little as $150 (2003) and next time $700 (2007). That gave me confidence that even with all odds in their favor, casinos are still afraid of calculated gambles. I don't drink while gambling and consider myself a very well behaved person. Money part should be just a joke for those casinos, then why? It was calculated gamble part that probably bothered them. I do not try to remember all results, only the last few winning numbers that are displayed on the table, not even the numbers, just the color (red or black). If logical outcome is 18 reds, 18 blacks, and 2 greens (0 and 00) it can only happen in three ways: 1) more red than blacks first, 2) more blacks than reds first and 3) alternate black and red. Even days on a roulette table is not a "long term", therefore I just follow the "leader" in blocks. First in a block of nine, then the leader in block of 16, ..... so on. No need even for a simple calculator or a pen, just memory of an ordinary person. Looks crazy, but try this at home, it is simple and works more than 50% of the time.

Casinos often observe unbroken sequence of one color (red or black) upto early forties (41 to 44). There will be broken sequences but "trends" are very probable. If there are trends, then there has to be counter trends and also consolidation periods even in roulette. I am not sure if I could articulate what I wanted to say but bottom line is trend lines work even in pure gambling. This is a very deep world and I can write books on this topic.

Yes, support and resitances are the key.
The Thief of Wall Street
  • #206
  • Aug 23, 2009 4:12am Aug 23, 2009 4:12am
Unless a roulette wheel is physically biased in some way, or there are otherwise 'external forces' that defy mathematical reasoning, then a distribution of spins must respect both expectancy theory and the law of large numbers. The latter dictates that any apparent trendlines or S/R levels have neither mathematical basis nor objective statistical significance. To suggest otherwise effectively means that, for whatever reason, the wheel does not deliver a truly random 50/50 red/black outcome, which is crucial to casinos' continuing to grind out a profit. Has any player succeeded in gaining a long-term edge over the house by identifying and exploiting non-random patterns (trendlines, S/R, whatever) in historical results? Despite the millions of punters that have attempted to win at roulette, I haven't heard of any such occurrence.

Exactly the same applies to a fair coin toss. Provided that one has any kind of faith in mathematical theory, then any trendlines, S/R or non-random patterns are either illusory, subjective, or (statistically speaking) a 'fleeting coincidence'.

In complete contrast, the markets are not random because they are created by human prerogative and premeditated agenda. Trendlines, S/R and other patterns skew probability wherever enough traders believe in them, creating clusters of orders at the appropriate levels, that have a self-fulfilling effect.

Of course I don't claim to be infallible. But, at least for me, the math will always speak for itself.
  • #207
  • Edited 6:15am Aug 23, 2009 4:41am | Edited 6:15am
Hi All...
Very interesting thread and a fascinating discussion going on here, thanks for starting it FX_Petra!

I've been also experimenting with similar modeling of synthetic data. The S/R zones and apparent trends in the "random" data is suggestive of weakness in random process underlying the data i.e. the random number generator (computer algorithm & the used hardware as well as the human in the case of physical coin flipping) is generating a number series which is from a pure random point of view "weak" and hence displaying a regularity.
This is good news since capital markets display similar characteristics that provide an opportunity to create and use an edge in periods of weak randomness i.e. "regularity" to make extensive profits in these markets.
I'm in complete agreement here with the viewpoints of daytrading, pipmutt, and hanover -- and
Scientific studies and experiments have been ongoing regarding the concept of "artificial randomness" in stochastic calculus at various research institutes as well backing up anecdotes of successes from everyday market practitioners.

...but of course everyone has to make that leap for him/herself and make the connection to translate this into a profitable trading strategy.

Trade Smartly!

regards,
GM
Hear the music before the song is over ...
  • #208
  • Aug 23, 2009 5:50am Aug 23, 2009 5:50am
Quoting hanover
Unless a roulette wheel is physically biased in some way, or there are otherwise 'external forces' that defy mathematical reasoning, then a distribution of spins must respect both expectancy theory and the law of large numbers. The latter dictates that any apparent trendlines or S/R levels have neither mathematical basis nor objective statistical significance. To suggest otherwise effectively means that, for whatever reason, the wheel does not deliver a truly random 50/50 red/black outcome, which is crucial to casinos' continuing to grind out a...
Have you verified this Hanover? Are those clusters of orders you speak of really there that supposidly make these areas bounce price? I submit to you that you are incorrect in that common assumption. The large orders are ussually tucked towards the 'middle' of the price structure rather than being on the fringes.
  • #209
  • Aug 23, 2009 5:55am Aug 23, 2009 5:55am
Quoting gbppip
It doesn't matter whether on not the market is random. What matters is what happens after you enter the trade, Mate.

I have a hard time believing experienced traders think they can predict the outcome of their next trade. It is the belief that the market can be predicted that is part of the reason 95% of traders fail.

I'll apologize to the thread starter if the discussion of expectancy is taking a backseat to randomness at the moment.
I read about how you trade here. I suspect that your edge comes from your breakout entries, and their timing around market open times, as much it does as the trade management, in fact possibly more so. It wouldn't be hard to prove this: you could apply the same exit strategy to completely random entries, and compare the resulting performance.

There are plenty of other systems whose edge comes from the trader's ability to predict the market direction. Not with 100% certainty, of course, but on the balance of probability that the market is more likely to move further one way, than the other, given a certain type of setup pattern.

There are many potential ways to profit, hence generalisations can be misleading.
  • #210
  • Aug 23, 2009 6:40am Aug 23, 2009 6:40am
"Unless a roulette wheel is physically biased in some way, or there are otherwise 'external forces' that defy mathematical reasoning, then a distribution of spins must respect both expectancy theory and the law of large numbers. The latter dictates that any apparent trendlines or S/R levels have neither mathematical basis nor objective statistical significance. To suggest otherwise effectively means that, for whatever reason, the wheel does not deliver a truly random 50/50 red/black outcome, which is crucial to casinos' continuing to grind out a profit. Has any player succeeded in gaining a long-term edge over the house by identifying and exploiting non-random patterns (trendlines, S/R, whatever) in historical results? Despite the millions of punters that have attempted to win at roulette, I haven't heard of any such occurrence."

While the trendlines and S/R generated from coin flipping or whatever else may have no 'objective statistical significance,' they do exist. The question is what do they mean? What are they? Are they of any value? Just because the math you know so well can't explain them, it does not nullify their existance. They're still there.

While I agree that over time you will have your 50/50 distribution in roulette for example, trends and S/R will play out and appear in the mean time. There will always be better times to take bets than others. I'm not a vegas guy, so I cannot say that these things will work in roulette, but if I were betting on a coin toss and I had a 'coin toss chart' in front of me, I'd probably take a crack at it.
  • #211
  • Aug 23, 2009 6:48am Aug 23, 2009 6:48am
Quoting Chicky
This is a very deep world....
Ok, now this is turning into something out of an episode of X-Files!

Predictable roulette, seriously? Hmmm I suppose anything is possible, have you found any correlation to the number of gamblers who bet red or black and the outcome of a spin, perhaps there's an element of telekinesis at work here.
  • #212
  • Aug 23, 2009 10:00am Aug 23, 2009 10:00am
Another point which comes to mind is that there are absolutely ZERO long term winners in casinos, and certainly over much less than 10000 bets. The very fact that there are 5% winners in the market (11% if you believe the NASS) goes some way to refuting the current premise. If the game truly is negative expectancy for all players regardless of what they do, there should be far few winners than there actually are, especially amongst traders with a higher turnover rate, since the house edge will not take long to show itself and bust them out.
  • #213
  • Aug 23, 2009 12:35pm Aug 23, 2009 12:35pm
Quoting hanover
There are plenty of other systems whose edge comes from the trader's ability to predict the market direction. Not with 100% certainty, of course, but on the balance of probability that the market is more likely to move further one way, than the other, given a certain type of setup pattern.
There is no need to predict, Mate, the market shows you what it is doing. My "edge" comes from the simple fact I do not try to beat the market because I do not want the rat to beat me.

When price hits the Asian Box low during the London or New York session, one of 2 things can happen to my trade when I enter short:

1) Price moves up and hits my stop loss

2) Price moves down and provides profit.

All I do is keep my stop loss a constant. The winners can return a risk / reward of 1 to 1 or more. Once I enter the trade, I can not make it a winner. I can not make small winners turn into big winners. All I can do is keep my losses a constant size. Doing this keeps me even with the rat.

I do not need to do anything else. I do not need any indicators to place lines on or under my chart. I fully understand market dynamics and am not "fooled by randomness". I refuse to complicate that which is simple. I'll never have a following because what I do isn't "sexy", it is flat out boring, Mate.
  • #214
  • Aug 23, 2009 12:43pm Aug 23, 2009 12:43pm
Quoting pipmutt
Ok, now this is turning into something out of an episode of X-Files!

Predictable roulette, seriously? Hmmm I suppose anything is possible, have you found any correlation to the number of gamblers who bet red or black and the outcome of a spin, perhaps there's an element of telekinesis at work here.
Actually Mate, there is one thing about roulette that can make it predictable in the short term - the dealer.

If the dealer spins consistently, you can watch where the ball is released and "predict" where the ball will land within 5 spaces on the wheel. This is shown when you see the same number appear twice within 3 spins.

I have exploited this to pay for lunch and dinner.
  • #215
  • Aug 23, 2009 12:46pm Aug 23, 2009 12:46pm
Quoting ThePipDoctor
Another point which comes to mind is that there are absolutely ZERO long term winners in casinos, and certainly over much less than 10000 bets....
Can you truly assign an expectancy to trading?
  • #216
  • Edited 1:31pm Aug 23, 2009 1:25pm | Edited 1:31pm
Quoting pipmutt
Ok, now this is turning into something out of an episode of X-Files!

Predictable roulette, seriously? Hmmm I suppose anything is possible, have you found any correlation to the number of gamblers who bet red or black and the outcome of a spin, perhaps there's an element of telekinesis at work here.
There is no correlation whatsoever and cannot be one because each spin is an exlusive independent event.

The only edge a "punter" has in such gambles is that no matter how long they are in the game, this length of time cannot be "long term" and therefore lesser probability of getting a 50:50 red and black. However, if they select the "leader" after every spin, the switching of the "leader" from red to black (or vice versa) will definately eat their bets in no time. Therefore I use blocks of spins to select the "leader" to minimize number of such switches. However, this puts me into yet another similar vicious cycle, but this time a longer one and enough for me to get out before it balances out. This I believe is the extreme edge, if any, one can get in such games.
The Thief of Wall Street
  • #217
  • Aug 23, 2009 1:28pm Aug 23, 2009 1:28pm
Quoting Chicky
No brother, I was not sarcastic. This topic may not attract many, but very interesting for me because I extensively researched on gambling (my family business). I was twice thrown out of casinos, both times in .......(very famous place). This does not surprise me, the surprising part was on earning as little as $150 (2003) and next time $700 (2007). That gave me confidence that even with all odds in their favor, casinos are still afraid of calculated gambles. I don't drink while gambling and consider myself a very well behaved person. Money part...
I am not sure what you are getting at... It seems like you are talking about some combination of mean regression and streak theory. I have studied both fairly extensively and can report both are very seductive. I have tried both in the casino and in the market and could make neither work.... but both because of capital restrictions. It is a scientific fact that the mean will regress to neutral (in oscillator terms), the probelm is there is no way of knowing how many trials it will take for that regression to take place. If after 100 trials you have flipped 60 heads and 40 tails, you would expect things to even up the next 100 flips. They may, or they may remain skewed for 1000 flips, or 10,000 for that matter. The capital requirements needed to adequatley fund such an exercise are far beyond me... even starting out with a .10 pip. Streak theory is also intresting, and I have used it in some previous posts without calling it as such. There is no question that streaks do occur, long ones, even with small sample sizes < 1000. The problem is, how do you know when a streak is starting? I am presently doing work in this area and feel I am close to being able to apply streak theory to forex--- but I have been close with other theories and never had a breakthrough. If anyone out there has any thing to offer along those lines, I would appreciate it.
  • #218
  • Aug 23, 2009 1:58pm Aug 23, 2009 1:58pm
Quoting FX-Petra
I am not sure what you are getting at... It seems like you are talking about some combination of mean regression and streak theory. I have studied both fairly extensively and can report both are very seductive. I have tried both in the casino and in the market and could make neither work.... but both because of capital restrictions. It is a scientific fact that the mean will regress to neutral (in oscillator terms), the probelm is there is no way of knowing how many trials it will take for that regression to take place. If after 100 trials you have...
Same analysis here. That's why I tend to be at bookmaking side of gambling and do not punt, except for fun. One of my few gambling rules:

Chicky's Gambling Rule #2: Disaster is the ultimate result of any kind of gambling. Because, there is a limit to how much one can lose, but no limit to how much one wants to win.
The Thief of Wall Street
  • #219
  • Aug 23, 2009 2:01pm Aug 23, 2009 2:01pm
Quoting ThePipDoctor
Another point which comes to mind is that there are absolutely ZERO long term winners in casinos, and certainly over much less than 10000 bets. The very fact that there are 5% winners in the market (11% if you believe the NASS) goes some way to refuting the current premise. If the game truly is negative expectancy for all players regardless of what they do, there should be far few winners than there actually are, especially amongst traders with a higher turnover rate, since the house edge will not take long to show itself and bust them out....
The only long term winners in a casino are those that have changed the negative expectancy of their game to a positive one. One such method is counting cards at blackjack, under the right conditions. I was a serious part-time counter for almost 2 years and made some sweet coin--- until pit bosses started moving the cut card on me. 2 casinos even told me I could not play blackjack at their casino anymore. Edges do exist in this world. I am looking for one in the spot retail forex market, but to this point I have not found it. If anyone out their has found it, they are keeping it to themselves. As I have stated earlier, I do not believe any such system exists that is presently available to the general public. I know there are a lot of traders who think they have a system that is positive expectancy. They are basing that belief on personal experience and past results. They are experiencing short term positive variance while operating underneath a negative equity curve. If they continue to trade, they will be wiped out. For some, they are gone in the first 100 trades. Some make it to 1000. Fewer still survive to 5000. The number still around continues to decline as large trade numbers amass. That is not to say that the person who makes it to 5000 trades is not a better trader than the person wiped out at 100. The 5000 trade trader is probably more astute in managing their equity and market volitilty. The most astute can continue on to the 10,000 trade level, where only about 5% are left. In the end, negative expectancy will find us all. By all, I mean small retail traders. George Soros (and his friends) had enoungh capital reserve to lean on the Bank of England until the Pound popped and they could collect their prize. People like that are not staring at a graph and hoping to predict the next move. They have the capital to make that move happen. Not so with us small fish.
  • #220
  • Aug 23, 2009 2:09pm Aug 23, 2009 2:09pm
Quoting FX-Petra
The only long term winners in a casino are ..................Edges do exist in this world. I am looking for one in the spot retail forex market...
You may like to:

1. Read "Trading in the Zone" by Mark Douglas
2. Click the link in my signature below.

I think you have found what you were looking for. GL.
The Thief of Wall Street
0 traders viewing now