Probability System Theory and Money Management | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Alternative Trade Management/Money Management Methods 3 replies

Money Management Theory 94 replies

Money Management / Risk Management 24 replies

Probability, Pyramiding and Money Management 37 replies

Lessons in directional probability and risk management 29 replies

Probability System Theory and Money Management

  • #41
  • Dec 15, 2006 7:24pm Dec 15, 2006 7:24pm
Before I go up to Nashville to play some tunes, my Friday night mellowness is giving me a different take on this...

We can gamble and win. I've done that. Back in 2002 Euro was a sure long, it was ridiculous, and I went all in all the time, with absolutely no knowledge whatsoever of fundamental. technical, or any kind of analysis, I did what they told me on Reuters and Bloomberg. I made a killing, at first. I thought it was free money.

Then I lost it all. I used to figure out how much of my account I could put on a trade without getting a margin call and didn't use stops. I'd see a news spike and didn't even know what a news spike was. I'd just buy, buy, buy on the spike and lose my ass on the retracement. I had never seen an economic calendar. I didn't know what a moving average is.

The point being, you can make this gambling if you want. You might win, but you could win a million bucks on the lottery, too, it's 1 chance in 40 million. I prefer the odds being with me and not against me, however. I go a lot slower now, use a lot less leverage, and let compounding do the work for me...
"The time to act is when others show signs of tire." --W.D. Gann
  • #42
  • Edited 1:06pm Dec 16, 2006 12:52pm | Edited 1:06pm
Quoting Monaco
Here is my live experience with Martingale from a long time ago with horse racing...

When I was very young I heard about a way to gamble that was guaranteed to win. This guy I met told me that in British horse racing all you have to do is choose a course say for example Epsom. They will have 5 races or thereabouts a day and the odds are that the favourite will win one for those races. The strategy was to keep doubling up until you win. It made sense to me and I thought lets go for it. Sure enough I walked into a betting shop in London put a bet on and the favourite won the race. Delighted I walked out of the betting shop with my winnings. Next day I went in and the favourite lost the first race but won the second race - again I was delighted. This went on for several days until one day I walked into the betting shop surrounded by old guys smelling of yesterdays whiskey and stale cigarettes (that should have been a clue to me right there to stop doing this). Anyway I put a GBP10 bet on the first horse - LOST. 2nd race GBP20 - LOST. 3rd race - GBP40 - LOST. I had never been past the 3rd race before and I started to panic and I was broke. I ran home and got some more money. 4th race GBP80 - the bet is on and the race starts...my heart is pounding as fast as the horse is running - LOST. I was devasted but I could not stop now, I knew I had to win. 5th and final race of the day. I approached the guy behind the counter to give him GBP160 and he told me that he had been watching my bets and asked me what I was doing. I told him the system and he told me to walk away as he had over the years seen many systems come and go and many more losers. I looked around to see the other gamblers and decided I never wanted to be like them at all...walked out the betting shop and haven't been back since. When I got back home I wanted to see what would have happened if I had stayed for the 5th race - the favourite won the race and I would have been vindicated but it would have meant I would have risked GBP310 to win just GBP10. I thought to myself I should have had the guts to do that but then again was I willing to experience a cardiac arrest in that 5th race and all for GBP10! My conclusions were that what may be good in theory is not healthy in reality.

By the way if anyone wants to adopt this strategy for horse racing then I will gladly be your bookmaker!!

Have a great trading day!

Monaco
Since the start of this thread I've since abandoned a martingale strategy, but not because it doesn't work. It works in theory. If you have $100,000 could you not trade it successfully with a microlot account on 400:1 leverage? You probably could, but whats the point? That is why I gave up. To be able to survive such a strategy would require that you trade for profits that are very small. If you have a $100,000 account why are you bothering to trade microlots? It would take forever to start accumulating substantial gains in your account.

I'll just defend martingaling in its theory though. The above horse betting scenerio is not very martingale friendly. This is because a coin flip has a 50/50 shot at being heads or tails. In the horse race your horse really only has a 1 in 6 chance, or however many horses are in each race, of winning. So it is much more risky than a coin flip.
  • #43
  • Dec 16, 2006 1:03pm Dec 16, 2006 1:03pm
Quoting dukeofdents
Here's my down home redneck thinking on money management, I'm not a math whiz, or anything, just a blue-collar dude,


I put 5% of my account on a trade. I only take 1 trade at a time. With 50:1 account leverage I have 250% in the market. That's 2.5/1 real leverage, sometimes I will go all in at 10%, being 5/1 real leverage. That's all I do. Why would the fact that you lost your last trade have any effect on your next trade whatsoever? Again, I'm not a math whiz, but, if I flip a coin right now, it doesn't matter at all how many times I flipped it before. What changes about the coin between the first time I flip it and the 20th time I flip it? It will still be a 50/50 bet. If I use a different coin every time it will be the same.
Again, yes on each individual trade you have a 50/50 shot of being right. When you correlate the trades to make a pattern then it becomes very unlikely. What are the chances of getting heads, then tails, then heads again in that specific pattern? Well first you need heads. You have a 50/50 shot so far. Then out of the 50% time you get heads you need tails on the second flip. So 50/50 shot of the 50% = 25%. Then to get heads on the third flip to complete the pattern you now have a 50/50 shot of the 25% so far = 12.5%. That's the idea of how martingale is succcessful.

Again, I've moved on from martingale since, while profitable if you have the funds, does not bring in profits large and quick enough to be worth it. Thanks for the thoughtful replies.

Right now im fancying the idea of how to get 25 pip profits per day. You always read posts about $1000 to $1,000,000 in 3 years and how they outline the perfect scenerio of not losing any trades and how long it takes. Well, just how can we practice that theory. If I use a 50 SL 75TP will it come out to very similar results? All depends on how often I can accurately trade to win. 50% is all I need a day right? I only ask because of avoiding getting stopped out. What is the minimum SL/TP combination to average 25 pips more on the winners without having it too small to get stopped out from general market noise.
  • #44
  • May 30, 2008 11:45am May 30, 2008 11:45am
hello guys!
I have a few thoughts and hope we can have a little discussion!

Lets look at one good trade, while i was in a trade the maximum i was down was 1%, the total profit was 20%! Thus under a normal distribution, the probability was only 5%.

Now its starts to become tricky! Lets imaginate the underlying only moves in the direction of our successful trade forever, and a drawdown happens every once in a while of only 1%, the probability of our trade would be 100%, thus our trade would be very probably and it would not be too much due to skill!

My question is can the level of skill be determined by the level of probability of the event happening? In this case it would be a negative correlation, thus the more unprobable the event, the higher the skill?

Furthermore, what are the timeframe and the probability were luck turns into skill? Is achieving an event/events of 1% in two luck or skill? This is based on the assumption of that there is a normal distribution and you would do two trades were could loose only 10% of what you win!

Hope to hear from you guys! My structural thinking and explain isnt the greatest, so please ask if i should re-word something.
Best regards
Paul
  • #45
  • May 30, 2008 5:24pm May 30, 2008 5:24pm
Martingale is a good strategy used only in consideration of analysis and price movement. I think it's ok to use it, but when you do you have to consider the doubling up. So let's say you lose your first trade, and double on the second...that second trade (which is executed by your trading system) should be considered TWO trades. So if you lose you have THREE consecutive losses and not just two. That way, according to your system at five or 10 losses you have the ability to walk away and start fresh..just like you would if not using martingale. In this case 10 losses would be approximately three trades. The fact is with a positive expectancy system, Martingale should increase your chances of recovering a loss dramatically. At some point you have to admit a loss (as your pockets aren't infinite). Plus, using martingale in the first place is a risk itself.

I've been pondering this for about two weeks now and that's the only practical method of useage that I can come up with.
  • #46
  • Nov 14, 2009 1:35pm Nov 14, 2009 1:35pm
A martingale strategy WILL fail at some point this is a statistical certainty. However, there is a way to profit from it nonetheless. It is also a statistical certainty that you will have winning streaks and some of those will be dramatic.

Therefore we can conclude that if we have say, $10,000 to invest, we can start in increment lots of $1000, knowing that some of those 1k lots will be wiped out. We also know that some of those lots will show astonomical profits. By segmenting our risk into consecutive parcels in which we start over after a wipeout, we greatly increase the odds of success.

The trick is to set a reasonable limit on how much profit we will settle for before we retire that lot, and start over with a fresh $1k Let's say we settle for a triple. If your 1k lot triples to 3k, retire that lot and start over with another 1k lot.

I'm not sure how to calculate the math on this, but my gut tells me it must increase the winning odds by some order of magnitude. Let me know if I'm missing something, but this looks like the way to play a martingale.
  • #47
  • Jan 24, 2010 2:09pm Jan 24, 2010 2:09pm
Quote
So on four trades you made 900 pips, where if you had won all 4 you would have only made 400. You were willing to risk more on each consecutive trade knowing statistically that it was very improbable to continue to lose. So you almost want the system to lose a couple times in a row once in a while. It increases your probability of success and more profit potential.

There's your problem. You think the more you lose, the greater the chances are of winning. On the contrary every time I flip a coin it is a unique outcome, not dependent on the last flip.

While the markets oscillate somewhat, it is easy for them to trend for years against the direction of your martingale system.
  • #48
  • Jan 24, 2010 7:15pm Jan 24, 2010 7:15pm
Trading in the Zone is a good book which I think talks about the probablity system..
  • #49
  • Jan 25, 2010 5:25am Jan 25, 2010 5:25am
Quoting doos
There's your problem. You think the more you lose, the greater the chances are of winning. On the contrary every time I flip a coin it is a unique outcome, not dependent on the last flip.

While the markets oscillate somewhat, it is easy for them to trend for years against the direction of your martingale system.
The angle that your reply is negating is the example of flipping a coin 100 times and receiving 100 heads... each coin flip is considered unique and its own beginning, middle, and end (this is your correct stance in your reply)... the problem is that when you observe a series of events that each have only 2 possible outcomes you expect that about half would have one of the outcomes and the rest would show the second outcome.

To observe a series of wins vs losses and see it showing a serious bias would give one reason to consider that the second outcome is 'due'. Yes, each time there is a new event it has a 50% chance of being either outcome, but the observed series would suggest the expectation of a reversal (so to speak).

(seeing such a series should make one consider the possibility that perhaps this 50/50 play is not 50/50 after all, and something else is in play here making this not a 50/50 at all but a completely different beast)

The problems with all of this is that markets are not strictly random. There are reasons the market moves. The problem with THIS though is that to the individual the knowledge of why the market is doing something may not be known, so in essence the market to the individual BECOMES random. Mix with this the fact that for any move the market makes it will at some point make the same move in the opposite direction for new reasons and now you begin to negate my first sentence above, and so now you have counter-moves which both have reasons non-random... but overall to the unknowing individual the moves remain random because their cause continues to remain unknown.

The market is not a 50/50 play though at any point in time. Foreign exchange is influenced by more than just the 2 countries which the trade is basing its currency exchange on... outside factors, countries, events, influence the bid/ask and the exchange rate. It is more of a living animal than just a simple heads vs tails game.

The biggest problem of all will always be the spread. That is where the casino (broker) wins, and continues to beat your account to death and smites all probability plays. Even if the markets were a head vs tails game the spread chomps into your game and makes many a ;-( for 'gamblers' and many >:-] for the brokers.

some things to ponder.
Congratulations ScalaFX for coming out of the closet!
  • #50
  • Jan 26, 2010 2:10am Jan 26, 2010 2:10am
Quote
but the observed series would suggest the expectation of a reversal (so to speak).

1) No it wouldn't. You would have lost 100 times in a row thinking like that. Every coin flip is completely unique. 50/50 everytime, even if there were 1billion heads in a row. Still 50/50.

There is a good example of this.

A game-show with three doors. Does anyone know what this example was called? It is very famous published in a NY newspaper some years ago.

2) The markets are not random you are right. Market movers obviously buy and sell around monthly pivots and big SMAs very often. Interest rates, carry trades, and the dumping of oil and gold obviously have big correlation too. So you're right I don't think the markets are very random once you are looking at 3 or 4 of the correct variables.


3) Spread is an issue when scalping. Say the spread is 2 pips and you want 2 pips, then they have a 50% grab at your cash. So instead of scalping why not trade for larger amounts like 50+ pips? These trades are usually higher probability AND beat the spread (or at least make the spread rather meaningless)
  • #51
  • Jan 26, 2010 2:13am Jan 26, 2010 2:13am
Quote
Mix with this the fact that for any move the market makes it will at some point make the same move in the opposite direction

I think this a very short-term assumption. Price action is not stuck in bounds like the RSI is. It's in an unbounded arena, with trends lasting longer than account balances.
  • #52
  • Jan 26, 2010 4:56am Jan 26, 2010 4:56am
Quoting doos
I think this a very short-term assumption. Price action is not stuck in bounds like the RSI is. It's in an unbounded arena, with trends lasting longer than account balances.
i do not mean an equal move, just an opposite move for new reasons to make it do this.

back to the 50/50. i agreed with you that each flip is its own event and has an equal chance of being heads or tails, but that was the point in that the original post was not talking about individual flips and was talking about a series and its probability as a whole. if after 99 flips you get heads you would either look at it as being really overdue for a tail, or you realize it is rigged and there is no 50/50 going on here at all.

i'd be interested in someone bothering to do the math on the probability of 100 heads in a row. i'm sure it is not hard, but i'm no mathematician or statistician so would rather someone cut the turkey who knows how to handle the knife. does a series such as this become infinitely unlikely to occur the further in the 'trend' you go? is this not where people start discussing things such as 'bell curves' etc?
Congratulations ScalaFX for coming out of the closet!
  • #53
  • Jan 26, 2010 5:03am Jan 26, 2010 5:03am
try making 100 positiv trades in a row... should be easier than flip the coin
StopLoss is for none belivers... Trust the Gurus..
  • #54
  • Jan 26, 2010 6:44am Jan 26, 2010 6:44am
Quoting nubcake
i'd be interested in someone bothering to do the math on the probability of 100 heads in a row.
It is 1 in 2^100, i.e. 1 chance in 1,267,650,600,228,229,401,496,703,205,376.

Making 100 consecutive winning trades would be significantly more difficult, due to the spread (assuming random entries and exits).
  • #55
  • Jan 26, 2010 7:14am Jan 26, 2010 7:14am
Quoting hanover
It is 1 in 2^100, i.e. 1 chance in 1,267,650,600,228,229,401,496,703,205,376.

Making 100 consecutive winning trades would be significantly more difficult, due to the spread (assuming random entries and exits).
so, you are saying it is pretty likely then?

ALL ON RED!!!

Hanover, now please calculate for me the chances I will some day soon make some serious money in forex. thanks in advance.
Congratulations ScalaFX for coming out of the closet!
  • #56
  • Jan 26, 2010 7:38am Jan 26, 2010 7:38am
ofcourse you can make 100 trades positiv in a row...
just not take random entries...lol
StopLoss is for none belivers... Trust the Gurus..
  • #57
  • Nov 26, 2010 9:44pm Nov 26, 2010 9:44pm
[quote=permanentjaun;156373]To those that would be turned off by sacraficing your typical trade size to survive using these principles think about this.

What if you had enough leverage, access to micro lots, and enough of a bankroll to play the doubling down as if your SL was 100 and TP was 50?

To break even, and only break even on losing trades in that system your double down line would look like this:

1 - 2 - 6 - 18 - 54 - 162 - 486 - 1458 - etc.

But if you apply this to your equal losses gains system of 100 SL and 100 TP then not only does it make it seem like the losing trades never happened, it makes it seem as if the losing trades were actually winning trades. Not only that. Look at what happens if you lost 3 trades in a row and won on the fourth. You would have lost 900 pips, but gained 1800 for a total profit of +900 pips.

So on four trades you made 900 pips, where if you had won all 4 you would have only made 400. You were willing to risk more on each consecutive trade knowing statistically that it was very improbable to continue to lose. So you almost want the system to lose a couple times in a row once in a while. It increases your probability of success and more profit potential.

Like I said, money will be slow in the beginning, but eventually when you have enough to handle the risk you slowly increase the multiplication factor to bring in massive profits. Imagine making 5 bad trades in a row and winning on the 6th. You would have lost 8100 pips total, but banked 16200. Do you see why I insist this type of money management be played out on a micro micro lot broker? But hey, you just made $81 using 1 cent pips. Compound compound compound. Invest invest invest. Eventually the money will come.

Anyways, I need criticism. What am I missing? In the last thread I completely missed the concept that using a 100 SL and 50 TP would require more than just doubling on the trades to break even. This thread changes the variables to make it more plausible with a bankroll and lot sizes. HELP! Matt[/quote

Yes, 50% chance when using random entry and equal stop,take profit.
So, use fast stochastic indicator, when price action is below 20 or above 80 on the stochastic, put your trade on with egual stop and take profit. About 30 pips for each. Do not increment your bet by increasing lot size after a loss. If stochastic has any statistical benefit this trading method with will show it does, by having a growing equity in the account. In other words, your 50/50 idea maybe be profitable long term if entered in oversold or overbought condictions. Maybe use 1hr chart on euro.
Goodluck!
  • #58
  • Jul 31, 2011 8:35am Jul 31, 2011 8:35am
thanks for sharing this, it does take a bit time to digest
  • #59
  • Last Post: Jul 31, 2011 3:24pm Jul 31, 2011 3:24pm
Quoting glennw
A martingale strategy WILL fail at some point this is a statistical certainty.
Not true; this statement is too broad. It is possible to implement effective martingale strategies on the markets.

Quoting doos
A game-show with three doors. Does anyone know what this example was called? It is very famous published in a NY newspaper some years ago.
Monty Hall Problem.
1 trader viewing now