why have you accepted the price-predictability assumption? | Page 104 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

So I accepted a Prop Trading job in South Beach Miami 44 replies

Why you shouldn't show how much you have made in FX 35 replies

Incorporating Slippage Assumption in backtesting 4 replies

Market Predictability - Technical Signals 25 replies

Contesting the Accepted View of Bernanke 0 replies

Attachments: why have you accepted the price-predictability assumption?
Exit Attachments

why have you accepted the price-predictability assumption?

  • #2,061
  • May 14, 2014 5:28pm May 14, 2014 5:28pm
Quoting Proximus
{quote} It is ok to be skeptic, and if you believe me or not, you can always test this yourself, but eventually you will come to the same conclusion as i did.I`m just saving you time and telling it to you now. {quote} Then you are a gambler not a trader.You try to exploit the volatility or the structure of the market, which is unpredictable.You may be succesful for now, but this wont last forever.Focusing on the direction however and succesfully predicting it, will give you the so seeked edge, and after that, the RR will be irrelevant. {quote} Yes,...
actually my strategy is fading moves/reversal/retracement, not sure how you got the idea it's a breakout..

I can't deny a huge part of the edge comes from the lines I draw... and I don't trade eurusd.

I don't think it's gambling; I know because I gamble and I trade - the distinction is pretty clear to me.

hmm... haha I am still going to test this probably, just not sure when..
Build good relationships with others.
  • #2,062
  • May 14, 2014 5:33pm May 14, 2014 5:33pm
Quoting shellsnail
{quote} But if you can prove to a certain degree of confidence that a high RR outperforms a low RR in a random trade scenario, you can potentially make a case to abandon all low RR trading systems. I'm not looking for a standalone trading system from this - just possible verification of the hypothesis that low R:R does not perform well. Do we have enough evidence/data to make that conclusion? And why am I doing this? For curiosity sake really. I have a working method and I am not actively looking to change my system, but if something interesting...
this might be along the lines of what you're looking for

http://forums.babypips.com/newbie-is...anagement.html
tradewith60
  • #2,063
  • May 14, 2014 5:37pm May 14, 2014 5:37pm
Quoting shellsnail
{quote} actually my strategy is fading moves/reversal/retracement, not sure how you got the idea it's a breakout.. I can't deny a huge part of the edge comes from the lines I draw... and I don't trade eurusd. I don't think it's gambling; I know because I gamble and I trade - the distinction is pretty clear to me. hmm... haha I am still going to test this probably, just not sure when..
Oh right, i misread that i though you meant shock not aftershock.So you trade the GBP/JPY, which is a very volatile pair and somewhat acceptable spread, but the volatility is a 2 edged sword, if you have no clue about the future direction you can just lose as much as you can win.

Trading can be gambling too, if you have no clue about what you are doing.And if by gambling you mean poker, thats not really gambling, but stuff like roulette or baccarat is, but thats another topic.

But i prefer trading, and knowing the future direction, so i think its good to follow the macroeconomics aswell, because all movement in the market has a reason behind it.
"There's a sucker born every minute" - P.T. Barnum
  • #2,064
  • May 14, 2014 5:38pm May 14, 2014 5:38pm
Quoting 60minuteman
{quote} this might be along the lines of what you're looking for http://forums.babypips.com/newbie-is...anagement.html
Hey thanks, will take a good look at that
Build good relationships with others.
  • #2,065
  • May 14, 2014 5:44pm May 14, 2014 5:44pm
Quoting 60minuteman
{quote} this might be along the lines of what you're looking for http://forums.babypips.com/newbie-is...anagement.html
Ah c'mon guys, some seriousity please, look at the sample size: 500 - 1000

I had atleast 200.000 trades in a 10 year period, and cannot support his hypothesis.

Besides how come that i can make good profit with 1:2 RR
"There's a sucker born every minute" - P.T. Barnum
  • #2,066
  • May 14, 2014 5:59pm May 14, 2014 5:59pm
Quoting Proximus
{quote} Ah c'mon guys, some seriousity please, look at the sample size: 500 - 1000 I had atleast 200.000 trades in a 10 year period, and cannot support his hypothesis. Besides how come that i can make good profit with 1:2 RR
Maybe because your system relies a lot more on getting the direction right? Have you tried using larger RR though?

I wouldn't dismiss the results so quickly since she did use all available data for EUR/USD. There seems to be something systematic about the market being exploited in the profitable samples (i.e. the fat tails perhaps)... This does appear to confirm my hypothesis (superficially at least) - I would have to get in touch with the TS to find out more about her estimation strategy etc.
Build good relationships with others.
  • #2,067
  • May 14, 2014 6:05pm May 14, 2014 6:05pm
Quoting Proximus
{quote} Ah c'mon guys, some seriousity please, look at the sample size: 500 - 1000 I had atleast 200.000 trades in a 10 year period, and cannot support his hypothesis. Besides how come that i can make good profit with 1:2 RR
you're clearly lucky as well as a know it all
tradewith60
  • #2,068
  • May 14, 2014 6:12pm May 14, 2014 6:12pm
Quoting shellsnail
{quote} Maybe because your system relies a lot more on getting the direction right? Have you tried using larger RR though?
Yes my system is actually the prediction of future price with various tools, fundamentals, sentiment analysis, volatility, etc.
And sorry i wasnt very clear, what i meant is my average RR is 1:2, so i have anywhere from 1.5:1 to 1:3 RR, and these numbers are not fixed, they vary and doesnt influence my performance.I put a 1:3 RR just because my risk management is asking for it, but if i`d ignore that, it could be anything with the criteria that the targeted/predicted price would be encompassed in the RR configuration.

Quote
I wouldn't dismiss the results so quickly since she did use all available data for EUR/USD. There seems to be something systematic about the market being exploited in the profitable samples (i.e. the fat tails perhaps)... This does appear to confirm my hypothesis (superficially at least) - I would have to get in touch with the TS to find out more about her estimation strategy etc.

Yes, it is exploiting the volatility as i said in previous posts, but it will fail horribly in trending markets like 2002->2004 EUR/USD.Read my previous post please.

And its still gambling if you put trades randomly, you can trade with 3:1 RR but you need a real edge to do so.Poking blindly on BUY/SELL buttons with predetermined RR ratio,is not an edge.
"There's a sucker born every minute" - P.T. Barnum
  • #2,069
  • Edited 10:50pm May 14, 2014 6:17pm | Edited 10:50pm
Quoting 60minuteman
{quote} you're clearly lucky as well as a know it all
Yea 87% winrate is really a matter of luck.Besides my expectancy +5 pip/ trade...How much is yours?

Edit: Linked in my practice account in my signature for proof that i`m not talking BS here
"There's a sucker born every minute" - P.T. Barnum
  • #2,070
  • May 14, 2014 6:23pm May 14, 2014 6:23pm
Quoting Proximus
{quote} Yes my system is actually the prediction of future price with various tools, fundamentals, sentiment analysis, volatility, etc. And sorry i wasnt very clear, what i meant is my average RR is 1:2, so i have anywhere from 1.5:1 to 1:3 RR, and these numbers are not fixed, they vary and doesnt influence my performance.I put a 1:3 RR just because my risk management is asking for it, but if i`d ignore that, it could be anything with the criteria that the targeted/predicted price would be encompassed in the RR configuration. {quote} Yes, it is...
But how can you argue it's not an edge when it has been clearly demonstrated to be an edge over a statistically significant sample? (come on be realistic, 1000 trades is really quite a large sample on daily time frame)
Build good relationships with others.
  • #2,071
  • Edited 6:56pm May 14, 2014 6:38pm | Edited 6:56pm
Quoting shellsnail
{quote} But how can you argue it's not an edge when it has been clearly demonstrated to be an edge over a statistically significant sample? (come on be realistic, 1000 trades is really quite a large sample on daily time frame)
It's just not, its that simple , if trading would be that simple everyone would be a billionaire, because in every trading book you can see that "3:1 RR advice", so it cant work if everyone knows about it,yet 90% retailers lose...

But enough of speculation, please test it yourself to make sure

No edge, nor disadvantage come from the ratio of the RR, its just irrelevant, but proper analysis and a good predicting ability will be an edge on the other hand.The RR serves only as a mean of controlling risk, so you dont want to have a 1:10 RR because that might end in a margin call, but if your risk management is ok, then i see no advantage in the RR ratio configuration.

Besides 1000 trade is nothing, if you find a high volatility pair like EUR/JPY the market can move 80 pips a day, thats like 1 candle on the daily, so if he tested the combinations under 100 pips, then 1000 trade is like 3 years or so.

Which is only catching the recession period and the late crisis period until at 2011.But what about the gold ascension in parralel with the bull EURO markets in the 2000's and the 90's Dow Jones / USD bull market before the high frequency traders.

You have to test it in all enviroments to make a proper assesment.It is like learning to swim in a pool, but when you go into the ocean the sharks eat you
"There's a sucker born every minute" - P.T. Barnum
  • #2,072
  • May 14, 2014 11:20pm May 14, 2014 11:20pm
Quoting PipMeUp
{quote} I made the same kind of experiment with the same conclusion. One trade at a time, spread taken into account, repeating the backtest several times. The expectancy quickly converges to minus the spread. Worse, I got the same conclusion when always opening the trade in the direction of the slope of a MA instead of flipping a coin. I even tried very small TP/huge SL to see if trading the noise was an option: http://www.forexfactory.com/showthre...12#post6512412
Interesting... Depends on which timeframe you placed the MA, if it was on a higher timeframe, it could actually catch a longer trend, but leaves you vulnerable in ranging markets, if on lower timeframes, than exactly the opposite, it cathces the ranging nice, but when a trend starts, it will just give improper signals.

And the MA is not a bad indicator afterall, mean reversion does occur, but if you choose to trade with it, then a mechanical approach is not good, because the market structure constantly changes and the indicator cant adapt to it, unless you want to re-optimize it after every trade.

It needs a little bit of distretionary element, pure mechanical approach is just flawed, it only works for high frequency trading where the edge is the speed of execution, but on high timeframe, only proper analysis works.
"There's a sucker born every minute" - P.T. Barnum
  • #2,073
  • May 15, 2014 2:45am May 15, 2014 2:45am
Quoting shellsnail
{quote} Yes, I am aware the expectancy might converge to minus the spread, but my question is in two parts... 1) Does any RR have an edge on its own? 2) Does high RR outperform low RR? 2b) Under what conditions might one outperform the other?
I never had Proximus' results. I may have not been as thorough as him. I didn't try each possibilities. For me not a single of the {SL,TP} gave an edge. Very small SL kills the account quickly. But the others are just edgeless (E=-spread). Sure you can get positive realizations but in average it's zero.

Also big RR usually comes with low winrate. For a given expectancy, the higher RR the lower the WR. So for the same expectancy a big RR system has longer strings of losers and thus worse equity swings. The risk of ruin is higher because a random walk, even with a drift, may reach any level. This include all the levels below the zero line.

WR in itself isn't enough. RR in itself isnt't enough. MM in itself isnt't enough.
No greed. No fear. Just maths.
  • #2,074
  • May 15, 2014 2:56am May 15, 2014 2:56am
Quoting PipMeUp
{quote} I never had Proximus' results. I may have not been as thorough as him. I didn't try each possibilities. For me not a single of the {SL,TP} gave an edge.
There were, but as i said they had whole years in minus, and in the other years compensated for the losses + profit, which is unacceptable.But it could be that my sample was also small [10 years of H1] and maybe on a 20 year scale it would also lose money, in a weird distribution as getting the long winning streak and then losing it all in the next couple years.

Its all because of the weirdness of the economic cycles, and the crossing of a small cycle with a large one.

And if you want to be a trader you probably look for a 10-20 year old carreer so, you want a very durable strategy for that.
"There's a sucker born every minute" - P.T. Barnum
  • #2,075
  • May 15, 2014 2:59am May 15, 2014 2:59am
Quoting Proximus
{quote} Interesting... Depends on which timeframe you placed the MA, if it was on a higher timeframe, it could actually catch a longer trend, but leaves you vulnerable in ranging markets, if on lower timeframes, than exactly the opposite, it cathces the ranging nice, but when a trend starts, it will just give improper signals. And the MA is not a bad indicator afterall, mean reversion does occur, but if you choose to trade with it, then a mechanical approach is not good, because the market structure constantly changes and the indicator cant adapt...
It absolutely doesn't depends on the timeframe but on the timeSCALE. That is the cut off frequency of the filter (MA) verse the frequency of the main swings. It was applied on H4 and it was a kalman filter so it re-optimized at each bar.

If the MA is vulnerable in ranging market it should be no worse no better than flipping a coin, so I should have had zero expectancy during the range periods. If it gave a slight edge during the trending periods, I should have had overall a positive result made of zero in the ranges plus a little something in the trends. That wasn't the case.

Distretionary is made by your brain. Human brain is the worse thing when it comes to combine probabilities. Want to try?
""
Mary is pregnant of her third child. There will be two boys in the brotherhood. What's the probability Mary is expecting a girl?
""
No greed. No fear. Just maths.
  • #2,076
  • May 15, 2014 3:09am May 15, 2014 3:09am
Quoting PipMeUp
{quote} It absolutely doesn't depends on the timeframe but on the timeSCALE. That is the cut off frequency of the filter (MA) verse the frequency of the main swings. It was applied on H4 and it was a kalman filter so it re-optimized at each bar. If the MA is vulnerable in ranging market it should be no worse no better than flipping a coin, so I should have had zero expectancy during the range periods. If it gave a slight edge during the trending periods, I should have had overall a positive result made of zero in the ranges plus a little something...
Yep, that was a lack of discretionary intervention.I have to disagree on that with you, the human brain is the best signal service out there.It can handle subconsciously very complex calculations, which nor the best neural networks can handle.

Discretionary trading is much better than mechanical, its not optimal because humans do mistakes, but its profitable atleast, while mechanical isnt.Unless you make a neural network like skynet, no chance it will predict the price reliably and for along time.

As in the case of Mary, thats a totally random event, unless you know the genetic conditions of her husband and hers, its impossible to extract an edge there, but in the case of the market, you can by aquiring external information like macroeconomic indicators, news releases, etc...
"There's a sucker born every minute" - P.T. Barnum
  • #2,077
  • May 15, 2014 3:44am May 15, 2014 3:44am
Quoting Proximus
{quote} Yea 87% winrate is really a matter of luck.Besides my expectancy +5 pip/ trade...How much is yours? Edit: Linked in my practice account in my signature for proof that i`m not talking BS here
139 trades not significant (can you see what i did there!) and it's demo...
tradewith60
  • #2,078
  • May 15, 2014 3:59am May 15, 2014 3:59am
Quoting Proximus
{quote} Yep, that was a lack of discretionary intervention.I have to disagree on that with you, the human brain is the best signal service out there.It can handle subconsciously very complex calculations, which nor the best neural networks can handle. Discretionary trading is much better than mechanical, its not optimal because humans do mistakes, but its profitable atleast, while mechanical isnt.Unless you make a neural network like skynet, no chance it will predict the price reliably and for along time. As in the case of Mary, thats a totally...
Our brain evolved to quickly find approximative solutions. Evolution taught the various species that it is better to cowardly escape a piece of rotten wood rather than taking the time to go and check whether it is indeed a predator or not.

Brain is wired to find patterns even if there is no pattern at all. Just look at the sky you'll see rabbits, witches, sail boat and dinosaurs, ... In reality that's just clouds in the sky.

Mary has 1 chance out of 3 to deliver a girl this time. Whoever she got married to.
No greed. No fear. Just maths.
  • #2,079
  • May 15, 2014 4:06am May 15, 2014 4:06am
Quoting 60minuteman
{quote} 139 trades not significant (can you see what i did there!) and it's demo...



Edit (making it a representative sample):
Build good relationships with others.
  • #2,080
  • May 15, 2014 4:09am May 15, 2014 4:09am
Quoting PipMeUp
{quote} Our brain evolved to quickly find approximative solutions. Evolution taught the various species that it is better to cowardly escape a piece of rotten wood rather than taking the time to go and check whether it is indeed a predator or not. Brain is wired to find patterns even if there is no pattern at all. Just look at the sky you'll see rabbits, witches, sail boat and dinosaurs, ... In reality that's just clouds in the sky. Mary has 1 chance out of 3 to deliver a girl this time. Whoever she got married to.
I didnt implied to find the solution in the chart.There are external signs which can predict the price: news, macroeconomic indicators.I didnt said that technical analysis solely is reliable.

As for Mary, it could be 1/3, if you dont know any additional information, i`m not a doctor, but i`m sure there are methods to measure the outcome more precisely than that, and i think some genetics still play a role here.But of course without those information you have no way of knowing.

Thats a good analogy for the markets too, too many traders trade blindfolded, while those who know ahead what will happen, will profit the big bucks.
"There's a sucker born every minute" - P.T. Barnum
1 trader viewing now