"The JUICE" - how do you separate skill from luck?

Hi,

I mean I have my ballpark estimates.. I would say, quickly.. if someone can make 100 full-size (100% account placed per trade) trades, with good leverage, trades to be almost evenly places in the 3-6months... and having like 100% return with max 50% DD... I think I would analyse it further
What would you say is a good

I mean yeah, there are statistical methods... People even play with golden risk:reward ratio... but if you watch closely, this is not a defining edge because otherwise.. one would just put both to almost limits... its like SL0.1 (spread paid) and TP100.. would than seems to give you enormous edge, but it is just not true..
Also some people TRENDSURF... or "see" the pattern that seems to work some of the time or for some time...

How do you know. I mean, I am interesting in your IDEAS and also if anyone of you here is very good in MATH-STAT that could PROVE/estimate the probabilty...
For example what are the odds... counting TREND (help) and all.. that someone can make 300% return with 50% DD, with 100 100% account size trades...
Placing only SHORTS in overall FALL market...

I mean I have seen, as I have said.. beginners that TRIPPLED account, they didn't place SL, so they exposed to margin call... very "unfavorable" risk ratio but in overall.. they did TRIPPLE the accoun in less than 3months... and then a few margin calls came and... OVER...

So please tell me... what would be your GOLDEN parameters to decide or even "KNOW" if someone REALLY have it outside of luck or temp-skill etc...
and if you can really estimate the probability.. For example.. what would be the "lottery" level probability... to make 1000% return with just 10% DD?
I have checked some of the TEs here, in the LeaderBoard and they seem to be outstanding, I wonder what is the real statistics behind it.
Also I have checked some boards on other forums and some have unusual good performance, even too good to be true, at least to me.
I have made more than 2000% return in a series... but I still can not confirm or reject if and how much of it was favorable setup, trends, etc.. and how much of that was THE JUICE..

Also, as some people say that in forex, everyone lose, it is also true that BOTH sides can win... There are many examples, even today.. if two people would place
the order at the same time, setting TP20... (not placing SL).... they would BOTH win... because market could swing to one TP and then to other.. And altough they were both placed TOTALY differently, one LONG one SHORT.. they both seem to "win".. So how do you select the real winner overtime.. By FEEL or by STATISTICAL proof (counting in the risk ratio, number of "trials", trends, "randomness" of NEWS etc.)

In the long run, everything cancels out and it comes down to statistics, so what would be the JUICE level without a doubt or very little.. let's say in a 5-7 sigma certainty? lottery certainty or Higgs boson certainty Well I know the obvious one... Placing 40 100% account sizes at 500:1, TP20, SL20, 100% win rate...
would make you TRILLIONAIRE after 40trades (40 profits, 0 loses)... This is clearly not possible or if someone can make it happen, doesn't matter even if it was just luck at that point But what would be the minimum parameters that you can say... "there might be something" there or "there is" beyond reasonable doubt?
100 trades, 100mio volume, 100:1, 3months+... 100% return, 10% DD?

You are mixing up the wrong numbers imho.

Keep it simple. Just filter every(!) system with a live trading history below two full years.. there you have your separation.

I was just giving some quick examples, to show the direction (parameters) that could be to look for.
Could you provide example of what you are saying? What do you mean exactly.. account that are recent but history should not be older than 2years?
And what to look for, how would you know which one of those "have it"?

I am all for simplicity, I primary look RETURN-DD, then timeframe, volume.... And then of course I would be interesting in analysing trade per trade, to see the price prediction "power", how many times were close to SL, how many PIPs were underharvested etc... But those are details..

For the ballpark, what parameters would you look for? We can get a clue for example what fxbook is looking for for autotrade... but still, if anyone has any ballpark parameter like accountants that usually have the feel of a companies records by checking the few numbers, but it takes years to get that feeel

Hope I was clear.

You are making it way to complex and unnecessary difficult.

I gave you the parameter to look for. Did not expect to be mistaken here.
All your mentioned criterias are more or less unimportant. It does not matter how many pips you are gaining per trade. You can not make any conclusions on the risk/reward ratio by analyzing the TP/SL relation as well. Stop that.

The only thing that counts is the durability of success! Whether you are trading manually with every entry based on different and evolving factors or you are relying on an automated algo that is doing the same thing over and over. ONLY if there is a profitable history of more than two years you can start to count that as skill. Everything below can simply be called as luck.

Well yea I do agree on durability, consistency... but few more parameters would be great... being profitable over 2years is not that big parameter to me since there is too much room there. One could trade different sizes, had "luck" in one period within 2 years and than reduce the sizes considerable so even if they lose overall... he will still be profitable. There are many tests for different areas so they can separate skill from luck. I just though any one of you know some, true one. I have my own set of tests, just wanted to hear some of yours. I also checked some opinion from statistics people but there different opinions of course.

I'd look for a Sharpe ratio of 2 or more, Sortino of the same. The volatility of the returns is also important - how smooth is the equity curve? Incredibly smooth returns are likely (but not necessarily) forged, very volatile ones suggest luck playing a factor.
As for track record, the time needed depends on number of trades taken - I'd want to see a few hundred. Low leverage. Low % max DD.

They do exist - check out some of the trading leader boards outside of FF.

What I would never do is simply filter by % return, unless you're keen to throw your money down the swanny

I learnt a quote " I am a great believer in luck, and I find the harder I work, the more I have of it." No doubt good luck can bring you fortune but it does not happen to everybody, yes if someone works hard luck shall surely come to him sooner or later.

If some trade with proper analysis and money management he can have good trades in his favor instead of one who purely wait for his luck to shine.

Do you belive this is luck or pure skill?!?







Yeah its just a really small account but the point is how many of us could trade 9x the deposited balance in just hours?it really crazy mind blowing scalping technique

Unlucky people think skilled people are lucky

tnx! this is the response i was looking for... much more in the direction I was hoping for... if someone could estimate/calculate real probabilities for different parameters you mentioned, that would be even better...

but could you maybe give sam example numbers for the parameters you mentioned.. not the extreme ones but entry level for strong chance of trading being skill vs luck? you mentioned Sharpe ratio of 2 or more.. so what would be your entry-level at which point you would be convinced someone has quite a skill (even on top of luck maybe)?

I do agree on the return part, not sufficient parameter alone, not by a far.. UNLESS it is very extreme... or whats more.. at extreme return, it doesn't really mater
For example, if someone could do FACTOR x1000... per year (and year after yea).. I don't even care how he does i and what the parameters are... not before I get that return, than I would of course want to know Because some returns are just very low probability.. I would like to estimate this, considering other factors... Very hard to model/isolate luck out tbh.

No hard and fast answer I'm afraid. So 2 for sharpe and sortino would filter out a great proportion of them. Smoothness of returns is a relative term, leverage I'd expect to see professionals using no more than 10, even that's pokey.

If you're looking to invest in someone elses strategy, the best you can do is compare it what's already out there, using those factors to work out which is better than another.

The best filter ever is to ask a trader why their strategy is profitable and durable. If the answer's not concrete, they're winging it, period. But if you're investing, that's not always possible to ask, so the above is a good shorthand. Oh, and never, ever, ever are videos of anything more than marketing spiel.

I also try to divide trading into components...

1. PRICE prediction.. did it price reach (at all)? How soon? How much/long it went in counter direction before it did go into the predicted one
2. HARVEST strategy... How much PIP (TP) was extracted per trade... and how PRECISE was it.. was it almost spot on (just reached TP before it went to other direction again) or was it much underharvested... For example, there were like 10-20pips (or more) left... Could this be solved by TS for example?
3. RISK management... what was exposure per trade... overall and also intratrade exposure... what TP/SL level was it... so exposure... Would it be different, overall picture, if different TP:SL ratio/levels would be used?
4. PLAN/DISCIPLINE... would the trade hit if the trader wouldn't manually close it.. how many trades were emotional ones, for example: chasing the loses or stopped the trade because of fear (but would hit) etc..

etc etc..

Because many traders HAVE some sort of skill... even though they are in a NEGATIVE P&L... The trader needs to have a good of components in order to be profitable.. in a longrun, somewhat consistent etc...
For example.. you could PREDICT that the price will go to 1.1190 but set SL to tight and the trade goes bust because of one spike/news or whatever.. and it could be seen is failed "signal"....

But yeah, before I would go to analyse in depth... would need at least some indicator that there might be some potential..
for example I would like to see at least DOUBLED (demo) account at 50:1 leverage, preferable trades made of for example (set) TP10-20pip (equal) trades... and 100+ trades would be appreciated. LATER I would like the analyses.. how many of those trades were helped or busted because of the major impact news etc...
Or was it simple trendsurfing?

I mean yeah there are some guys that start FX... at "just the right time" when for some reason there are some (temporary) patterns... for some time and they "recognise it" and trade it... This can lead to multiplication of the account balance... but when this pattern changes... BUST

So for a ballpark I would like to see
100+ (equal size/TP trades... not just for example.. few LUCKY and then added fillers just to boost stat)
At leverage 50:1 for example..
over the range of 3 months..
then compare the trades direction vs trend... was it just trendsurffing, did trend helped?

BTW if you can make 5+ (for example TP20) profitable TRADES in a row at 500:1 (stop-out at 10%).... you are a WIZARD for me... IF you can repeat this sequence for example once every 20 trials and do that a few times per year. But this would be so extraordinary that wouldn't matter if you are just lucky, just need to exploit it while it lasts, analyse later Send me a msg if you can do this

Before I give some examples is that at this point the question is only skill vs luck or if ONE has THE JUICE... Not what exactly is one doing but just if it is distinguishable from luck, trendsurffing, temporary pattern recognition etc...

Those are interested articles that I find useful:

COIN TOSS vs TRADING
https://www.investopedia.com/article...t-formulas.asp

IF (big assumption) we assume trades to be coin toss, then this is very useful tool to get the chance
https://www.wolframalpha.com/widgets...db1069cc52464f
For example if you can win 76 out of 100 coin tosses ((equal)trades)... that would give you probability of occurrence of: 1 in a 15million... (assuming trades=coin toss; no trend)
That would put you on some LOTTERY level probability... Trades should be equal.. TP20 or TP20 possible...
Here is the confidence interval, for ilustration: https://en.wikipedia.org/wiki/Standard_deviation

Can someone calculate (or provide the link to it) the TREND in so the trend can be calculated-in? Because sure, if someone is trendsurfing (at those levels), he might get this kind of win-rate...
I have been searching for this correct probabilities (for forex)... but haven't found the good probability calculator/example for trading in trend...
Would like to know what kind of win rate, out of a 100 equal trades... would give you "benefit of the doubt" and which level... "THE JUICE".
And much more I would like to read the assumptions.. for example proof of coin toss, corrected for trend...
If anyone strong in math-stat, much welcomed, can msg me privately also. Or if anyone have a good link for the matter (simplified, conclusions), you are most welcomed to post here.

WHY this is needed? Because there are many ways to be temporary (very) profitable... For example, you can make ONE trade... and have 10 fold return..
All you had to do is SHORT/BUY at the right time... TP200... at 500:1... AND hoping it does not get to -20PIP first. And since you can open many accounts, this ONE TRADE...even though being 900% would mean absolutely nothing! It is much harder to get that return with many many different trades and climb there slowly... if you can get that far at all... So I like to have some example to ballpark the game.. It would be good if someone would make equal trades, using same strategy...
and if above certain probability level... using certain strategy (PA, news, tech advantage, scalping, fundamentals...).. I think we would have the picture of the player's real ability after about 100trades over 3months... This seem to be the test I am the most fond of... If anyone can make 100+ equal trades, over 3 months... calculating TP/SL (ratios) AND TREND! in.... and prove win... there might be something to work with there. But must be very careful with attributing WIN rate to probability because for coin toss calculator.. TP and SL would have to be even and no trend.... TP, SL, TREND can be different but have to be calculated-in properly...
For example if you are using different SL levels to TP level... the probability for TP changes per trade but in the endgame... this have to be adjusted... so if you get more than 50% WIN rate but SL>TP.. this would have to be adjusted...

Ok so if anyone strong in this or have some good example/source, with summary and optional additional info... welcome..

“Luck is not chance, it's toil; fortune's expensive smile is earned.” ― Emily Dickinson

Good luck.

The thing is also to CUT out the timeframes when something does (not) work..
For example, if we would filter out just the ones that survived after certain period.. we might miss heavy volume/profit trades that someone have/d that were extremely profitable... it was just that the trader overtraded.... through different times where "he hadn't"...

The market was (I think) proven that is NOT RANDOM... but CHAOTIC... meaning.. some things works some of the time...
I compare this to "burner" strategy... I see certain strategy over the span of a year... If someone could roll 10FOLD (per week).... 5+ times a year... it would be a good chance to take a chance every week... because in a year you would get overall profit... So that would be a skill to me... although it it would have NEG TIME return.. Why repeat, because it only works in a very good/smooth condition, because when trading for that high return, this ussualy means tight leverage.. so there has to be as little noise as possible (unless you trade the noise) The strategy would be burning deposits to the ground unless it gets to 10fold.. and repeat...
This is where the usual report would (maybe) fail to recognise the potential The true profit would show in absolute return.... That is why the chart is very good recognising this... If someone have spikes, run-ups... maybe we could leverage it up... burner strategy.. and harvest (overall) profits...

I am saying that it is hard to recognise skill/juice.. because someone could just be trendsurfing and have good stat, profitability etc... but could be blown after (changes) and never recover... It might take long time to that, specially with open trades.... But you could also have a trader that is also in a neg.. but he might have excellent certain trades... IT taks a lot of components to be profitable... just one paramater that is criticaly wrong could make overall NEG.. O I watch many things, but trader must get to my attention first otherwise I would just be wasting time analysing every one that just wanted to play a little.

So if anyone have some test/model, welcome...

P.S. but for REAL juice.. if anyone can get 10x TP20 within SL20 (at least once a year; after many trials)... he has the juice in my book and PLEASE msg me immediately... it would be worth to try/burn (deposits) some/many times for this... srs