Forex Factory (https://www.forexfactory.com/)
-   Trading Discussion (https://www.forexfactory.com/forum/11-trading-discussion)
-   -   How can an "edge" wear off - Shouldn't it be the opposite? (https://www.forexfactory.com/thread/1024492-how-can-an-edge-wear-off-shouldnt)

ryuryu Sep 24, 2020 12:30pm | Post# 41

I don't understand how releasing a working strategy to the public can make it stop working
The question is not how, but why or what for?
For example you spent 3-5 years of hard tests, lost thousands money and finally found you strategy. Now you... yes! Exactly. Now you go to FF and start sharing it, debating, demonstrating, arguing.

Oh wait! There is another option. You start to sell your great EA for 30$ lifetime license. Or start your learning course. Because you don't want to run your great strategy and earn money, you want to help others and make the world better!

And of course while you creating your great golden deluxe strategy you are thinking is there any liquidity in the world that can handle my great strategy? What if (it is so great) it can break the market? Or programmers will stole it... or whales will find it and change their algos! Oh my god so many questions!

Tmaniss Sep 24, 2020 1:40pm | Post# 42

You can either have a permanent edge or a temporary edge over the market. The temporary edge is a fad. It doesn't wear off, it was never relevant. A permanent edge can only be built on the characteristics of the market (both explicit and hidden).

havo Sep 24, 2020 1:50pm | Post# 43

No one wants to listen or do their homework.. amazing

the edge its in the price in itself, not even "fundamentals" or whatever you want to think matters, the PRICE leaves you clues of what its going to do next; if you learn how to read it you will always are going to be at the right place and time, thats all you need to make money (well, that and know how to handle a trade but thats a different history)

And yes, its that simple!! i never understand the desire/necessity of over complicate this activity. there is no need for that and you wont get better resuts doing it, period

Seneca pilot Sep 24, 2020 1:50pm | Post# 44

You can either have a permanent edge or a temporary edge over the market. The temporary edge is a fad. It doesn't wear off, it was never relevant. A permanent edge can only be built on the characteristics of the market (both explicit and hidden).

There is a lot of insight in this post.

It would surprise most traders how often someone can get lucky for quite a long period of time. The trader who adds to losers can survive for quite a while when volatility is low and markets are trading in ranges for extended periods. A lot of stock traders thought they were market gods from 1997 all the way to late 1999. All you had to do was buy an internet stock, any internet stock. Most of them went broke in the summer and fall of 2000.

godlike Sep 24, 2020 5:45pm | Post# 45

In my opinion, people who say sharing their strategy will remove their edge are either lying or do not understand how the forex market works. The idea that sharing your strategy will create competition in your entries is pure crap. There are lots of great traders here, who have been sharing their strategies for years and not once have I seen any of them complain on it losing it's edge or whatsoever.

So I don't think sharing your strategy, makes it lose its edge or effectiveness except there was never one in the first place. The best strategies I've seen, are based of the price itself, just following what it's doing, so I don't see how any kind of edge will be lost if more people knew about it.

joyny Sep 25, 2020 3:32pm | Post# 46

1 Attachment(s)
I'm a coder. When I see any "promising" strategy... which performed great 10 years... I view it as an investor. I view any strategy as "just another business opportunity". It's like investors viewing regular companies how they performed in the past... ok, great. Your company sold products and made some x% yearly profits. But well... that's the past. Who knows the future? There might appear any obstacle in any business... and everything can (and will) go wrong with more or fewer drawdowns.

In the banking business, there are theories of how portfolios of stocks must be made. In short: never rely on 1 strategy. Diversify. Any strategy can/will fail. Make packages of running strategies with various risks/rewards, entries, exits etc rules.

A package of 10 EAs will have 1-2 winners, 1-4 total losers, and all the rest +/- nothing (not good and not bad or compensate each other). Those 1-2 winners will cover losers and make some profit.

Then comes another layer... you need to have ~10 packages with EAs - so total about 100 EAs setups. Because there can (and will) fail the entire package... and then like it is in the insurance business (which 100% relies on statistic too by the way) there comes next layers... any insurance company is reinsured by bigger. And any bigger is again reinsured by a huge one. So it seems needed to have about 1000 EAs setups... well that's too much for 1 trader. It can be done by hedge companies or banks which have teams of coders for this.

In statistics there are fluctuations. If any process runs with some success rate lets say in 10 years average 60% profitable trades then there will be months and years when that profit will be 45% and periods when that profit will be ~75%. So on average, it is 60%. But those periods can last months/years. You never know when that statistical fluctuation in which direction will happen to any strategy.

See here a screenshot from my ~90 EAs (different strategies/setups/symbols risk/reward etc - no martingale, no averaging, all with safe SL) running on 9 accounts on 3 brokers:

Click to Enlarge

Name: 90_EAs_8_accounts_3_broker.png
Size: 110 KB

And you see some packages perform bad/modestly and some quite good. Some of them attached to my TE (which are on MT4, the rest on MT5 which is must-have for optimizations/backtests).

I gave you "the edge" I am using but for you, it is impossible to reproduce (still you can build something similar - different sets of strategies, proportions, rules, etc). I built this system of EAs several years. Made a lot of optimizations/backtest. Made special tools to analyze data and calculate the combined risks of packages. I see that there is no get-rich-fast possible... with compounding maybe 7-10 years to accumulate something meaningful (if broker won't go to bankrupt... therefore diversify brokers too... split your capital)

hasib Oct 18, 2020 1:18am | Post# 47

{quote} What I wanted to say is that the psychological problems are not real problems. The are mostly a result of lack of knowledge and skills. The people who are not programmers think that they fail because they cant follow their own rules. Their common believe is: "I broke my rules again and I lost. My strategy is good. If only I could follow my rules perfectly I could make millions". On the other side, the people who are programmers and understand statistics and probability theory already made this strategy into a robot and backtested and forward...
I agree with you. Most of the trading systems don't have an edge. People who are trading demo and selling courses claim they can read price actions yes they can read but in the real world, they fail because so-called price actions give a lot of contradictory signals. When I see a man with too much confidence in their trading and start giving BS advice I immediately know he is a new trader. The market doesn't make him humble yet. I can assure you that the market will make him humble by taking his money.

Endoephemera Oct 18, 2020 4:03am | Post# 48

I don't understand how releasing a working strategy to the public can make it stop working If a lot of people trade at the same time in the same direction, price will go in the right direction strongly. The only problem can be a diminution of liquidity, but I don't understand how it could affect the edge ? It should enhance it at best, no? please enlighten me thanks Jeff
Any popular and easy-entry exit signal used by retails is used by larger participants against them. You see this all the time with trendlines and SR levels. Those levels and consequent patterns are created and then broken exactly to put common expectations on the same one side and then overrun their positions sending them in stop, once or repeatedly. Even when an easy entry on a "flag" retr. (for example) is played out as in textbooks, and leads to trend continuation, it is usually quickly tested again to shake out those who entered there.

What, ex post, turn out to be the best entries, at swing reversals, are the hardest to play because they happen when the vast majority is on that swing. Then something magic happens, somehow a mountain of liquidity appears and stops the move and reverses it, or leads to a fast and deep retr.

A lot of what goes on all the time in the markets is going against the easiest expectations of the weakest participants.


That said, you can publish a profitable strategy and have it working long term as long as few retails play it. Once it becomes really popular, it will be erased purposefully; not because the big players track the forums but because they see the retail positions.
A really significantly profitable entry-exit strategy, one that reverses the markets and dictates the swings (by definition) cannot be in the public domain because then there would be no losers, which is not possible; you need loss takers, to take profits.

mr.brown Oct 18, 2020 4:12am | Post# 49

I don't understand how releasing a working strategy to the public can make it stop working If a lot of people trade at the same time in the same direction, price will go in the right direction strongly. The only problem can be a diminution of liquidity, but I don't understand how it could affect the edge ? It should enhance it at best, no? please enlighten me thanks Jeff
fap turbo was an asian session scalper robot that was trading session slow moves with eurgp. then suddenly it becomes famous and many robots took the same trade at the very same time so it created its own slippage and become unprofitble
if you are trading hft eg then there isnot much room for others to joın

however medium and long term strategies might work longer but for whole, wear off is inevitable

every system works, until it does not

mr.brown Oct 18, 2020 4:14am | Post# 50

Quote
Kaufman: I did that for years. Absolutely yes. But during the 1990s, I was successful in what is called opening-range breakout. With an opening range breakout, you let the market establish an early range in the first half hour, hour, or hour-and-a-half, and then when prices breaks out, you go that direction. That worked really, really well for maybe 15, 20 years, from 1981 to 1999. Now it doesn’t appear to work at all.

hasib Oct 19, 2020 7:39am | Post# 51

{quote} Richard Dennis and his famous trend following turtles story is full of shit. Blind Trend following based on technical analysis does not work in today's markets and it barely worked 50 years ago and this is super easy to proof for anyone who can write code and know how to build and test automated systems. The trend following systems in general are the easiest systems to code because their rules are very simple and straightforward. There is no complicated pattern recognition. The definitions for trends are simple and clear. So if these systems...
As an Algo Trader, did you backtest Pivot based trading strategy? what is your opinion and findings of the pivot? if you share your statistical findings it will help us.
Thanks, brother


© Forex Factory