Important – Why systems are profitable (or not)
I’ve just had what I believe is a significant "Aha! moment" that I’d like to share with you all. It came as a result of my reading the posts here and here.
These guys are (apparently) trading profitably by turning conventional "wisdom" on its head. FXTitan is advocating that we "cut profits short and let losses run indefinitely" and StevieT is using an RR profile of 1:20, i.e. he is setting profit targets of 50 pips and stoplosses of 1,000 pips. I asked myself "How can these guys break the established rules, and still be profitable?", and attempted to probe deeply enough into their rationale to learn why.
My conclusions follow below. Be warned: to fully grasp all of this, you must think beyond all of the "textbook" maxims surrounding entry, exit, RR and MM, and apply lateral thinking, just as I did. If so, I can guarantee (well, almost) that what follows will revolutionize the way in which you think about trading methods. I hope I can express myself simply and clearly enough to make this post a worthwhile read. OK, here goes:
1. The types of order that we use (buys; sells; market or limit orders; profit targets; stoplosses; hedging by taking long and short positions in the same currency pair) are effectively irrelevant. All order types simply add to, or reduce, our net overall position (long or short).
2. To profit, we must be net long while price is rising, and/or net short while price is falling, frequently enough to overcome costs (spread + swap). This applies to all methods, "conventional" or otherwise, without exception, at all times. As long as we are achieving this, our account equity (which is the sum of realized P/L from previously closed positions, plus unrealized P/L from currently open positions) will rise; otherwise it will fall.
3. For as long as our account equity continues to rise overall, we may justifiably assume that we have an "edge". To achieve this, we need to somehow time our orders (i.e. adjust our net position) accurately enough around market reversals, on balance, to be net long while price is rising, and net short while it’s falling. This is the bottom line regardless of whether we’re (in conventional terms) entering, exiting, scaling in, scaling out, realizing a profit, realizing a loss.
4. The result of any one trade is insignificant (if we are "hedging", or scaling in/out, with positions offsetting each other, the idea of an "individual" trade is effectively meaningless, anyway). It is the overall effect on account equity, i.e. net P/L, that is the bottom line.
5. RR (in the conventional sense) is effectively irrelevant. Firstly, TPs and SLs are effectively "offsetting orders" that bring our net long/short position back to neutral. Secondly, everything else being equal, the further away we place our TP from entry than our SL, the less likely the TP will be reached before the SL. All this ultimately amounts to is a compromise between win size (RR), and win rate ("batting average"), i.e. everything else being equal, win size and win rate always operate in exact inverse proportion to each other.
[Note: Since markets are driven by intangibles (mathematically) like greed and fear, the probability that a trade will succeed can’t be calculated in advance. Mathematical edge (or expectancy, or profit factor, call it what you will) can only be measured in hindsight, as the product of win rate and win size.]
I’ll try to explain my point about RR with an example. Let’s say we believe price, currently rising, will likely reverse at heavy overhead resistance. So, following conventional maxims, we set a tight stoploss ("cut losses quickly") just above the resistance point, and then let profits run, as price falls, using a trailing stoploss. However, expectancy is the product of win size and win rate, and the latter is determined by how frequently our assumption of a reversal is correct. That relates to the efficacy of our analysis and forecasting. It's the extent to which the forecasting goes beyond the inverse balance of RR and win rate that provides the edge.
Since exit is effectively nothing more than an offsetting order, entries and exits are ultimately mirror images of each other (just like Merlin said here). Hence, to improve overall expectancy, exits must be likewise somehow be timed accurately around "probable" market reversals, i.e. as described in point 3. (The only difference being that, if we use "conventional" RR/MM techniques, stoploss combined with position size is used to allocate a pre-determined maximum risk with each trade).
So what about "let profits run, cut losses quickly"? Again (if we revert to thinking in "conventional" terms) "let profits run, cut losses quickly" is going to operate profitably to the extent that prices trend. In a trending market, it is a profitable strategy; in a ranging market, the exact reverse applies (more on all of this here). The key is to somehow apply the correct paradigm, as explained in points 2 and 3. (To simply apply the conventional maxim in all situations – assuming that it in itself provides an edge – is to presuppose that forex "trends" more frequently than it "ranges").
If I may digress further..... if I understand correctly, the "let profits run, cut losses quickly" maxim originally applied to equities, and has been "transplanted" into forex. More on this here. My observations are that while stock trading is purely speculative, allowing prolonged trends while positive sentiment escalates (e.g. the dot-com boom), forex tends (on balance) to revert to a mean, (1) in the long term, due to national economic boom/bust cycles, and government/bank intervention,when economic indicators reach "undesirable" extremes, and (2) in the shorter term, there is mean reversion unless/until news announcements generate the sentiment necessary to move price up/down to another level. (However, I realize that "trends" do occur in forex, on every trimeframe, and that the reality is much more complex than this generalization).
Finally, I'd like to recommend two posts that (IMHO) are both excellent, and relevant to this discussion:
"……It is a MIRACLE that we are making ANY correct statements about a market that was at one point thought to be inherently chaotic and foolish to participate in. Remember that it took several academic papers in the Journal of Finance to prove that the market is not a random walk model.
In fact, the market (and any similar exchange) is inherently chaotic, due to the large number of unrelated participants whose actions both commercial and speculative cannot be previously determined, and whose collective effect, weighed by volume, is even harder to predict. It's a crazy game we are playing if you think about it. Just because you have charts that seem to tell a picture in hindsight does not mean that anything makes sense at all. This is an unknown function with a built in random factor, and we're trying to ride it. F*cking crazy.
At the same time, there seem to be recurring patterns in the market activity that repeat themselves, and which do form both a statistical and an intuitive basis to make time series propositions (aka trades), but that is a topic for another discussion. ......"
"I believe most traders - regardless of them being forex traders, futures traders, commodity traders, or stock traders - fail because they don't have a real edge.
Most of the reasons I read from forums, books, articles only focus on discipline, money management, adequate capital, experience, etc...
However, if you only have discipline, money management and enough capital to trade and you still are missing a real edge, it'll only be a slower death.
I also believe it's a lot easier to be disciplined when you have a real edge than when you don't have any edge.
If you seem to be on the wrong side of the market most of the times, the temptation to tweak your system around is always around the corner. How can you be a disciplined soldier if you have no reward from what you are doing? Unless you are a masochist, I can't really see how someone can stay disciplined using a losing strategy.
What about experience? Well, experience can just help so much. If you still have no edge, your experience will be brutal and will only tarnish your confidence permanently.
In conclusion, a trader fails because his strategy doesn't provide him with an edge good enough to make money.
Many losing traders think that money management can fix this handicap and try their luck with "averaging down" until they blow up their account and go broke.
If you don't have an edge, don't trade. Forget about ratios, money management, discipline, experience...it's all useless if you don't have a real edge."
What provides that edge? Refer back to points 2 & 3 above.
Hope all of this makes sense.
Thanks for the deep thoughts. Especially the above quoted section is totally new to me and really Great. I know that a lot of analysis will be different for Trending and Ranging markets, but never thought it will apply for SL/TP also. Thank you.
David, I've updated that post to include something I've incorporated since the New Year. http://www.forexfactory.com/showpost...postcount=3201
Very well written post again but I wouldn't expect anything less of you
I have just begun to look through posts on Forex Factory again since a long time. I gave up on this site for a while, just checking everyone once in a while. Its true that even logging onto FF can be a paradox. On the one hand you are finding systems that may or may not be profitable, however without a system one is completely hopeless. But on the other hand FF is so cluttered with undefined systems and only a few that seem to be established.
Anyway I don't know why that blurted out but I have been reading alot of your more recent posts. The older ones about the mathmatics and logic, seem to be over my head unless I were to read the whole thread which I currently don't have time for. I appreciate you and your opinions that you have expressed and wil be coming around more often now that I have found that you are one of the great minds here at FF. Thanks again
Some of the reasons
Once you do find a system that does work, don't forget Axiom 3:
Then the question becomes discipline vs. tweak the system....
Regardless, you have to be quick to react to the market, on many timeframes. By this I mean that if your system works well for 3 months, then doesn't for 3 weeks, 4 weeks... how long do you trade it before you change it? Or do you change what you do based on the way the hourly chart is behaving? Or do you ride out several losing months and hope your year ends up OK? How long do you keep faith in your system when it fails, just because you manually backtested for 3 years and demoed for 3 months before using it live? Maybe you stick with it based on Axiom 6:
It can be difficult.
A person never stops learning. However, any time you study something new, you can't help but have it enter into your trading to some degree. So your system is constantly evolving as you grow. (Unless it is a pure mechanical system, in which case it will not work over time anyway - see Axiom 7
http://www.forexfactory.com/showpost...5&postcount=62 ) .
I think my problem currently is a habit of forcing trades. I have two systems that is and of themselves are inherently profitable. If I were able to just be patient. HOwever when I get in front of the computer at the wrong time during a session I find myself gazing looking for an area to place a trade. I WILL remedy this. I know I can and will. Its frustrating to know you broke a rule and will probably do the same thing at least a few more times befor eyou learn your lesson.
I agree that the "Trading Systems" area is littered with a myriad of different ideas, many of which are conceptually mutually exclusive. With the vast majority of these, the pattern seems to be that (1) a trader wannabe posts his idea, seeking confirmation; (2) over several dozen posts, many well-meaning folk suggest untested variations that cause the original idea to be distorted - an attempt at "conceptual curve-fitting"; then (3) interest eventually dissipates, and the system is (apparently) discarded. But the uneducated can spend a huge amount of time being a faithful disciple of (m)any of these systems, believing the current one to be the HG, only to have their hopes dashed over and over. At least, that's what happened to me. To speed up the process, I attempted to forward test perhaps 20 of these ideas simultaneously, only to become horribly confused and mentally exhausted.
I am continuing in my quest to find that profitable methodological edge.
One of the best threads on this forum in my opinion. Thanks for starting it.
Sorry for the long disertation below:
As much as I hate to even say it, there is NO system or methodology that will give you an edge. The only edge we have as individual traders is PATIENCE! Taking only high probability trades takes patience. Most of the traders that I know personally trade only 1-3 pairs of currency. WHY?? Most brokers have at least 20. Limiting yourself to a small number of pairs can create frustration if you are only looking for higher probability trades. Mentally you start to struggle with yourself thinking that you HAVE to make a trade. Don't limit yourself in your trading.
Second... Most "systems" either on here or other places are very limiting in their methodology. "Only use it on this pair or on this timeframe". To me that is not a system but merely a fluke of nature that will eventually correct itself and make the so called system obsolete. One of my two mentors has shown me the importance of working within multiple timeframes and with any currency. Sure each of the indicators should be adjusted for the timeframe or even for the currency pair (due to the differences in volitility or daily price movement). When I developed some custom indicators for him, he requested that each would adjust automatically for timeframe and currency pair. Trading in multiple timeframes is where patience really starts to play out. Try this sometime, place an 8,3,3 stochastic on a currency pair (doesn't matter which one), do this on three different timeframes (15,30,60 for example), place all three side by side on your screen, watch what happens when all three line up either overbought or oversold. It usually happens 1-4 times a day (on a 15,30,60 less on higher timeframes of course), but the results are a higher probability trade.
Next, we spend far too much time looking for a PERFECT system. Notice I have said "higher probability trades". Even when multiple timeframes line up occasionally it will go against you. So what? One of my mentors calls this "paying rent for your business to operate". LOL! It is how you manage those losses that matter.
We tend to give up on systems too quickly. Even when one trade goes against us it is human nature to start questioning the system. Does it really work? etc. Two trades in a row going against us and most likely (if we don't fully understand how a system works) we are looking for a new system. It took nearly 4 months of FORWARD testing on a demo account for me to fully believe in the trading environment that my mentor set up for me. I kept detailed records of wins and losses and how big the wins were and how bad the losses were. Somewhere during the third month I started to see the magic of consistency.
Can a purely mathematical or mechanical system be traded profitably? Yes, absolutely. I have a clearly defined set of rules that I trade by, if all do not line up, I do not trade, PERIOD! None of those rules are subjective, all are based on indicators from multiple timeframes. Because of my software developement agreement I cannot disclose the system I use, sorry.
I could go on and on and on. LOL! Maybe it is all the drugs I am taking to get over this stupid cold I have that are making me ramble on. Who knows, hope I made sense back there. LOL!
By "system" or "methodology" I didn't mean to exclude approaches where one must wait for several ducks to line up, before pulling the trigger. I just finished watching a coaching video where the instructor said that the most successful trader he knows trades only one setup, and will wait weeks, if necessary, for it to occur.
Conversely, there are folk here who believe that we can profit using completely random entries, i.e. managing exit in itself will give us a decent enough edge, but I can't see how that's possible.
Having said all that, I think that there's possibly a tendency that when we find a method that works for us, whatever it might be, we develop a kind of "tunnel vision" where we start to believe that ours is the only profitable way to trade. Hence there may be "systems" out there that deliver profit in ways that we thought to be impossible, but by ceasing our search we will never discover or develop them. Who would have thought that FXTitan, StevieT (and possibly others) could be profiting by flouting some of the most widely held trading maxims? Hence there may also be reward at the end of the tunnel for those who are willing to think outside of the conventional squares. As Jack Schwager says in his Preface to New Market Wizards "There are a million ways to profit from the markets, but the irony is that they are all hard to find". The book was written in the early 1990s, but I assume that its wisdom still applies today.
And, what if, after weeks of waiting, the guy misses the trade? DOH!
(I expect if it's that much of a long term strategy, the entry isn't terribly critical, so maybe the guy's allowed to sleep during those weeks....)
Hey, whatever works...
If an EA can do it, then it's mechanical. If it's mechanical, every trader should get the same results with it as long as the rules are followed to the letter. As soon as someone doesn't, they're a discretionary trader.
That's my take on it.
No doubt there are some great stories out there! Read into this whatever you will. I'm just the messenger..... please don't shoot me!!
Yeah, I think I could live on 1 good trade per week, with 200 - 300 standard lots, hell yeah! Count me in! Just gimme 2 pips per trade and I'm good to go! (Actually, maybe he isn't that much of an expert trader now that I think about it.... but that doesn't mean he's not extremely smart.)
Reminds me of skunny saying you can trade 100%, but you'll lose a lot of pips doing it. Hey, it's a good strategy if you have the system and the capital - be the sniper.
The more I think about it, the more I'm starting to like it. Less time in the market = less stress. One pair, just check it every hour.... entries could be automated when the setup is near complete.
btw thanks for the thread, very interesting. I just discovered StevieT method yesterday, and I'm planning to test a similar but more conservative estrategy with EURAUD , maybe with RR of 1:5
sorry for the non useful superficial post :S, I'm just learning...
The danger is that you experience a lucky streak during the test, then believe that it works.
you may want to read David's (hanover's) other posts on the subject. It has been mathematically and logically proven that this will not work.
That might be more than you want on the subject, but overall, the reading should be informative.
IMO, which you already know , he's doing a lot more than just MACD divergence. I'll bet with trades that large, he has a couple favorite entries that are phenomenal, to avoid drawdown at $2000 - 3000 per pip!
LOL, it's worked for the last 6 months over 750 trades but OK
I know it works for me, my investor knows it works for him - 6 consecutive profitable months so neither me or he is complaining.
Are you saying you use completely random entries?
Is this the sl 1000, tp 50 method I read somewhere?
It is this method, however entries are not random - I trade within a range taken from the daily charts and am looking for specific points or areas of entry. However, due to the large stop loss (pip wise, not % wise), entries are far less important than most conventional methods.
© Forex Factory