website header 2

website header 2

Monday, January 23, 2017

YES, THERE REALLY ARE “NON COMPUTABLE” PROBLEMS IN TRADING

“Hey, let’s go hiking!”


Today’s blog post is really just a continuation of the ideas & thoughts from the weekend’s special update; specifically today, I want to write about the 2 dynamic variables that occupy our thoughts continually [whether you admit it or not] when trading, and that is “time” and “price”. The concepts seem simple enough; trust me, they’re not.

There is a very simple reason why I have never been in favor of EA’s [expert advisors], which are of course the automated trading algorithms that the brokerage house executes on your account’s behalf without any input from you; the dirty little open secret is that neither you nor anybody else can “input” all the data or parameters necessary in the computer code to make it functionally profitable like you want. It should be obvious to everyone [but isn’t of course], that no matter the genius of the computer code or those that write it, the code simply executes [in a linear fashion] mathematical and/or mathematical logic [e.g. “if”, “and”, “or”, “then”, “else”] instructions based on the “input” guidelines. Nothing more, nothing less.

So, for example, let’s say my algorithm for the scalper USDJPY algorithm has been meticulously and scrupulously proof read, thought about, agonized over, and finally coded into an EA. And since you know the “rules” for buying, let’s set up the following hypothetical trading situation.

All 3 lines on the algorithm are sloping up; USDJPY has 3 M1 down candlesticks in a row, and in the 4th M1, it is sitting just a couple of PIPS in the red from the third down candlestick. USDJPY is trading at 113.10, with the day’s high = 113.36 and the day’s low = 112.55. From the algorithm’s rules, you know that if the M1 goes green to initiate a long position, and so you’re sitting there with the order box up, waiting for confirmation. 

If the algorithm was “EA’d”, the computer code is also waiting to execute as well. Of course, you have no input into this as the code will determine mathematically the course of action, while you sit comfortably at the beach drinking a strawberry whatever, and basking in the future knowledge of untold riches.

Without warning, the market wires receive unexpected “news” from the BOJ [Kuroda & crew] and the market reacts violently; a few milliseconds earlier, USDJPY is trading at 113.10. Now, a few milliseconds later USDJPY is bid at 114.25. You immediately see this and your first natural reaction is, “WTF is this”? I doubt very few of you, and for sure me, are NOT going to be hitting the “buy” button; I want to know what the hell happened. In any event, I’ll wait for some kind of break if I’m going to buy.


But here is what your “EA” does; 1) it sees a higher M1 that goes “green” and it places a market order to buy, 2) the LP fills you at the high offer price, with a 9 PIP spread at 114.36, 3) you have instructed the EA to use a 7 PIP stop loss, 4) once filled the market gaps lower in 20 PIP increments back to 113.80 on “knee jerk” selling, 5) your calculated stop by the EA at 114.29 [market order to sell] gets “run” because the next bid after 114.30 is 114.07 and then a nanosecond later its 113.80, and finally 6) your stop loss fill is at 113.80 for a loss of 56 PIPS.
When you get back from the beach, what’s going to be your reaction to this? [“Cuz I know what mine would be”.]

So you tell me, “well –vegas, I would have code put in that prevents this; if the market “bid” goes up 20 PIPS or more, then it would not execute. Problem solved I think”.

MY RESPONSE: “Ok, the market just jumped exactly 19.9 PIPS on the “bid” side from buy stops above the previous high of the day; you are going to get filled because 19.9 < 20. And again, your stop of 7 PIPS gets immediately taken out and you still lose anywhere from 15 – 30 PIPS from 1) wider spread, and 2) lousy fill from the LP because you were buying at the market on the way up and your stop was selling at the market on the way down; EXACTLY OPPOSITE OF WHAT THE ALGORITHM CALLS FOR FROM YOUR TRADING.

YOU: “Ok, I’ll just widen the stops a little bit, from say 7 PIPS to 20 PIPS. Again, problem solved I think”.

MY RESPONSE: “Ok, you’ve just decided to “hand” the market more money. Your buy still gets “screwed” no matter what number you fill the code with; what happens if the market comes back exactly 20.1 PIPS from the high bid to stop you out and then spends another 3 M1 going down 4 PIPS before gapping higher AND doing all of this to you again for the second time”?

Meanwhile, as a human not subject to “forced actions”, I can sit back and decide what the EA cannot under any circumstances calculate; which is to say, “do I want to be a part of this or do I want to stay away? Is this going forward a good trading environment or one that is almost guaranteed a losing proposition?” 

And so it goes “ad infinitum nauseum” adding code to cover the “what ifs” to prevent your account from getting wiped out while you are away playing and counting monopoly money; “I hate being the one to break the news to you, but it will forever be like this because these “non computable” problems will always be here waiting for you. THEY ARE INFINITE IN SCOPE & YOU CANNOT ESCAPE THEM! … deal with it, please”.

And all of this brings me back to “time” and “price”; “what’s acceptable in terms of measurement and what isn’t? What is ‘volatile’ and what isn’t? Where is the ‘line’ for chop and when isn’t it there in the market? What time is ‘right’ for making money and what time isn’t? Where, oh where, is the line for optimum profits, cuz I most definitely want to be on the right side of it? And so here we are; you can’t code these questions because in essence they are simply probability wave functions where, until you finish a trade and ‘book it’, infinite possible outcomes are viable and present. 

In other words, you can’t code “common sense”; all I or anyone can do is build a “framework” from within which you can thrive based on favorable historical market probabilities that have existed and repeat consistently. With this proper “framework”, you can then do things that further increase your probability of success; 1) trade liquid markets only, 2) trade only those markets that have the lowest “net” cost to your account, and 3) use the volatility of the market you’re trading to your advantage by setting a PIP [or $$ per OZ.] goal and remaining disciplined. Nobody ever suggested you have to be in every squiggle 24/5; pretty much every day there is a “move of the day” to capture.


The M1 signal parameters have been changed from time to measure deviations in price from a slightly longer time perspective; in essence, every minute the algorithm is plotting approximate 15° and 7 ½ ° changes in Gann’s ‘Square of Nine’ to give us [in my opinion] a very optimal look at changes in price momentum as we go through our trading day with as few “false positives” as possible given a minimum level of intraday volatility.

Obviously, if the momentum of price is “up” [3 lines sloping up] we want to be long, and if momentum of price is “down” [3 lines sloping down] we want to be short. If any of the 3 lines are not in agreement with the others, we sit and do nothing. However, we don’t want to be “long” or “short” from just anywhere, we want to be “optimally” in a position that has historical probability on our side [70/30 to 80/20], and so our threshold is 3+ M1 closing candlesticks contra-trend to short term momentum before we enter.

Now comes the fun, because with each person sitting out there reading this comes probability wave functions based on 1) your risk profile, 2) your short term profit objective, and 3) your PIP goals, etc. Throw in the element of the “free trade” concept when appropriate, which I have mentioned before, and the results can be as varied as the people trading the same algorithm. So, when I get people who ask, “hey –vegas, what’s the exact, right way to liquidate?”, only semi in jest do I say, “tell me where you got out, and let me look at a chart, and I’ll tell you if it was the “right way.”

This is why, in an earlier blog post I said, “have no memory”. This trade, that trade, this day, that day … whatever … the only thing that matters is did you make money today and did you achieve your PIP [or $$ per OZ.] goal? If the answer is “yes” … no worries, no introspection, no “what ifs”, no nada … go live life [it’s called “balance”] and come back tomorrow and do it again. Cuz if you do, you’ll find yourself smack dab in the middle of that “earnings projection” table I have posted several times in prior blogs, and by definition you have “escaped to success”.

If the answer is “no” from above, it isn’t most likely going to be the market that spoiled the day … it’s gonna be you! When I talk to people, invariably something went wrong in implementing the algorithm … and then to compound the problem it was time for “Plan B”, even though there was no “Plan B” thought out beforehand. 

If you spend your time trying to figure out where any market is headed before you trade, you’ll win just enough to keep you coming back to the analysis while your account is getting bled to death. The only “analysis” that is proper is to judge yourself on how well you followed directions; and make no mistake, it isn’t just with my algorithm. If it pleases you, there are thousands of trading methods, EA’s [God forbid], systems, what have you with their attendant requirements and technical indicators; the vast majority you don’t use because you’ve looked at them and their respective limitations are glaring.

What makes people “uneasy” about trading, either as a professional trader who makes his living off of it or just being successful with a “plus” balance, are 2 things; 1) people don’t like to think their lives are “probability wave functions”, and 2) nothing … and I mean literally nothing … in your life either prepares you in an educational sense or allows you to experience what trading is like before you get into it. 

And while people dance around the edges by saying, “well, I heard of a guy once who did it … I read about a guy that did it … and every once in a while I’ll read about a guy who did it”, what they are really saying is that they can’t possibly figure out, even with shiploads of alphabet degrees or Pudding Business experience, how this works cuz there is no parallel in the universe they can relate to or go to for answers … so it stays that mystical mystery.

Of the really good traders I have known over the years, all had that “go with the flow” mentality inside their trading methods; if things changed, they went with it. They didn’t sit there and try to “intellectualize” remaining the same, while at the same time watching their incomes drop. You want to see a really pissed off trader’s wife or girlfriend? Go ahead, tell her she has to “cut back” her lifestyle for a bit and then give her the “intellectual” reasons why the market is not behaving. Yea, call me and let me know how that goes.

Bottom line is recognizing limitations that heretofore were unknown, and then doing something about it. You can have the worst algorithm in the world, and if market conditions are right for an extended period of time, you can make stunning amounts of money. Where many traders “trip up” is their lack of realizing this is not the end of the road but just the beginning; there will be many periods ahead that will bring out into the open your algorithm’s flaws. Question is, will they be tiny ones that you can live with, or ones that seriously hurt you when conditions change?

What I am most interested in as a trader, is getting to the market early enough so there is plenty of time left in the day to make money, followed by successive profitable signals that I can capture profits, even if I have a losing trade or two. In the meat of the day, where liquidity and volumes are highest, I want as few “false flags” as possible, knowing that double and triple reversals in price are rare and usually accompany news events. Once I achieve my goals, there isn’t much chance I’m going back in and potentially give it back.

With the changes I have made, unless intraday volatility really falls off a cliff in gold [some days it has] & USDJPY, the “meat of the day” should have the fewest “false positives”; that doesn’t mean there won’t or can’t be any, it simply means there will be much fewer than before. And now that President trump is in office, let the volatility games begin!


Turning to markets today … it’s all Trump folks … whatever he tweets … whatever he decides to do. Here at mid morning markets are a complete mess, with gold & USDJPY putting in rallies & breaks all over the place; gold as illiquid as ever, with not much better conditions in USDJPY, as I have seen panic on both sides of the market within about an hour. Directly below, earlier today, a very good sell signal in USDJPY.
 


Beach beckons … I’m outta here … until tomorrow.

Have a great day everybody!
-vegas
OPEN A DEMO AND/OR LIVE ACCOUNT [CHOOSE THE “ZERO ACCOUNT”; REFERENCE ID# 6952 WHEN OPENING YOUR ACCOUNT] AT THE LMFX LINK IN THE “DOWNLOAD LINKS” SECTION OF THE WEBSITE TITLED “OPEN TRADING ACCOUNT – DO IT NOW!”
 

 


 

No comments:

Post a Comment