optimal numbers for stop loss and profit target?
- Thread starter Wach80
- Start date
Another reason of learning to program for a year, then come back to trading. All these questions you ask the forum, most of the answers comes from you back testing. Like I can dream up of a system in five minutes cause I know what works and reason I know is tens of thousands of backtests. Get like Ninjatrader or Multicharts and start learning, ask around support groups on mentor on teaching you how to program, spend a solid year on learning as you then have knowledge how to do it well enough to gain knowledge on how to trade.
Learning how to trade well is not cheap any more, although in the 80s wasn't cheap either as PCs be 2-3k, data far more expensive than now. But you are trading against the brightest people on Earth who know charting like back of their hand and you trading against programs who were designed by same people. So you have to back test well to know the answers before the questions.
I dont think it will take a year to learn program into ninjatrader but yes trade well will take a lot of time , i have learned alot but there is still alot of terminology and concepts i dont understand and the bad thing is that there arent specific rules about anything , it is interesting if you ask me because it involves strategy but it doesnt involve theoretic perfect rules or an ultimate strategy or even an approximation of it , its like you said , you can determine something after a ton of tests only , i have different understanding of strategy thats why i make all these questions. In a multiway game of pure psychology no math can solve anything at all , so just test a ton to figure out that psychology...its a pure reaction game.
thousands of backtests resolves psychological issues in trading .when someone speaks about mental pressure of trading ,they actually speak about lack of enough backtest .Dr. Van Tharp underestimates number of backtesting & he claims 100 of samples would be enough .that's a joke
however after 200 samples you know almost how good trades goes against stop loss and vice versa .but further backtesting like thousands makes self confidence in trading system as well as optimal stop loss and profit targets
Jason Rogers
ET Sponsor
But unless you know the game well, every time there is a reaction will be hundreds of trades a day and I wish I was your broker. Not all reactions are the same. And there are many rules in trading and the fun part is you have to develop your own set of rules.
As far as sample size, last time I thoroughly back tested an ES system over 12 years of day session tick data had little over 160,000 sample size covering 8 signals. The only other method that I know more about is my Long Term Commodity Method tested over 80 years in some markets...some of the data was spotty cause of World War 2. When you get sample sizes this large, then plugged in your risk management, there isn't much you don't know or when someone says this has never happened before, it has to a degree of different times with exceptions of New Highs.
I laugh when I see someone say they back tested something less than 8 years, much can happen in 8 years. Obama been President for eight years, the economy is in worst shape but Mark Brown is selling guns and making big bucks, I most likely going to start buying assortment when I buy my farm in East Texas, they have those wild hogs out east Texas, mean critters with tusks.
thousands of backtests resolves psychological issues in trading .when someone speaks about mental pressure of trading ,they actually speak about lack of enough backtest .Dr. Van Tharp underestimates number of backtesting & he claims 100 of samples would be enough .that's a joke
Actually, having read most all of his books (some several time since) I am pretty certain Van Tharpe recommends 400 samples for sufficient knowledge of a system and he recommends not to trade with less than 200 samples but you can get some idea if a system is worth pursuing after only 100 trade samples, however he would not trade it at that point. He also recommends a Monte Carlo simulation of those 200 or more trades run 10,000 times.
in the book "trade your way to financial freedom" he states minimum of 100 samples which is not sufficient at all .maybe at least a thousands in different years may work.he insists on trading psychology which means he encounters lack of enough backtesting .also he underestimates how long does it take to find an edge .his writing are great (80% of them ) but also in some parts he has some misleading.i did not read other books by him and my judgement is only by above book.
I am not a fan of Van Tharpe or Roosevelt or Douglas, that not to say they not helped others. I only believe in what I can see in the past, then forward test it, paper trade it then do a one lot for few months how I start, am very conservative. Many have PM'ed me in past to say hi and how risky I trade, but what is risk to one is very little risk to another. I often think how others view me in a way they would never trade as I seldom take the "feel good" trades of going with trend, but what many don't understand about "Futures", Commercials use futures as a Hedge to cash. And options on futures is like a double hedge to cash. I have seen plenty of Cash charts of the Futures underlying and I be more of a trend trader of those charts. Day trading Indexes, it is more of a dance back and forth around the "mean", don't think New York wants to be too far away from the Mean, and why reactions are all different.
Reactions can be a simple "BTFD" as retracement in a trend, and some are intra-bar, I know some trade "Fibs", I have a rule of pivot low to current high has to be so much range, so if current bar opens and price goes higher to exceed my "range min" then comes back before close of bar, I will buy at 61.8% of this baby reaction, and should price be pushed back up as all the fib traders are pushing it higher, the close can be at the highs of the bar. So this is exactly best to back test on tick data, cause using one minute bars would never show this trade.
But if you don't have the knowledge which is gained by studying charts, you won't understand why it is important to find the "mean" Swing average length, on ave, like the ES will go in one direction so many points before reversing, the very best you can gather is 50%, but your risk management rules can keep you out of trouble by using time stops, don't want to get married to it, just ask Paul McCartney. Back testing will show if you ask it how long is too long, I have found best trades take off in my desired direction quickly.