Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Is back-testing really necessary to develop a consistently profitable trading strategy?
With that being asked, I went way back in time on the ES and noticed little volume and volatility- compared to now at least. 20-40 point range of movement in a day, if even. Within the past couple of years though, volume and volatility are at an all time high. Is there any merit in going back in time (like 2004-2012), where market conditions were obviously different (climbing bull market for the most part) to determine credibility of a strategy? Has anyone here pondered deeply about the utility of back-testing in general?
If you think back-testing is absolutely necessary, please explain why. Maybe the past actually is indicative of the future to some degree. I just want more information on this.
OK, let's say you do not historically test your approach, whatever it may be. How do you know the method is any good?
Let's say you have an entry based on XXXXX, with a stop loss of Y and a profit target of Z. If you never have evaluated or tested it, how do you know it works (or not)?
With backtesting, you can at least see if that approach worked historically. Does that mean it will work going forward? Not necessarily, but which choice do you prefer:
A. Trade a strategy that has historically worked
B. Trade a strategy that you have no idea if it worked historically or not
The option you choose will tell you if you should backtest or not.
You're making an excellent point here. Things do change; going back years is probably not the answer. How bout staying closer to home, say a month or a quarter. Then it's like a MA, most important data is most recent data. Personally gave up on BT because the systems were difficult to use or gave poor results. If you have one that you're confident in, why not compare him with Forward Testing. Lots of solid demo systems around. Fwd testing for a week or a month seems a lot more practical.
Every year there are Black Friday sales. You can't predict with certainty what will be discounted, but you can make an educated guess based on historical evidence.
The ATR on ES is at a historic high right now, but as a percentage of the daily close we've seen these levels before. So there have been similar market conditions to this one in the past. You may not be able to guarantee the future, but why wouldn't you want to know what worked in the past in similar market conditions, you know what I mean?
If you believe Daniel Kahneman, he says intuition doesn't work with trading - it's an "illusion of skill". He later says “to maximize predictive accuracy, final decisions should be left to formulas, especially in low-validity environments.”
More data is generally better than less.
As you pointed out above the market constantly changes. Volume and volatility won't stay this high forever. Then what?
Two approaches are to identify historical regimes, and backtest only within those regimes. So you will switch strategies on/off when the market changes.
Alternatively develop strategies with data from all regimes, and review performance within each regimes. Something that performs across a range of regimes will give you further confidence the strategy will work going forward.
Read @kevinkdog thread, so you learn to back-test well from the start - and avoid a lot of wasted time and mental roller-coasters.
"We look into a mirror darkly".
History may not be the best way but we can't see the future so it is the best we have.
Also, looking back to go forward may be imperfect and you may stumble at times.
If you try driving a car with the windshield covered and only use the mirrors, you can be somewhat successful.
However, you must go slow and when you see you are going off the road, correct your path.
If you try to go fast, you will hit something or end up in a ditch but a slow, steady movement forward looking in the mirrors and out the side(present) tells you a lot about the road ahead.
The trip will not be without incidents but driving this way does not mean it has to plunge off a cliff.
FWIW
Rejoice in the Thunderstorms of Life . . .
Knowing it's not about Clouds or Wind. . .
But Learning to Dance in the Rain ! ! !