Forex Backtesting

[Br88dy 10]

There are many plans or systems that are sold for $200 and virtually guarantee wealth. Do you think that these plans, if followed with perfect discipline, will really achieve good results?

 

I wasn't planning on buying a strategy, I was going to develop one on my own.

 

Markets trend a lot if you consider intraday activity.

 

Well I want to be an intraday day trader, so is fading a breakout not a good strategy if I want to stick to the 15min timeframe?

[joshdance 10]

Well I want to be an intraday day trader, so is fading a breakout not a good strategy if I want to stick to the 15min timeframe?

 

Well, the market doesn't really care about what chart you use; it's doing its thing regardless of what method of presentation you choose. What matters is whether you can successfully trade from it. I don't use time-based charts much these days, so it's hard for me to really say whether the 15 minute will be suitable for you. You should just look at some different presentation methods and see what speaks to you. Good luck and keep us posted.

 

 

I wasn't planning on buying a strategy, I was going to develop one on my own.

 

Yes, but you said that the reason 95% fail is that they don't stick to their plans. My thought is that part of the 95% do in fact have a plan, and do stick to it solidly, but that does not make them successful because their system/plan/model/whatever stinks. Many "systems" out there are just braindead, and are not based on a solid premise, or have been designed to free the trader from thinking. Having a model of the markets that is logical stands a better chance of allowing one to profit from it, whereas a plan to buy above the high of XYZ every time is bound to fail over time. Just my as always.

Edited February 22, 2012 by joshdance

[afalde 10]

Less Than Random does back testing, optimization, custom indicators, etc.

[4EverMaAT 10]

The article you referenced is precisely a "set and forget" type of approach. You are not supposed to add discretion with this type of approach. No need to watch charts with this. You could always do a hybrid method of a mechanical signal as a guide, and then your discretion to actually trade.

 

 

There are NOT scientific patterns that the markets have followed since the "beginning of time." The market does not offer a scientifically-based approach that performs consistently (such would be relatively risk-free, a no-no for the market) for all eternity. The article you reference develops a system that back tests well for a year and a half, but fast forward 8 years from when it was written and I'll bet it has consistently lost money recently. I suppose you could stop at a max drawdown and have made some good money though. But it is not forever.

 

Interesting point of view. Science is just another belief system anyway, but the fundamentals of buying and selling goods hasn't changed. The buyer and seller agree on a price, and product changes hands. Perhaps a small "rake" or commission for the intermediaries

 

..........

From your first post on this thread it seems you are using metatrader. It does not, to my knowledge, offer any real tools for back testing. MT is sort of the "Yugo" of trading software. It's cheap, and you have charts, but other than that, it's very barebones and not particularly known as a platform that a serious trader will use. Tradestation, Ninjatrader, Investor R/T, and many others will have a pretty robust set of tools for backtesting. Ninjatrader is free to use without live trades, but you will have to get data. You will find a few data providers out there who provide a trial. This may be a good start for you.

 

You'd be surprised how powerful metatrader is. One of the advantages of metatrader 4 is that it has a built-in programming language. It uses a small memory footprint and it encouraged custom indicators/automation from the beginning. The trader that aligns themself with a reputable broker could then focus their efforts on their strategy. This created a huge following that became forex-tsd, forexfactory, etc. as users could showcase their own creativity. You can always request API from a broker that offers it and pay a programmer to bridge it to your favorite trading package. Most popular brokers will have already set something up, like MBTrading >> API >> Multicharts. But you could also use their own proprietary platform, or Metatrader 4.

 

In futures, just about every broker went the proprietary route, but API access was usually standard (just had to ask for it) because everyone gets the same feed and trades the same prices. So "bring your own charting package" was an unwritten de-facto rule for many of the discount brokerages. An individual firm having their own charting package was a pretty big selling point, especially if it had robust programming capability built in. Tradestation actually could have standardized the industry with their software about 10-12 years ago if they licensed it out wholesale to brokerages, but they chose to keep it in house, forcing users to maintain a tradestation account to get access to it. Multicharts came late, but I think they did well to make Tradestation-like features available to the masses without the Tradestation bureaucracy. Ninja trader, strategyrunner, etc and others are similar also, but one has to pay for these IN ADDITION to normal trading costs.

 

Another advantage of MT4 is that it is a de-facto standard among retail forex brokers. This is great for a trader that has their own Expert Advisor (EA)/indicator; they just copy and paste their files to a new broker's mt4 installation and they are set to go. A broker can "easily" create their own custom indicators if they want to "brand" their version of mt4 beyond a simple logo swap. IBFX has done just this. They have a lot of custom indicators that are availble ONLY through their platform. So it is easy to switch brokers, and easier to compare strategies among a common community. To keep up with 4-5 different platforms for a futures broker; it is a nightmare to port your indicators unless you are an expert programmer. For a professional trader it's not the end of the world, but it does take considerable time to test several platforms, especially if you want custom programming. Good luck if you decide to make a platform "switch".

 

If a forex broker comes out with its own proprietary platform ONLY (not offering mt4 along side), it is much more difficult for that broker to break ground, unless that platform blows the competition away. A proprietary platform with programming capability is an expensive startup cost, and time consuming to bring to market. Plus you must convince your customer to only use your brokerage (or increase business at your brokerage)?

 

That standard is also making it difficult to switch to MT5, as the languages are not compatible. Mt4 works a little too well, and even though the changes in mt5 are significant, it will probably take 2-3 years for the masses to catch on. mt4 will be arould for many years.

[4EverMaAT 10]

For the OP, ForexTester is what you are looking for. It's not perfect, but the best thing going right now for walk-forward market simulation. You'll need their VIP service to download tick data, which is appx 2 years with several brokers. (the guy collects the ticks himself with is own capuring tool). For the tick data, it takes up about 1.2 GB per pair on IBFX data for the past 2 years.

 

The program allows you to play data tick-by-tick if you wanted to, and you can do this over the weekend or whenever you want. If you have an automated strategy or custom indicators, you may also program them and import them into the program.

[joshdance 10]

You'd be surprised how powerful metatrader is.

 

I really wouldn't be. I used it for several years, and have written more than my fair share of MQL.

 

Your argument in favor of MT4 seems to be largely that they have succeed as a business in getting their product into the hands of a very large percentage of retail forex traders. I agree completely with that, and see all of its benefits in this regard.

 

Also, I always did appreciate that MT4 is lightweight and fast, uses little memory and the little that it does, it does do well.

 

However, from an actual trading standpoint, it offers very little other than being free, and providing access to free data. Probably the most pronounced of its deficiencies is the inability to do two very important things (to me anyway):

 

+ Define trading sessions:

This should be basic to any charting software, IMHO. With MT4, your bars start and end according to your broker's time zone. I have seen many users downloading 4 different broker's versions of MT4, so that they can see a 4 hour bar which starts at 12:00, 1:00, 2:00, and 3:00. If my broker's daily bar starts at midnight EST, too bad; never mind the fact that no trading session in the world starts and ends at midnight EST. This also means that you cannot hide certain periods of trading. If I trade from 8am to 5pm BST and wish to hide post lunch NY trading, as well as pre-London Asian markets, I can't, as far as I remember anyway.

 

+ Define custom bar periodicities

Never mind the fact that non-time-based periodicities are not possible (range, volume, tick, P&F, etc.) -- I can't even create minute-based multiples (2 minute, 3 minute, 90 minute, etc.) with MT4. It's possible that with MQL some users have implemented ways to do this, but I really don't want to trust some 3rd party code to perform what should be a basic functionality.

 

So it may be free, and have free data, but you really do get what you pay for in this case.

[4EverMaAT 10]

I really wouldn't be. I used it for several years, and have written more than my fair share of MQL.

............

So it may be free, and have free data, but you really do get what you pay for in this case.

 

It's extremely difficult for any charting package to be all things to all people. MT5 addresses the "+ Define custom bar periodicities" , but I doubt they added the functionality to hide trading periods. You could create an indicator to highlight/hide the bars that you want to ignore, but it would still leave gaps on your chart. For these types of requests that metaquotes won't honor directly, you would need to either request API access or bridge the incoming tick data with a custom indicator and bring-your-own-charting package to meet your demands.

 

But having a robust ability to program workarounds was a refreshing change for something that is free for the end user. It was ahead if its time and even now it is progressive. After managing a few programming projects myself, I respect the fact that it takes a lot to make a stable trading platform for the masses. There is no open source charting packages in this business (that I know of). Who would pour thousands of free hours into creating a package like sierra chart, metatrader, and then offer tech support it also? A dealbreaker is a dealbreaker, but I think that a serious trader would do their best to workaround cosmetic limitations.