KeltnerFlow Omówić

Good points, btanalysis.

Frank, you have posted a 15-year EU backtest. What is it exactly that prevents you from also posting the other pairs' 15-year backtests?

If you need them made in high quality, I can take my time to do them for you on tick data via the Tickstory suite. All pairs, for the maximal possible period each (since 01.01.2008 until now). I will just need a personalized copy of your EA for it which wouldn't be tied to specific trading accounts and which I have requested from you by email.

It's up to you whether to accept my offer - I just somehow thought you were interested in creating a true one-off with this bot. The better you make it, the more sales you'll make.

RedRhinoLab posted:
I'll gladly send you a Backtest Only version later this week so you can run your tick data backtests with the volume filters.

That should be fun.

RedRhinoLab posted:
Does anyone know why KeltnerPro only trades these selected pairs? Anyone try other pairs?

Because other pairs have higher spreads and the central banks' policies behind the other pairs' currencies are not divergent enough for frequent big price movements.

I have added new backtests to my profile after completing KeltnerFlow.

Version 1.0 will include a variety of settings which should allow us to trade almost any currency pair. I would like to encourage the use of a backtest ONLY version by providing a 10% discount for each Successful Currency pair that you optimize and backtest.

The backtest version PDF manual will explain how to optimize the EA and the pairs selected are up to you however there are a few limitations to earn the discount.

1.) Data type requirement, TickData (dukascopy) OR https://www.forextester.com/data/datasourcesb -

This will allow all members the ability to reproduce your backtest results. The choice is yours. Some already have tickdata and for those that do not, the data provided in the link above are much better than metaquotes data. If TickData is available, you can test the volume filter, if not leave the volume filter == 0.

2.) You must upload the backtest to the myfxbook strategies section and make public. This allows us to verify the quality and it provides better analytics. Please name the strategy backtest ' KeltnerFlow' - 'Years of Data' - ' Currency Pair'

Example: https://www.myfxbook.com/strategies/keltnerflow-3-year-gbpusd/80511

3.) If I've already posted backtests for a specific currency pair, in order to qualify for the 10% discount you must have a higher Profit Factor and higher number of total trades for the same data period.

Thanks for being apart of the backtesting beta group. Software will be available shortly.

regards,

Frank
https://redrhinofx.com/keltnerflow/

RedRhinoLab posted:
Software will be available shortly.

Please, let us know once that is done. I feel like I'm increasingly missing the point here - after seeing those 3 new 3-year backtests with profits much smaller than those of the original KeltnerPro bot. Especially, +30.25% on EU over three years. 😲

I want to run on many currency pairs with the highest probability of making a profitable trade. If patience is needed by getting less signals for better results than that is perfectly fine with me. I am interested in a better profit factor than KeltnerPro's PF1.23 on EURUSD - https://redrhinofx.com/keltnerpro-review/

More trades with a smaller PF can result in more profits overall but a better system comes with higher ProfitFactor and less trades.

RedRhinoLab posted:
I want to run on many currency pairs with the highest probability of making a profitable trade. If patience is needed by getting less signals for better results than that is perfectly fine with me. I am interested in a better profit factor than KeltnerPro's PF1.23 on EURUSD - https://redrhinofx.com/keltnerpro-review/

More trades with a smaller PF can result in more profits overall but a better system comes with higher ProfitFactor and less trades.

That would hold true if the higher profit factor led to higher fixed-lot profits, too. That's something I fail to see from your backtests - rather the opposite takes place, like those meager 30% over 3 years on EU.

Breakouts are some thing that can happen at any moment or don't happen for over a year. The markets are normally efficient and most say range more than 70% of the time.

To try to improve a breakout strategy only leads to LESS trading opportunities and I don't think you realize that. Improving the keltnerPro EURUSD PF of 1.23 to a higher number will result in LESS trades and not more.

AUDUSD had some nice break outs with 150% gains - https://www.myfxbook.com/strategies/keltnerflow-3-year-audusd/80510
GBPUSD had some nice break outs with 150% gains - https://www.myfxbook.com/strategies/keltnerflow-3-year-gbpusd/80511

EURUSD and GBPJPY gains are less, but also drawdowns are under 10% for the 3 years which indicates we can crank up the lotsizes if we wish.

These backtests are NOT the only setting as well, I didn't optimize all the parameters to show the holy grail, I want my forward test to mimic the backtest results and in order to accomplish this , the optimization should be selective.

Frank is risking only 1% at fixed lot, that's why the profits look so small. I agree with him that it's better to trade more pairs with fewer trades yet a higher profit factor per each pair. I will gladly run the comparative tick data backtests of his bot and the original KeltnerPro EA with the same risk settings for all of you to see his bot's advantage over KeltnerPro.

I have my doubts about the relevance of aiming to increase the PF at the expense of the number of trades when it comes to this type of strategy. Markets don't treat the same things the same way over time. Just some years ago we would have wild rides on such Greece news while now the market hardly moves. So it's not unusual now when an incidental breakout does end up with trades hitting TPs due to something like a thin market while a perfectly healthy, logical breakout from the standpoint of the past soon reverses because the market players now realize in increasing numbers that both the US and Europe are going to the dogs, quantitative easing doesn't help much, and liquidity deficit was merely an outcome of the previous 2008 financial crisis mistaken for its cause.

There is some truth in your words, btanalysis. I've noticed this bot enters the market following a 45-pip (or more) bar. For the purpose of stressing your point, let's now say we've increased this minimal bar size prerequisite to a whole of 80 pips. Then the bot will be entering the market far fewer times per year and will in all likelihood demonstrate a higher PF/accuracy rate on backtests. Yet the modern choppy markets leave us with a far higher chance of even such a big breakout reversing instead of forming a new trend. Thus, ironically, we would end up with less profits due to fewer entries despite the higher PF on backtests.

Better yet - pick the biggest single breakout over the backtested period. Set the bar limit to the size of that breakout. PF on backtest? Intense! Accuracy? Extreme! Chances of a similar outcome on a forward test? Next to none.

@forexstriker - there are a total of 16 preset logics within keltnerFlow, some do use the candle size and some do not. The 45 pip candle size bar in the backtests may or not be used depending on the logic version.

I've added some more backtests that have a lower profit factor and can have up to 900+ trades over the 3 year period. For example
https://www.myfxbook.com/strategies/keltnerflow-3-year-eurusd-pf172/80577

Looks like you're using a lower time frame (M5 instead of M15) there. Anyway, I'm still waiting for a backtest version of your bot - will then run its comparative backtests with the original KeltnerPro EA and share their results in this discussion.

Today, the original KeltnerPro EA took losses because of a late trade opening due to slippage. KeltnerFlow has demo-profited. Such a feast!

Setfiles can be downloaded here - https://redrhinofx.com/keltnerflow-version-1-0

KeltnerFlow Backtesting version can be downloaded here - https://redrhinofx.com/product/keltnerflow-bt/

You must enable ' Import DLL Allowed' for backtesting to work. Tools --> Options--> Expert Advisors --> Import DLL checkbox

More currency pairs and more setfiles will be provided in the future.

Suggestions are highly appreciated and features will be added as requested.

EA is not locked to specific timeframe, but all my backtests are on M5 TF only. It is universal EA to trade on other timeframes and ANY Currency pair.

Enjoy the weekend.

I can't copy the EA file from the ZIP archive without entering a password. What's the password?

RedRhinoLab posted:
Setfiles can be downloaded here - https://redrhinofx.com/keltnerflow-version-1-0

KeltnerFlow Backtesting version can be downloaded here - https://redrhinofx.com/product/keltnerflow-bt/

You must enable ' Import DLL Allowed' for backtesting to work. Tools --> Options--> Expert Advisors --> Import DLL checkbox

More currency pairs and more setfiles will be provided in the future.

Suggestions are highly appreciated and features will be added as requested.

EA is not locked to specific timeframe, but all my backtests are on M5 TF only. It is universal EA to trade on other timeframes and ANY Currency pair.

Enjoy the weekend.

I am prompted to enter a password while copying the .ex4 file to the 'Experts' folder. What is it?

follow instructions please. If you do, you will get the password via email in auto responder.

https://redrhinofx.com/product/keltnerflow-bt/
Must include Correct Email address
Password to unzip folder is sent to your email address. You must scroll down to the bottom of this page ( Footer) and insert your email address inside the Sign Up Box. Without doing so, you will not receive the password to unzip the folder automatically.

I have not got password either - like anyone else - while following the original instruction... I guess all of us are simply dumb...