MTF HMA + VHands Trade = Holy Grail?

Sorry I couldn't help myself by jumping on the "Holy Grail" bandwagon and the numerous recent threads about discovering the "truth" of the market.

In all seriousness I was testing a MTF HMA indie in VHands trade and every time it changed colors I would either buy/sell. It was incredible, huge wins after huge wins. It was like the Nostradamus of indicators predicting bull/bear runs and even predicting when I should close for a small loss. It certainly wasn't behaving like any other MTF indie I've seen in real-time. Nonetheless, I slapped it on my charts and watched it real-time for a few weeks and yeah it doesn't work the same...BUT I still think it's pretty good as a trend indicator and would like to continue working with it w/o spending days on end making observations to draw any conclusions about the utility of this HMA.

I attached the HMA indie below and found the indie here. It's non-repainting and one of the best HMA indie's I've looked at. Is it simply the way the programmer programmed the indie why it won't work properly in VHands or should I try dl'ing a different simulator like the one here? Is there any other MTF HMA indie that doesn't repaint that I could re-trade in VHands that will behave the same as in real-time?

Essentially all I'm trying to do is load up a MTF HMA indie in VHands (or any other simulator), spend a few days retrading months of data, note the particular idiosyncrasies of the indie, and draw my own conclusions about developing a strategy using it.

Any assistance would be much appreciated! Thanks!!

Well tried the other simulator, which TBH looks to be more user friendly than VHands and took all of 3-4 minutes to set-up. Anyone that's loaded VHands knows that one takes a little longer, but still quite a useful tool IMO. In any case, even with the different simulator, I'm running into the same problem. It's a little ridiculous when the HMA changes in the direction of news spikes prior to the news.

Hmmmmm...I have a feeling that even if I do find another HMA indicator, it's still going to act the same in the simulators. Anyone know why this happens or has a MTF NRP HMA indicator that'll behave as close as possible to real market conditions in a simulator?

i'd be guessing your demo runs are profiting when news chaos ensues which will be demoing on a tight spread, but irl the spread goes to crap and so your real life results also do.

if the bulk of your demo'd profits are when irl the spread sucks then there's the problem.

I'll do my best to offer some comment, but much of it will necessarily be guesswork, given the info that you've provided.

Firstly, I'd need to see the source code (MQ4 file) to ascertain how the calculations in the code are performed.

Secondly, you haven't described the parameters that you used in your testing, or the pairs or timeframes that you tested.

Thirdly, you haven't described how you manage or exit your trades, after an entry has been made.

Anyway, some general comments. The Hull MA operates according to a specific formula** (that should NOT cause re-painting, if coded correctly), so I would ask why (everything else being equal) one version of the Hull MA would outperform another. Also, I doubt that the problem lies with vHands; it's been widely used and tested here, hence a change in simulator shouldn't make a significant difference.

From what I've read elsewhere, and tests that I've conducted myself, it's unlikely that any MA would provide an edge of "Holy Grail" magnitude. Hence I think one likely reason for the difference in the back and forward testing results could simply be a lack of statistical validity. You haven't specified how many trades were made in each case. Several hundred would ideally be needed to get a reasonable idea of likely profitability, drawdown, etc.

If your back and forward testing were performed over different sources of data (e.g. demo vs live), then, as nubcake pointed out, 'br0ker conditions' (spreads, fills, etc) might be very different, as might be the accuracy of the data. Some br0ker's demo accounts are riddled with bad or missing data. The backtest should have given a 'modeling quality' value, but you didn't specify what this was, so once again I'm only guessing.

** Formula for HMA (n=period):
HMA = WMA(2*WMA(PRICE,n/2) – WMA(PRICE,n),SQRT(n))

Here is a non-repainting, email alerted HMA Indy. I used it for a long long time but like all MA's it's stopped working after a while.

Hope it helps you though...I know having email text message alerts to my cell is the only way I could trade.

-Darren

It doesn't repaint at all but if you want to see the actual gap where it changes from a long signal to a short and vice versa then change the "Color bar back Mode" from 1 to 0...

A phenomenal indy based on code guru Mladen

Thanks for the reply guys.

Indeed Mladen has coded up some pretty nifty indie's over at forex-tsd. You guys can go over there and see if they offer any value to your trading.

But Hanover has also coded up some very useful indie's and offered it to the community for free.....thanks Hanover! By the way, you've had some very good posts here at FF over the years, and 'm sure I speak for other traders, we appreciate it!

Anyways, I spent the past week retrading the HMA strat in Vhands and it's no Holy Grail. lol. To answer your questions, MLaden didn't provide the source for that HMA so I couldn't include it. But there are definitely HMA MTF indie's that will change colors, indicating a trend change; however, if you close MT4 and open it again, the colors will switch back to the original trend color if the market just had a temporary deep retracement and continued in the original trend direction. The HMA indie's by Mladen don't do that. If they switch, they will remain that way.

In any case, I can tell you the outline of the strat but it's not profitable over hundreds of trades as you suggested. It has streaks which when I first looked at it, I was "fooled by randomness". My initial testing was not over a large enough sample. There is some discretion in it too. I was looking at EU, 15 min, with a 4 HR MTF HMA 14,0,3 (the settings don't even matter much anyways). Long position opened on color change. Pull a fib from high that triggered the long to the low of the swing. Limit entry @ 20% retracement level. Target 1.618, SL is .786. Stops moved to BE once price closes above the 1.382 extension (vice versa for shorts). Again it doesn't work over a large sample of hundreds of trades.

Lastly, I agree MA's are definitely not a grail, and no grail exists (except arbitrage or some sort of HFT algo), but even then you have high overhead maintaining colo's, hardware (arms race?), etc..., but still those strat's are quite profitable if done correctly. I can't compete on that playing field so don't even try. MA's can show the market trending which I think is an edge retail traders can attempt to exploit. The new strat I'm working on is based off "Trading for a Living" by Alexander Elder and "Trading Day by Day - Winning the Zero Sum Game of Futures" by Chick Goslin where I'm utilizing HMA's as the long-term trend and trying to catch retracements via an oscillator into the major trend. I think the secret sauce (Holy Grail if you will lol) is adding on to winning positions which can only be done trend trading (not covered in Trading for a Living, still making my way through Chick's book for any hints in there).

Hi,


if u want to add this indicator in EA, this way:


Later