Duxon's Archive

[studentofthemarkets]

only increased my bias against looking to a hypothetical past for proof of validity and reliability as opposed to real-time trials in the actual present—at least when it comes to using the MetaTrader4 program.

This is a bit concerning. I'm not sure what you are meaning by a hypothetical past for proof of validity. I'm guessing you mean that the backtest would not give the same results as if you had traded the same system live at the same time. For example, suppose you tested your system live on Tuesday of last week, and then decided to run a backtest on it this weekend. The results should be identical. If they aren't then it calls for finding out why. Could the results be affected by less accurate data because it is only using the OHLC of a candle instead of intra-candle tick data? Are the indicators calculating differently in a backtest than in real time...such as waiting until the close of a candle to evaluate, whereas in realtime the indictors update continually? I too have been very hesitant to trust the accuracy of my own backtests for these reasons, and there are probably more others have found.

I'm not sure if you can go back and look at how the backtesting entries were placed using metatrader, but with Ninjatrader you can make sure the trades were taken right because the back test lets you see the candles, indicators, entries and exits. This is really helpful to gain confidence that your backtest took trades in the right places.

I hope learning to code goes quickly for you so that you can do automated trading soon. Seems like it could be easier to get statistical data on your real time trading with a live automated system than with manual trading. At least you will know what the stats are without a human element involved in the decision making....well, I guess you could run your automated trading system for alerts but only take the trades you decided to take and that would defeat the purpose of finding out how your system performs on it's own, but it might make more $$ with a human element added.

 

[expiated]

This is a bit concerning. I'm not sure what you are meaning by a hypothetical past for proof of validity.

It concerned me too, so while you were typing this response, I edited what I wrote to "automated" past, which I changed before I even saw your comment. I evaluate the past all the time, but not a hypothetical one--rather, a historical one. I'll go back and read the rest of your post now.

 

[ph1l]

For example, suppose you tested your system live on Tuesday of last week, and then decided to run a backtest on it this weekend. The results should be identical. If they aren't then it calls for finding out why.

A backtest is almost never identical to actual trading because you have to make assumptions on whether orders were filled and what prices they were filled at.

Metatrader 4 backtests have a poor reputation.
https://www.forexfactory.com/thread/post/11114843#post11114843

Standard MT4 backtest quality is crap and cannot be taken as source for any assumptions.

 

[studentofthemarkets]

Most definitely. I don't do back testing. I analyze using historical data and I test in real time. Here is an example of a baseline that looks to give a consistent edge when evaluated looking backwards, and which actually does give a consistent edge in real time:

View attachment 265380

The last chart you posted, showing the buy and sell areas looks really good. Price is generally keeping within the boundaries of the envelopes and the direction for trading is usually clear.

 

[expiated]

Monday / August 9, 2021 / 7:45 AM PST​

Having reached what more-or-less amounts to the end of this leg of your journey, in the final analysis, for you personally and under normal circumstances, swing trading is pretty much out of the question. It just doesn't make a lot of sense from where you're standing.

The seventeen-minute measure is the ultimate backbone of your intraday system (kind of) or to be more precise, the 17-minute price range envelope at 0.04% deviation. If candlesticks break out of this zone, the corresponding pair is trending strongly. On the other hand, rates will almost never violate the 0.20% deviation level of this measure in the short run.

When rates are not trending so strongly, the general intraday direction of price (and intraday reversals) is revealed by the positional relationship between the 23-minute and 45-minute (a compromise between 40 and 50 minutes) baselines, with the faster measure obviously leading the way.

To decide when to enter positions and when to take profit, consult the five-minute baseline (is it reversing direction?) and the relative position of candlestick formation within the seventeen-minute price range. Also check to see whether the guidance provided by the five-minute measure is backed up by the 10- and possibly even 15-minute trend lines.

For an idea of how much room price might still have left to run, you have the following with which to gauge potential distances:

• The two-hour price range envelope at 0.11% and 0.20% deviation
• The four-hour price range envelope at 0.30% and 0.50% deviation
• The 16-hour price range envelope at 0.14%, 0.30%, 0.45% and 0.60% deviation

These three measures can also give you a sense of the general flow of price from successively broader vantage points.

And of course, to see where price is headed from week to week and month to month, consult the four-day price range envelope at 1.50% deviation and the 12-day price range envelope at 2.85% and 5.00% deviation.

 

[expiated]

These three measures can also give you a sense of the general flow of price from successively broader vantage points.

Today was not an easy day to trade, but was a great day to adjust anything that needed additional fine tuning in this near-final version of my system. I've drilled down to the point where I am totally immersed in the nitty-gritty—the most important aspects or practical details—of this methodology... information so consequential that I think it begs to remain "classified."

I will therefore be interested to see if I have anything more to add to what I've already typed in the past, only parts of which I consider to still be valid. But at this point at least, I have nothing more to say about NPP.

 

[expiated]

If you find anything interesting, like if you find that futures are harder to make money with than binary options, I hope you post something about it so the rest of us don't have to put in the effort to find out which is easier to make money with.

Late Sunday night saw the last little piece of my system fall into place, which involved understanding how my two-hour and 30-minute measures work together, so that I now know what to expect foreign currency pairs to do at all times (and how to interpret what they are doing when I see that this is changing.)

I just now took these same measures to the major U.S. indices, and found that though they are just as valid in this context, the indices are simply too "wild" for me to feel any motivation to trade them.

Forex is nice and orderly, whereas the futures tend to swing wildly and quickly. Also, the amount of territory covered by the typical index candlestick spans much greater distances than the typical Forex candlestick.

An example of the difference is that when trading Forex, I almost always know when there is the possibility that the trend might reverse direction soon, and of course, I then recognize when it does. But with the indices, reversals often occur instantaneously, so that while I might recognize when it happens, because I did not see the possibility of it coming ahead of time, I might have gotten stopped out of a trade that I would not have entered if the market had cautioned me that a reversal might be coming.

And on top of this, futures might reverse in one direction as describe in the above manner, and then turn around and reverse in the other direction shortly thereafter, relatively speaking. Foreign currency pairs do not behave in this manner. If a currency pair adopts a new trajectory, it will typically maintain the new course for a decent amount of time before turning around and going the other way again—not so with the Dow, Russel, S&P or Nasdaq.

The image below illustrates at least one of my points. The red moving average is one of my key baselines. But it is "unstable" on an index chart due to the behavior of price action in the futures markets. I have used arrows and ovals to highlight spots where the line goes up, and then down, and then back up again in a relatively short period of time. This simply does not happen in the Forex market.

The fact that it DOES happen in the index markets makes DAY TRADING indices a very unattractive prospect to me personally.

 

[studentofthemarkets]

And on top of this, futures might reverse in one direction as describe in the above manner, and then turn around and reverse in the other direction shortly thereafter, relatively speaking.

LOL. Yep. It can be a crazy roller coaster ride. Up, down, up, down, and sometimes no clues as to what it will do next. Here's a pic from the NQ 1 min chart that I would find difficult to trade. There are clues on the chart....if you look at the how it responds around support and resistance levels and it's general rhythm, but they are much easier to see in hindsight than real time.

Thanks for sharing your observations comparing the indexes to currencies!!!

 

[expiated]

Friday / August 20, 2021 / 9:45 AM PST

I've drilled down to the point where I am totally immersed in the nitty-gritty—the most important aspects or practical details—of this methodology... information so consequential that I think it begs to remain "classified." I will therefore be interested to see if I have anything more to add to what I've already typed in the past, only parts of which I consider to still be valid. But at this point at least, I have nothing more to say about NPP.

Well... what I typed in the past was valid, but I had to break it all down and then put it back together again. It was only this morning that I was able to really "see" how it all fits together (properly this time, I hope).

It took a week to put this "icing on top"—from last Friday to this Friday—but it looks finished in that I have the same measurements I had before, but am simply viewing them from a different perspective...

It took the first four trades during this last 24-hour market cycle of the week to arrive at my epiphany. The last nine trades were made after this (final?) insight was put into application. Going forward, the winning daily percentage should jump from 69% to anywhere between 85% to 100%.

My hope is that next week, I will continue to repeat the kind of performance achieved with these last nine trades without any addition modifications to my charts whatsoever (for the most part).

 

[expiated]

J.M. Hurst

The manner in which I executed my last nine trades this week used the components highlighted below as suggested by at least two views I share with J.M. Hurst:

1. The Principle of Cyclicality: Price movements consist of a combination of specific waves and therefore exhibit cyclic characteristics.
2. The Principle of Synchronicity: Waves in price movement are phased so as to cause simultaneous crests and troughs.

Presently my charts include the two-minute price range envelope at 0.03% and 0.05% deviation along with the one-minute baseline (see image below). These measures of short-term price fluctuations play only a limited role in any decision-making process by partnering with the five-minute price range envelope (at 0.05% deviation) when it fulfills its primary function as the announcer of reversals in the short-term intraday trend.

The ideal time to enter a position is when the above-mentioned reversals are observed to occur at statistical support or resistance (depending on whether the general overall intraday trend is rising or falling, as conveyed by the 45-minute baseline) in the form of the 45-minute price range envelope at 0.12% deviation.

Again, the five-minute price range envelope (pictured below) tracks the general flow of the short-term intraday trend.

The 17-minute baseline conveys the direction of the intermediate intraday trend. Its associated price range envelopes at 0.05% and 0.12% deviation work in concert with the 45-minute price range envelope at 0.12% deviation to identify potential entry levels—not to mention reasonable stop loss values and realistic/rational take profit targets. On the other hand, this same measure at 0.20% deviation tends to represent the maximum distance rates are typically willing to separate themselves from the underlying baseline before reversing toward the mean...

As already suggested, the 45-minute price range envelope at 0.12% deviation reveals the gist of the trajectory of the general overall intraday trend. The band opposite the slope of the baseline serves as a reasonable stop loss. The band at 0.32% deviation tends to define the maximum distance rates are typically willing to separate themselves from the underlying baseline before regressing toward the mean...

As previously stated, the key to successful intraday trading is entering and exiting positions when these four (or five) measures are bottoming out or maxing out all at the same time.