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)
I will use this thread to share some of my thoughts and write about things that interest me. I will not be posting any trades or daily market opinion or commentary. Just random stuff that I feel like writing about. Not sure how much I will post, but thought it would be nice for not so busy days.
I do have a first post I would like to share and will do so in a few.
Also, I will not be answering any questions. If someone else with knowledge would like to answer, please do. If you don't know, don't post. I specifically choose a journal hoping it does not get spammed
Thought I would share the following.
It is from my private trade journal, although I have removed a lot of the detail to condense it and I have also made it more reader friendly. Maybe it will help give a foundation for new Ideas to a struggling trader or possibly even a successful trader with a similar edge/ideal.
I will start by saying…this is merely my attempt to show peeps some type of proof that tradable edges do exist in the market. This is in no way the only or best way I have traded the markets! But many edges can be built on this very concept.
If it were the only thing I had, would I trade it? Absolutely. I did for over a year.
I have chosen this method as my example, because its easy and can be shown without giving away too many details. This was one of the very first methods I derived nearly 3 years ago. No one showed me this method, I derived it out of pure thought from many things I have learned from intelligent people. In no way do I take claim (Only for my own ego), I am confident it was well known before I arrived! LOL
I am going to use the concept of Volatility reverting and understanding that the Mean is only relative to current conditions and is nothing more than Volatility Neutral. I will not give the volatility calculations away…find them on your own. What is more important here is the method itself.
Although I disagree with him on things he said in the interview, I believe this is in some way what Tom Sosnoff, in chat with traders, was referring to (Expansion and Contraction of Volatility). Of course…My approach to this method is probably entirely different, so keep that in mind. Options traders might respect what I am about to show with this method.
See this thread for the interview.
This thread starts from a discussion that started in "the small exchange" thread, and in particular in this post
In that post I mentioned some points that Tom Sosnoff made during the interview with chat with traders
1) there is no way to …
The method will involve one upper band and one lower band. They will be my measure of volatility. If the market is to hit one of these bands, an event has occurred. You can trade the event, or you can wait. I will try to explain this the best I can later.
***As we know, not all events in volatility will return to a mean and nor do they return in some set amount of time. News event traders pay attention.
Remember “Past performance is no guarantee of future results”. Thank you, SEC. But really…How often do you drive home from work a different way. During an “event”, right?
So…an event (when a band is hit) is a moment in time where volatility is out of line with the markets current posture. You can think of it as a sudden drop or spike in momentum that is relative to the current markets condition. The Distance between the Upper and Lower band will determine the current posture of the market.
The key point here is… I am using volatility that is currently relative. No different than the concept of options traders using Volatility Rank/Percentage to base current volatility.
Although my example will not include it…Look into volatility skew. 😊
*These volatility bands have no parameters that are user adjustable. So, nothing in the way of optimizing…we will use them the same way to calculate volatility no matter when it is. Also, the bands are in no way calculated with any directional bias (Upper and Lower Bands are calculated the same, just Opposite).
Bollinger Bands and other box ready indicators are all lagging, you need to think in terms of predicting the future.
One other thing that must be explained is the event sequence. If an event happens, the revert will not be complete until an event happens in the opposite direction. See Chart below.
Revert in Volatility (Re – Latin for back, Vert – Latin for turn)…So Turn Back
For all you advanced guys (I know you are thinking, so I will explain). For this example, I will not include concurrent events in the same direction. These concurrent events will allow for possible scale/average in methods to strengthen a position. For newbies, please be careful with this. See chart below for concurrent events.
Ok, a band is hit, what now?
We will draw a series of lines starting at the band and going lower or higher depending on which band was hit. These lines could be almost infinite (at least to the tick) but let us be realistic in the method.
I call them offset lines. They are always calculated in the same way for each event, so you will not have to optimize. The calculations for my example will not be given away, so figure it out on your own. You need to learn your own way. Remember the concept is what’s important.
Some traders may think that no matter where you place your offset lines, they will be random. Believe what you want, but for me it is all Math. If you have a mathematical reason, it is no longer random. Without math, you then can argue Random Line Theory.
See Big Mike’s interesting thread on Random Line theory here.
I want to test something. It's simple. Place some random lines on your chart prior to the day opening, and see if at the end of the day you feel like those lines were important (try to imagine they weren't random, but some expensive or complicated …
See next chart for an Event Short with offset lines.
Reference your offset lines by number. I start with 0 where the Event happens. Each offset line will be a trade opportunity. You can trade line 0 at the event or wait to see how far the market will continue in the current event direction.
No matter what offset line becomes your trigger, the trade is buy on a short event and sell on a long event. This is a reversal strategy by reversing Highs and Lows.
Offset line 0 obviously has a 100 Percent chance of happening, but if you wait for price to reach other offset lines, the chance of a trade triggering will grow smaller as you get further from the event.
EXAMPLE RESULTS:
Considering a higher timeframe…I have used my method to build a NQ swing strategy and will list backtest results for all offset lines below. The results assume every trade is taken and no slippage has accrued. Slippage for this timeframe would be minimal to nonexistent. Commission have been included. I have also built a specific rollover schedule that the strategy adheres to.
The following results are for all trades from 1/1/2008 to 6/2/2021 (13.5 years). Looking at offset 0, which will keep you in the market continuously when not considering rollover, the strategy takes 387 trades.
On Average:
28.67 trades/year
7.16 trades/Contract Month
.11 trades/Day
Some things to consider when looking at the results:
There is no stoploss used in the example, so MAE should be considered.
Using a StopLoss will improve MAE and Drawdowns at the cost of Profit. There are ways to include logical stops but that would be a good topic for a whole different post.
Using Options or spreads, a trader could easily hedge this trading strategy and eliminate the need for a stop.
Volatility skew was not considered in this example and would improve results. This also should be a topic for a whole different post.
Notice every offset (that triggers trades *0-17) is profitable. In fact, this is the Holly Grail! LOL.
Special thoughts:
Offset line 0 is your control and will show all volatility events within your chosen timeframe. The shorter the time frame and the smaller the volatility event needed.
With the chosen timeframe for my example…I used one that will only pick up larger events. Any major event over the 13.5 years will show in the trades log.
Twilight Zone:
One event that I would like to point out is Covid. Most news reports list the Covid selloff starting on February 20, 2020.
I am Not going to argue but…
Using my timeframe of reference, my strategy shows a short event happening on January 26, 2020 and does not trigger a long event until middle of April (Start of the “V Shaped Recovery”).
It just so happens that January 26, 2020 was only 2 days after a classified Senate briefing on the threat of a coronavirus outbreak.
According to financial disclosure forms, many of the Senators and high officials in that meeting sold hundreds of thousands of dollars in stock within a couple of days.
Hmm…wonder how many of their family members and close buddies did also?
The only problem with my observation is that the NQ market slowly rose nearly 500 points between January 28 and the selloff of February 20.
I have a hunch that big money was only shining the apple during this time period and waiting for the big bite.
Or you can think they were bracing for losses. I dannae is she can take any more, Captain. Aye, the haggis is in the fire now for sure!
But please, do not attempt to answer this in my journal thread. Start a new thread…Thanks.
Anyways:
I traded this method in a few markets for over a year and did really well. The only differences between my trading and the example, was I used a volatility skew calculation and a maintenance margin stop. Together, both of these will greatly improve the method. I have not traded it in some time now. I have found many strategies that I do prefer and perform better. Most of my current strategies are breakouts, although I do trade something similar to this with one market.
Thought I would add the results with Skew included.
Although this needs it own post to discuss, Skew works best right after the event happens and gets less significant to nonexistent as time decays after the event. Easy to see if you compare the results using and not using skew.
Thought I would show a method I use in walk forward optimization.
For me…WFO is single greatest tool that a systematic trader has in their toolbox.
Its just a fancy way of finding the optimal parameters for a trading strategy.
To learn about some basics of it, you should really check out this series of videos. Hands down the best I found…
For clarity, other than these videos, I know nothing more about the author (Darwinex).
Here is a link to the 5 video Series....
My method:
Looking at this picture as an example:
Let us assume for the 6 optimization periods you are testing 2 parameters. Normally most would use the best performing combination of the 2 parameters for each period. I call this the POC (Point of Control) of your parameters.
So… what happens if you do not use the POC but use what I call the ROC (Rate of Change). The combination of your 2 parameters that creates the greatest change thru each of the optimization steps. Optimization 1 has nothing to compare to, so you will use the POC on that optimization.
To demonstrate how this works, I will show the following example for the NQ 03-21 Contract. This is a day trading strategy and is walked forward daily. There is only one parameter to optimize, and it has 50 values. The data set is large, so I will not be able to post in one pic. Sorry. I will show the first and last page of the excel file. Green demonstrates POC and Yellow demonstrates ROC. The light pink at the bottom is just the average return for all 50 values of my parameter.
Looking at the results:
Although all 50 values are profitable for the strategy. Picking the best value of 49 and not walking forward with optimization would have gave me a profit of $44,027.40 for the Contract Month. You have a 1 in 50 chance of picking this value. Picking the worst value of 33 and not walking forward would have made a profit of $15,967.70. Same 1 in 50 Chance.
Walking Forward with the POC (in green) will allow us to eliminate the 1 in 50 chance and has a profit of $39,068.52. This falls in the 96th percentile of all 50 values when choosing randomly. Not Bad.
But…
Walking Forward with ROC (in Yellow) will allow us to eliminate the 1 in 50 chance and has a profit of $53,067.50. This not only beats all 50 values on random chance but beats the best value of 49 by over 20 percent.
Any way you can outperform your strategy by 20%
I will take it.
Edit: Sorry the pictures are so small. I can not get them to load larger? You might use the CTRL + or - to zoom in and out using a browser.
I appreciate the fact that you are trading one contract. So many people want to use maximum size, load up 10 or whatever they can afford, and then find out that trades can go both ways, which I guess surprises them.
I know you said no posting, but I had to throw this in.
Bob.
When one door closes, another opens.
-- Cervantes, Don Quixote
No problem Bob, comment anytime. My intent was only to keep the journal from going other places...to keep it within my interests. As you know, sometimes threads can go astray.
I also trade copy 2 Micro NQ's with and add-on called Flowbots.
I hold X times the maintenance margin for each contract within my account and all the markets can be a little different. NQ and MNQ is 3X. My capital is kept spread out between other markets because it serves my equity curve better than just several contracts within one. I am currently trading 7 instruments...9 if you count the micro's.
In the beginning I traded only NQ and there were no micro's. I also was discretional trading. So my equity curve looked like a heart rate monitor where somebody was having a massive heart attack. LOL
But...NQ is definitely my love and I am looking forward to the time I trade 2 mini contracts. That will be a really nice milestone for me.
No matter the length of your walk forward interval (1 Day, 1 week, 1month, etc.). The POC will be Today's best parameter or combination of parameters for your strategy. Your ROC will be the parameter with the greatest + change between Yesterday and Today.
Yesterday and Today are representing the interval that you have chosen.
The following example is a daily walk forward with one parameter that has 45 values,
Example 1
Picking the POC (Although Good) is really just old news. You are chasing the lead if you will.
Picking the ROC...You are trying to read tomorrows news in advance.
All parameters have a drawdown and this method has a tendency to work better through those drawdowns...So think of it as optimizing to a parameter that is coming out of a drawdown.
Hope that helps a little?
EDIT:
My basic method to calculate the ROC.
I run and optimization daily in ninjatrader and export the results to excel. Once in excel I calculate my ROC.
