April 14, 2019
The subscriber platform updates below are important because we have come full circle on many things and there are big changes on the horizon.
Crude Oil Trade Alerts
In recent months we have been coding our oil software and this has caused most crude oil trade alerts to be machine related. This will be changing starting this week because we are as good as done the code testing for crude oil (see more detail below on this topic). This will allow me to personally execute more crude oil trading on a personal basis.
So starting this week and going forward there will still be machine trading alerts but there will be much more of my own alerts also.
I have spoken to a number of our clients that wish for this to return also (because I have personal trade processes that are time tested and not experimental or high frequency such as when we were developing the oil software).
Swing Trading Alerts
If you have not been following our swing trading alerts and you are swing trading you should be watching these alerts closely.
We have had a blow out year in swing trading – we are hitting over 90% win rate and the swing trading wins have been unusually large ROI.
I expect this to continue as the structure of charting we are using for swing trading is just getting better and better as time goes on. Eventually (soon) we will be coding many of the equities for swing trading also BTW.
Crude Oil Machine Trading / Coding
In short (point form), this is a where we are at with coding the machine trade for crude oil:
- Time Frames (charting).
- We initially coded the oil machine trade software for the primary time frames (on charting) and over time found that all time frames were required for the software to trigger properly, this has been completed with the exception of two time-frames – hourly and fifteen minute. We have tested a few variations of the fifteen and one hour time frames that we are not completely satisfied with as of yet. Eleven out of thirteen are complete. This has helped consistency / accuracy of trade execution. The last two time frames will be coded over the coming weeks.
- Risk Tolerance / Threshold (variance, draw-downs, returns).
- Recently we had the risk tolerance throttled to a return level of about .25% per day, then we moved it up to .5%. At .25% and at .5% the software executed well with very little possible down-side scenarios. Then we released the throttle to .6%, .75% and .9% and as the throttle was released above .5% the potential draw-down variance escalated to an uncomfortable point by mid Friday trade last week. Now, it is important to note that I am referring to a ten contract size account, on a thirty contract size account this would be different because the risk tolerance variance is different. A few contracts on a thirty contract size account that are in an intra day loss position is not nearly the same as a ten contract account. For the purpose of our current coding structure we are coding for ten and five contract account size. Soon we will code for twenty, thirty and so on.
- Our experience last week caused us by the end of the week to settle with running the software for a .5% throttle for the next thirty days. This will allow the accounts to build with the potential of very little draw down.
- This will provide an average return of .5% per day on an average day with opportunity for up to 2% return on better days and .5% or so on loss days. The loss days and higher return days will be anomalies. After thirty days we will reassess and any throttle adjustments will be made in much smaller increments in future.
- Target Returns.
- What is possible on paper (or paper trading) and what is reality when put in to practice (as it relates to machine trading) are two different things (you cannot paper trade test this software with great accuracy). There are many reasons for this, but the primary reasons are order fills and tight time frame high frequency trading (one minute time frame or less). We have completely abandoned the idea of HFT on one minute or less time frames. This is highly competitive, we believe most participants are losing in this space, location of servers relative to the exchange and more become variables. On paper HFT offers 100% return per month (the math works) but order fills (especially on lower time frames) are at issue.
- So what is a reasonable return expectation for the time-frames we have identified as the most consistent time-frames with the lowest risk thresholds? We know .5% per day or 120% per year is reasonable (based on real life experience) and we believe 1% a day is attainable. We have seen up to 2% per day (in real life scenario) but the draw down scenarios are outside of our current comfort zone. Maybe in future (as returns are realized) we will slowly release the throttle toward that goal.
- For now we are going to focus on .5% per day with opportunity for some better days that we believe over a number of weeks will result in .5% – .7% on average. Any changes to the code will be very small and incrementally slower.
- The primary target we use is Renaissance Technologies reported returns;
- Renaissance’s flagship Medallion fund, which is run mostly for fund employees, “is famed for one of the best records in investing history, returning more than 35 percent annualized over a 20-year span”. From 1994 through mid-2014 it averaged a 71.8% annual return. https://en.wikipedia.org/wiki/Renaissance_Technologies
- If we achieve 120% per year we are doing exceptionally well, we have shown over short duration while testing recently that we can achieve this, the unknown is whether our software can achieve this consistently over a quarter, and a year or more.
- Order Flow (IDENT).
- Our software triggers on structure (charting time frames), timing cycles, direction of trade (momentum) and order flow (the IDENT program).
- Specific to our order flow, it is important to note that as the throttle is released the primary cause of variance or risk threshold widening is the IDENT order flow technology we are developing. In other words, if the throttle is opened to a return of 1% per day vs .5% then the draw-down possibility is much greater because the software is weighing order flow positioning greater than say chart structure. In short the software is following position bias and size based on identified machine trading market players. At issue is the size of account (as I noted above). I mention (or give some description) of the IDENT program and subsequent sizing and positioning as it relates to throttle because the stops widen greatly. A throttle at .5% on average will allow about 11 tick stops (often dynamic trailing) but at 1% the field range in play can be up to 40 or more ticks. It is a structured environment that becomes complex to explain in short but I think you get the idea. The bottom line is that it is most comfortable to have a minimum thirty contract size account for this scenario (as mentioned above) so that when up to five contracts are dealing with a forty tick variance it is not as uncomfortable as a ten contract size account.
- We will leave the oil machine trade code risk tolerance throttle for now at where we are most comfortable and where we know it works to the highest degree of certainty, at .5%.
- Any changes will be small and incrementally slow.
- This will achieve low down-side risk and consistent gains.
- Crude Oil Trade Webinar
- We recently ran a one day crude oil trading webinar and after it was complete I stated that we would run a series of shorter crude oil trading webinars to complete descriptions of various set-ups that I wanted to be sure the attendees all had. Anyway, I had planned to do these on Sundays in the afternoons immediately following but I have not had time. Soon I will be continuing these short webinars to complete the rest of the set-ups (structures) for trading crude oil.
- Swing Trading Webinar
- With our very successful swing trading platform (we have achieved 100% per year every year and the first quarter of 2019 has been outstanding) many have asked me to do a one day webinar and I planned to get one in before the end of April. This is still my goal and I expect to get a notice out this week for the one day swing trading webinar. Also the P/L for the swing trades will be released soon (Jen is on it).
- Webinars for the other algorithm models will also be scheduled soon, watch for updates.
Going Private, Personal Subscriber Conversations, Alerts, Personal Trading Set-Ups
Over the last number of days I have had personal Skype conversations with a number of our retail trading clients and commercial enterprise clients. All of which have asked that we remain at least semi public and all have asked that I provide more personal trade insight and alerts. Even though I desire to take our platform 100% private and only bring our current clients with us I am going to make the following adjustments to our plans and announcements will be forthcoming.
- I will be doing more personal trading, providing more personal trade set-up insight and alerts over the coming weeks and forward. I feel an obligation to this and to be honest it is win – win because many of the subscribers become friends and day to day the reciprocation of trading set up ideas is beneficial for me also. It is a win win, so yes I get it and I will adjust accordingly.
- Second, I will make adjustment to our “going dark” and only make it semi private. The details will be announced. I think a happy medium balance is to have a basic application process at entry. My main objective is keeping the “casino mind” retail traders out of our midst because I don’t feel it is my skill to train that out of a beginner. My skill is after the casino mind has humbled a retail trader. In other words, teaching structured discipline rule-set trading. So I will make adjustments and announcements will be made over the coming days.
- Over the coming weeks I will be reaching out to every subscriber on a personal basis to touch base (as time allows I will contact each of you and there is no obligation, I just think its important to do this at least once a year).
Reporting this Week
- Reporting this week likely won’t start until sometime Monday or Tuesday but any pertinent charting will be sent in the meantime. We spent the weekend re-setting the code (as described above) and reaching out to many of our subscribers for input.
Thanks so much everyone for being part of this journey, your patience in our petri dish science approach, supporting our work and sticking with our efforts to develop a platform by traders for traders for a better trading future and associated freedom.
Any questions or thoughts send me an email email@example.com.
Best and peace.