One of the ideas I have for my new system (still in design phase) is to also have an eject lever in the the aircraft. To set a point below all drawdown levels obtained from backtesting that defines that something has failed to such a point that the integrity of the system has failed.I think one of Mr Skate's motto/rule is of the kind of
"Do not 'play' what you can not afford to lose"
Rule number 1 indeed
Haven't thought this through too much but I'm not sure setting a point below backtest drawdown levels is an accurate indicator of system integrity failure. Think about the Covid crash. Certainly many systems (particularly those based on longer time frames such as weekly or monthly) would have experienced much larger drawdown than previously shown by backtests prior to that crash. I know my weekly did--look at my recent post of my weekly's P&L through the crash. Reality is the integrity of my system wasn't broken we just had not seen the market crash before so quickly so my backtests on historical data gave no insight into the system's behavior in such market conditions. I've since tweaked my weekly to better handle similar rapid crashes. This is the problem with backtesting on historic behavior--it gives you no insight into new market behaviors that may emerge in the future. One of the things I think would immensely help system testers is the ability to generate random market data to augment your system testing with actual historical market data.One of the ideas I have for my new system (still in design phase) is to also have an eject lever in the the aircraft. To set a point below all drawdown levels obtained from backtesting that defines that something has failed to such a point that the integrity of the system has failed.
Thank you for your post, it shows me that this feature does need more depth of thought. The eject from the system feature may still be useful depending on the type of system a person is using. I'm planning on a system that trades both long and short so it will have a built in characteristic to be out of longs and looking for shorts when a market turns down. This characteristic will limit losses in itself from a market crash situation, but the area where it may be useful is from consecutive losses. So it's not the market crash that I'm most concerned about it's the change in the characteristics of the market that can shoot me down in flames.Haven't thought this through too much but I'm not sure setting a point below backtest drawdown levels is an accurate indicator of system integrity failure. Think about the Covid crash. Certainly many systems (particularly those based on longer time frames such as weekly or monthly) would have experienced much larger drawdown than previously shown by backtests prior to that crash. I know my weekly did--look at my recent post of my weekly's P&L through the crash. Reality is the integrity of my system wasn't broken we just had not seen the market crash before so quickly so my backtests on historical data gave no insight into the system's behavior in such market conditions. I've since tweaked my weekly to better handle similar rapid crashes. This is the problem with backtesting on historic behavior--it gives you no insight into new market behaviors that may emerge in the future. One of the things I think would immensely help system testers is the ability to generate random market data to augment your system testing with actual historical market data.
My system is planned to be based on weekly price bars as you have mentioned above and as you may have seen from my previous postings, I am very focused on managing risk. I don't think that I could do what you have done when you held to your system through such a long period. I would make sure I knew my system inside out as you did but I would also have my wife looking over my shoulder questioning me. So what I've got on the drawing board is limiting each trade to a 5% account exposure. I also plan to eliminate gap down risk by using options, this means that my 5% max risk per trade will be unbreakable.Think about the Covid crash. Certainly many systems (particularly those based on longer time frames such as weekly or monthly) would have experienced much larger drawdown than previously shown by backtests prior to that crash.
Having a wife looking over your shoulder is a good thing…keeps you accountable ?My system is planned to be based on weekly price bars as you have mentioned above and as you may have seen from my previous postings, I am very focused on managing risk. I don't think that I could do what you have done when you held to your system through such a long period. I would make sure I knew my system inside out as you did but I would also have my wife looking over my shoulder questioning me. So what I've got on the drawing board is limiting each trade to a 5% account exposure. I also plan to eliminate gap down risk by using options, this means that my 5% max risk per trade will be unbreakable.
Without giving away your secret sauce it would be great to hear a little more about how you envisage NTW operating. Also, are you planning to use AB or something else for system evaluation?I came to this thread to have a casual interesting read, ‘a drink at the bar’ so to speak, and found a group of people immersed into system trading. I’ve always loved systems, in my day to day life I create systems to do most things and not just a way to do it but the most efficient way. My wife’s systems of doing things use to drive me crazy and I’d try to show her a better way but I grew up and realised that you can’t move a mountain, I also found a warming richness that came from seeing things from another person’s eyes, she’s taught me a lot. My mind has been impregnated with the idea of building a system of my own, a new baby, he or she is only an embryo at the moment but I’ve already been thinking about names. I’ve come with the NTW system. It stands for Not The Worst system, and I hope it won’t be, we all want our kids to have a good and successful life don’t we. I’m not sure how useful or interesting this post would be to anyone reading it but like any new dad, you just want to go and tell everyone.
Go and create a thread based on your process.I came to this thread to have a casual interesting read, ‘a drink at the bar’ so to speak, and found a group of people immersed into system trading. I’ve always loved systems, in my day to day life I create systems to do most things and not just a way to do it but the most efficient way. My wife’s systems of doing things use to drive me crazy and I’d try to show her a better way but I grew up and realised that you can’t move a mountain, I also found a warming richness that came from seeing things from another person’s eyes, she’s taught me a lot. My mind has been impregnated with the idea of building a system of my own, a new baby, he or she is only an embryo at the moment but I’ve already been thinking about names. I’ve come with the NTW system. It stands for Not The Worst system, and I hope it won’t be, we all want our kids to have a good and successful life don’t we. I’m not sure how useful or interesting this post would be to anyone reading it but like any new dad, you just want to go and tell everyone.
It has helped me greatly in my own voyageGo and create a thread based on your process.
It is interesting for others, and a way to record your process, your thought at a time , a journal.
I would encourage you if you are dedicated to a long term involvment
I’ve come with the NTW system. It stands for Not The Worst system
To a great degree, our success or failure in the market is a function of our luck
We all like to think that our trading results are a direct consequence of our effort in selecting a position to enter, but to me, it's all about the ability to effectively manage luck. Small traders know that sometimes they will have streaks of bad luck, & other times everything they touch is golden. The key to surviving your run of bad luck is by minimising the amount of damage being inflicted. (how to minimise the damage, is for another discussion).
The other side of the equation is to capitalize when your luck is good
Designing good money management principles to a degree helps but our "success or failure" is having an aggressive approach toward selling but that isn’t enough. In designing a money management system, your first consideration should be how well it protects your capital should your run of bad luck continues. This should be the topic of the conversation.
Skate.
without disclosing too much (actually disclosing MUCH), one way is to be long on bears..This rather important point didn't seem to garner much interest.
Risk, what we all trade is defined as the 'unknown future'.
As such, that is equivalent to 100% unknown or 100% luck (good or bad).
A system or methodology seeks to (as Mr Skate stated) capitalise good luck and either capitalise or minimise bad luck. The system's traders on this thread, as far as I can see, essentially follow the model of (a) minimise through exit of position(s) if bad luck is an outcome and (b) holding a position (long) for as long as good luck holds.
Of course, pure luck, you would expect to follow a 50/50 type distribution, balancing out over time. Clearly again, the lack of long/short systems on this thread and others, there is an upward bias to (stock) markets vis-a-vis Central Bank interventions and arguably any number of other macro-economic variables.
The issue therefore is that trading that bias makes sense, but, increasingly, when bad luck intervenes, the break, due to a number of reasons, has increased in its speed and % decline.
Therefore 'exits' need (i) accuracy, (ii) speed of recognition, (iii) management of liquidity issues, (iv) whether the entire portfolio is exited and whether there is a hedged type of position or portfolio that runs concurrently.
I know that Mr Skate has emphasised exits far more than entries. I'm just curious to know whether the systems run, run a hedged system alongside/part of, the long system?
jog on
duc
without disclosing too much (actually disclosing MUCH), one way is to be long on bears..
What i mean here is some ETFs are bear and so a long only system including them allows for short play,
you can also see Gold miners positions as "short" etc;
So my hope is that in a crash, my long only systems will go long into the "artificial" shorts and allow gain in crash
Works in BT..but backtests...
I know that Mr Skate has emphasised exits far more than entries. I'm just curious to know whether the systems run, run a hedged system alongside/part of, the long system?
I’m a big fan of profit based exits. I introduced one into my weekly some time ago. The reason I have a profit exit is that it suits my style of trading—I don’t like having large unrealised profit open for too long. Works for me. I’ve also introduced a swing system which exit intraday. Being a swing system I don’t have the same concern that I have with my weekly. In the case of my swing system it is just there to realise larger than expected profit from big intraday moves.I'm a trend trader
I trade mainly one type of strategy that has been proven to be profitable over the long run. I'm only interested in making the most money with the least amount of time & for these reasons I'm drawn to Mechanical System Trading.
Hindsight
We are chart "experts in hindsight" but system trading picks the moves in real-time. Also using a "Take Profit Stop" these days is certainly a sign of the times by adopting this feature in your strategy allows you to advantage of the shift in trading since COVID.
A recent trade
This is one of my recent trades that indicates the use of a take-profit stop.
With a "Take Profit Stop" activated
The security is (VUL)
View attachment 132672
With no "Take Profit Stop"
You can argue that holding (VUL) a little longer (an additional 5 weeks) was fractionally advantageous but after the "Take Profit" exit & with the profits locked in we can "immediately" look for the next ride.
This is the point I want to get across (in Hindsight) the move eventually went higher but when we "Took Profits" we had "NO IDEA" how the next series of bars would have played out. (better safe than sorry)
View attachment 132673
Skate.
OK, luck. I totally reject the assumption that my trading outcome is due to luck. So, I disagree with the quoted statement. For me, trading luck is a random, spontaneous movement in price. Luck doesn't start a trend nor does it sustain one.
As a discretionary trader I experience lots of lucky moments. Buying CHN the day before their resource upgrade, that's lucky. Buying WBC just before prices tanked, that's unlucky.
The long term outcome of a trading system or methodology isn't the result of luck but the existence of an edge.
In order to profit the edge must be positive, like the edge in a casino. Trading in an undisciplined manner (gambling) generally has a negative edge. A gambler can be lucky for a short period of time but over the longer term it's the gamblers edge that determines the outcome.
The system traders in this thread have recently discussed the issue of timing risk. They incorrectly attribute their early results to "luck". Timing risk is merely one of the many risks to be addressed in a trading plan.
LOL, try managing the unknown future, no way, too many possible outcomes. Managing luck, more LOL.And what exactly is an edge but a systemised management of the unknown future or 'luck'?
Right or wrong it doesn't matter
The 'Dump it here' thread is not about right or wrong or whether you’re right & they're wrong that's not what is important. What is important is the choice of words members use to express an alternative point of view & expressing an opinion from experience in a respectful manner.
OK, luck. I totally reject the assumption that my trading outcome is due to luck. So, I disagree with the quoted statement.
And what exactly is an edge but a systemised management of the unknown future or 'luck'?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?