Dump it Here

Skate · 23 October 2022

InsvestoBoy said:
@Cam019 Imagine if he had let himself be influenced by the people on this forum to stop trading, or tweak his system or whatever, at pretty much the lows.

@InsvestoBoy made a great comment in "cam019-asx100-monthly-superannuation-portfolio" today about not being influenced by others. There were comments made at the time when his trading system was well underwater, raising the question, "when do you stop trading your system"?

When do you stop trading your system?
When @Cam019 was asked this question, he was steadfast in his reply:
"I can only speak for myself, but my answer to "When do you just Stop!" is; I don't."

Cam went on to say:
"Now would be the worst time to stop executing trades in this system, however it is probably the most likely time for someone to do so."

This allows me to make a few general comments
1. When you have a trading system that really suits you it is very easy to follow the system’s signals.
2. A trading system that fits you allows you to focus on trading it with conviction & doing it well.
3. The hold time in Cam's case "monthly periodicity" I'm sure it feels natural to him in a unique way.
4. If you believe your strategy has an edge, you will trade with confidence.

Summary
Kudos to Cam for sticking with the strategy & achieving good trading results since August this year.

Skate.

Skate · 23 October 2022

Strategy development, Backtesting & Optimization
This topic has been done to death but I still have the feeling that most traders optimize their systems & select the wrong metric to work with. In reality, you will make faster progress in the development phase if you keep your trading system code simple at first & learn how to correctly optimize your trading system to avoid curve fitting.

Let's address the Elephant in the room (Backtesting)
If your trading idea "does not work" in a simple backtest then "it will not work in real-time trading". Backtesting's unique feature is that it eliminates the bad ideas you can come up with that have no edge or the likely hood of being profitable.

Skate.

Skate · 23 October 2022

When a strategy backtests well
This is the time to start the optimization process. IMHO, the smartest way to start optimizing your strategy is to optimize a "single parameter" at a time so you can really see how stable the parameter value is as you vary it.

Curve fitting
"Optimization" is what gets most traders into trouble with curve fitting. My advice is to keep it simple, baby steps at first. The stability of the optimized process is more important than going straight for the best backtest result. So don't fall into that trap of immediately going for the best, "stability of results" is what you are after.

Skate.

Skate · 23 October 2022

It's extremely important to "step up" the optimization process
This means carrying out one optimization at a time because we want to ensure every parameter value is stable before moving on.

Position Size
When backtesting your system ensure that you select a constant position size (to remove the impact of compounding) ensuring your edge is stable.

Skate.

Skate · 23 October 2022

Let's talk about the maximum historical drawdown of a system
This metric "historical drawdown" is not given the importance it deserves & at times overlooked. The maximum historical drawdown is simply the percentage drop from a high in the equity curve to a subsequent low at any point throughout the whole backtest for every single bar.

Drawdowns are what cause your emotions to spiral
This percentage drop from the high point to the low can take on a new meaning after your backtesting is completed. I'm just saying, make sure you can handle this percentage when trading your system live as it takes on a new meaning when you do. Just don't be flippant with this backtest metric. If you do, you'll quickly find out that drawdowns will cause your emotions to spiral out of control. So don't let this genie out of the bottle, if you do, you won't stay in this game very long.

Skate.

Gringotts Bank · 23 October 2022

If using fixed profit and loss stops (and no regular or trailing stops), you can also position size in order to standardize the loss on each trade. eg. If you want to risk $500 per trade, the PS formula is:

Positionsize = ($500 / points from entry to stop) / $ per tick.

It's a nice way to do it because some trades can have targets/stops way off in the distance. You don't necessarily have to skip them. Just use a scaled down position size. Vice versa for trades with a nearby target/stop.

Skate · 23 October 2022

dyna said:
Old mate Nick Radge , too , has been fearless in his honesty. More power to him for that.
That six figure number with a minus sign in front of it, just turns my stomach over.

Trendnomics said:
I must be doing something wrong....

2020 was -9% closed profit, current year is +8% open+closed profit

Not seeing much change....just cruising....

@Trendnomics it appears over the last two years your portfolio has gone nowhere being down -9% in 2020/21 & up +8% for the current year which is on par with the "The Chartist" equity curve for the same period.

Being 8% up for the current year is impressive
During this period, "this year & the last" has been a tough slog even for the best of traders. Posting your returns in "percentages" is a true indication of how your system/s has been traveling during that period. Reporting dollar amounts are only relevant to the total funds of a trading account. Even so, @dyna & I thought "The Chartist" PnL for this calendar year to the tune of -$927,760 was stomach-churning.

"The Chartist" main account has gone nowhere over the last two years
No growth for two years puts real trading into perspective for most of us, meaning if it's been tough for "The Chartist" it certainly has been a tough year for us mere mortals. I'm just saying if "The Chartist" main account growth has gone nowhere over the last two years "something has changed". My guess is, "it's not the strategy" being traded.

Skate.

Skate · 23 October 2022

Gringotts Bank said:
If using fixed profit targets and profit stops (and no regular or trailing stops), you can also position size in order to standardize the loss on each trade. eg. If you want to risk $500 per trade, the PS formula is:

Positionsize = ($500 / points from entry to stop) / $ per tick.

It's a nice way to do it because some trades can have targets/stops way off in the distance. You don't necessarily have to skip them. Just use a scaled down position size. Vice versa for trades with a nearby target/stop.

@Gringotts Bank thanks for the input
Your alternative position size is a practical solution for trading. In my previous post, I was making the point of the value of using a constant position size when backtesting (to remove the impact of compounding) ensuring your edge is stable. Using compounding skews the results making the backtest results less than useful.

I'm amazed at how much we think
Being a wet day, I have been looking at a series of backtest reports on how having additional parameters affects the exposure percentage of a strategy. For me, protecting my capital is the prime objective.

My mind quickly shifted to excessive drawdowns
Sometimes excessive drawdowns can be stomach-churning, which gave way to my last post. Ensuring you don't suffer from a large drawdown, allows you to stay in this game. Keeping drawdowns in check always "comes at a cost". Keeping drawdowns to a "stomachable level" is hard to achieve but not impossible.

Being selective about when to take a signal
That was going to be my series of posts today until I was sidetracked by the comments of @Trendnomics concerning @Cam019's fabulous strategy turnaround. In my opinion, when market conditions change as they have in the last two years you simply have to be more selective about what & when to trade or your returns will suffer as shown in my previous posts.

Skate.

MurariMikeTrader · 23 October 2022

Skate said:
Hi, @MurariMikeTrader welcome to our community
Thank you for making your first post in this thread, minutes after joining. Over time I have made a series of posts on a 20-period breakout-out strategy commonly referred to as the "WTT Strategy".

To save me from posting repetitive information again
Use the search feature (RH Top corner of the menu bar) & you will find approximately 140 posts using the WTT keyword. I'm sure reading a few of my posts will answer questions you have about the strategy.

View attachment 148347

Enjoy looking around
If a thread interests you, read it slowly, don't fall into the trap of "speed reading" as it doesn't give you time to fully understand the post let alone the time to memorise all the important stuff. Also, make the search feature your best friend as "if you can think of a question", I'm sure it has already been answered.

Skate.

Thanks for the response Skate.
If you don't have or don't want to share the Amibroker code for your "Final Version" of Nick's WTT Strategy (with Lipstick applied) then just say so. I'm a big boy, I can take it.
But please don't ignore the question and instead send me on a wild goose chase through your posts merely discussing the subject.

Skate · 23 October 2022

MurariMikeTrader said:
If you don't have or don't want to share the Amibroker code for your "Final Version" of Nick's WTT Strategy (with Lipstick applied) then just say so. I'm a big boy, I can take it. But please don't ignore the question.

@MurariMikeTrader, sorry, I should have said that I don't give, sell, swap, or divulge any of my systems. As you are a new member of our community I merely suggested that you use the search function as my way of helping you try to find additional information about the strategy elsewhere. This way you help yourself by stretching your mind & research abilities allowing you to learn things more thoroughly.

Skate.

ducati916 · 23 October 2022

Skate said:
Let's talk about the maximum historical drawdown of a system
This metric "historical drawdown" is not given the importance it deserves & at times overlooked. The maximum historical drawdown is simply the percentage drop from a high in the equity curve to a subsequent low at any point throughout the whole backtest for every single bar.

Drawdowns are what cause your emotions to spiral
This percentage drop from the high point to the low can take on a new meaning after your backtesting is completed. I'm just saying, make sure you can handle this percentage when trading your system live as it takes on a new meaning when you do. Just don't be flippant with this backtest metric. If you do, you'll quickly find out that drawdowns will cause your emotions to spiral out of control. So don't let this genie out of the bottle, if you do, you won't stay in this game very long.

Skate.

Would you agree that the timing of the drawdown (maximum) is also an issue?

To turn the system on and encounter a maximum drawdown cannot inspire much confidence. Do you keep going?

jog on
duc

Skate · 23 October 2022

ducati916 said:
1. Would you agree that the timing of the drawdown (maximum) is also an issue?

2. To turn the system on and encounter a maximum drawdown cannot inspire much confidence. Do you keep going?

jog on
duc

@ducati916 they are two important issues you have raised.

Issue (1)
With Amibroker the maximum historical drawdown from a high in the equity curve to a subsequent low at any point throughout the whole backtest for every single bar the timing is irrelevant during backtesting. (a) The timing would be an issue if the maximum drawdown happened just after deciding to actually trade the strategy. (b) If a strategy established a profit buffer (closed profits) it becomes less of an issue. (c) Mature strategies are unaffected to a certain degree by the timing. After giving you those 3 scenarios the timing issue would hold 3 different levels of emotional variations. So yes, you are correct, the timing of the maximum drawdown can certainly be an issue.

Issue (2)
To be honest the maximum drawdown must be at a level that you can tolerate & tolerate easily. I'm not in any way being flippant but rather it depends on your tolerance for risk. If I may make a reference back to my boxing days, I'm sure there were some I fought who "felt no pain" & it's the same with trading, the pain threshold for losing money varies from one trader to another.

This calendar year backtest (2022)
I pride myself on coding strategies that exhibit drawdowns within my pain threshold. I also remarked that I couldn't trade "Dunn's Equity Curve" because of his "Average Maximum Drawdown" of 36.2%. I don't think there would be anyone who would be comfortable trading that system.

Don't kid yourself 2022 calendar year has been tough
The Hybrid Strategy Backtest is from one of my seasoned performers. It's stuck around a long time because it's a handy performer with a low maximum drawdown. Trading these metrics is easy because they are well within my comfort zone. When you are confident with a strategy "it's comfortable to trade".

Skate.

rcw1 · 23 October 2022

Good evening
Discovery / ex ante, inextricably interwoven, and the process may well have identified a threat (s), for example, mergers.
Real time testing, watch and wait and wait and wait some more, then spend, has merit for mine too. Get to know the beast you want to make coin from. Patience. Milk the cow, so to speak.

Have a very nice week.

Kind regards
rcw1

qldfrog · 23 October 2022

ducati916 said:
Would you agree that the timing of the drawdown (maximum) is also an issue?

To turn the system on and encounter a maximum drawdown cannot inspire much confidence. Do you keep going?

jog on
duc

So the luck part in a system trading, even for a system with an edge.
If your drawdown comes at the start of your investment period, and that DD coukd be higher, even higher than your backtest runs
And the problem is : how do you know if your system is failing and so you should stop, or this is just a timing issue.
So far the only way i can get some idea is a comparison between relatively similar systems, both between my own systems and reading the experiences of fellow ASF system traders

qldfrog · 23 October 2022

qldfrog said:
So the luck part in a system trading, even for a system with an edge.
If your drawdown comes at the start of your investment period, and that DD coukd be higher, even higher than your backtest runs
And the problem is : how do you know if your system is failing and so you should stop, or this is just a timing issue.
So far the only way i can get some idea is a comparison between relatively similar systems, both between my own systems and reading the experiences of fellow ASF system traders

I should also add that based on your system, there can be a ramp up phase, the initial sorting of the whey which generally results in underperformance in the first few weeks of starting a strategy..duration variable based on the ramping speed of your code

ducati916 · 24 October 2022

Skate said:
@ducati916 they are two important issues you have raised.

Issue (1)
With Amibroker the maximum historical drawdown from a high in the equity curve to a subsequent low at any point throughout the whole backtest for every single bar the timing is irrelevant during backtesting. (a) The timing would be an issue if the maximum drawdown happened just after deciding to actually trade the strategy. (b) If a strategy established a profit buffer (closed profits) it becomes less of an issue. (c) Mature strategies are unaffected to a certain degree by the timing. After giving you those 3 scenarios the timing issue would hold 3 different levels of emotional variations. So yes, you are correct, the timing of the maximum drawdown can certainly be an issue.

Issue (2)
To be honest the maximum drawdown must be at a level that you can tolerate & tolerate easily. I'm not in any way being flippant but rather it depends on your tolerance for risk. If I may make a reference back to my boxing days, I'm sure there were some I fought who "felt no pain" & it's the same with trading, the pain threshold for losing money varies from one trader to another.

This calendar year backtest (2022)
I pride myself on coding strategies that exhibit drawdowns within my pain threshold. I also remarked that I couldn't trade "Dunn's Equity Curve" because of his "Average Maximum Drawdown" of 36.2%. I don't think there would be anyone who would be comfortable trading that system.

Don't kid yourself 2022 calendar year has been tough
The Hybrid Strategy Backtest is from one of my seasoned performers. It's stuck around a long time because it's a handy performer with a low maximum drawdown. Trading these metrics is easy because they are well within my comfort zone. When you are confident with a strategy "it's comfortable to trade".

View attachment 148411
Skate.

1(a). It is the drawdown that occurs at the start of a trading strategy that (I would think) carries the most impact because putting aside any emotional issues, the loss of capital will start to impact the number of trades that can be taken.

From what I have seen from systems, this is an issue. The 'edge' (it would seem) relies on a minimum number of trades taken, which allows the numbers generated by your backtests to play out. Part of the 'edge' is the number of bets laid. N'est pas?

2. I agree that 'drawdown' is a major emotional issue and can play havoc with compliance to a system. I don't think that too much can be added to comments already made. You'll only really know what you can live with once you have lived with it a few times. So the 'Hybrid' at 5% is certainly tradable in the real world.

I'm sure there are systems traders that have different systems for different market conditions: big picture stuff, bull market, bear market? Or is the nature of building a system, a system for all market conditions?

jog on
duc

qldfrog · 24 October 2022

ducati916 said:
1(a). It is the drawdown that occurs at the start of a trading strategy that (I would think) carries the most impact because putting aside any emotional issues, the loss of capital will start to impact the number of trades that can be taken.

From what I have seen from systems, this is an issue. The 'edge' (it would seem) relies on a minimum number of trades taken, which allows the numbers generated by your backtests to play out. Part of the 'edge' is the number of bets laid. N'est pas?

2. I agree that 'drawdown' is a major emotional issue and can play havoc with compliance to a system. I don't think that too much can be added to comments already made. You'll only really know what you can live with once you have lived with it a few times. So the 'Hybrid' at 5% is certainly tradable in the real world.

I'm sure there are systems traders that have different systems for different market conditions: big picture stuff, bull market, bear market? Or is the nature of building a system, a system for all market conditions?

jog on
duc

On your last point, while i dream of a fit for all system..holy grail.., realistically I see that as a set of systems each good for specific market phases and un clunching reengaging when the conditions changes.
Ideally one optimised system per market condition at a time.
Semantically speaking, it could be a giant single code with
If market condition 1 then subsystem code 1
Or
If market condition 2 then subsystem 2
Etc
But truly different potentially:
Long trending, short trending, reversal, etc
That's my current aim..still an aim, not an outcome.
And the reason i currently have 3 active weekly systems.. but only one engaged

Skate · 28 October 2022

Trendnomics said:
I can see alot of effort being put into trading a lot of different strategies, but not much reward.

Maybe time to consolidate and re-assess? Or blame it on the market changing?

I can't entirely agree with this statement
I believe system traders spend an enormous amount of time trying to hone this craft. I do accept that the amount of effort put in doesn't automatically correlate to the reward side of the equation.

Skate.

Skate · 28 October 2022

Trendnomics said:
Maybe throw in some Jim Cramer quotes as well? We all know CNBC is a very reliable market reporter...

The role the media plays in trading is actually an interesting topic
Even Twitter is now viewed as a defacto media outlet. Traders when they are trying to understand the markets turn to market commentators, & news outlets. We can all have an opinion about why the market moves as it does but media outlets are simply better at making all this stuff up. The story is packaged in such a way you tend to believe it as gospel.

Media outlets & commentators do this on a daily basis to keep readers
Every time there is a move in the markets, suddenly stories get published explaining the reason. As @Trendnomics eluded to these media outlets vary in their quality of information & there's nothing wrong with that. I find different media outlets have completely different objectives. As traders, we're trying to make money in the market while the media outlet is trying to get people to watch or read their content.

Skate.

Skate · 28 October 2022

Trendnomics said:
I can see alot of effort being put into trading a lot of different strategies, but not much reward.

Maybe time to consolidate and re-assess? Or blame it on the market changing?

ducati916 said:
I'm not sure if you are legitimately trying to be helpful. Your ending comment however is most definitely not helpful.

Nobody cares what others think
At the end of the day, it will make little difference what you say because if the information is outside one's belief it will be rejected. Some members have open minds & display varying degrees of openness accepting some views while rejecting others & that's healthy. But, I've found there are a few members who have a closed mind to new information rejecting everything that's contrary to their own. One thing I do know is "if something resonates" it can make a big difference.

Quality posts
When members make quality posts it really does influence how others think to some extent. Expressing alternative views in detail is more valuable than just making one-line statements.

Skate.

Dump it Here

Similar threads

Privacy & Transparency

Privacy & Transparency