Dump it Here

[Newt]

1. Remember we are still in an elevated Vol. environment. Daily moves are still amplified compared to 'normal' metrics. The same things still make prices move. (a) The macro-environment, (b) the sector specific fundamentals and (c) the individual fundamentals of the stock. If the fundamentals of a sector are considered to be optimal, then price will move a long way past those fundamentals. Tech. being an obvious example, gold currently. Momentum is measured by any number of indicators. It is a different game. Momo Vol. is always higher (in normal circumstances) in an elevated Vol. environment, it is higher again. My guess is that your systems are written in such a manner as to look (in normal circumstances) for momentum based trades. The system(s) are still finding them: just at an elevated Vol., hence the positions as a portfolio, are far jumpier than they would normally be.

2. I think T/A is the starting point. Internals are the next step. Macro-analysis being either the starting point or finishing point.

3. This is where macro is so important. If you 'knew' that the market was going up, then intra-day volatility is less of an issue. Markets operate pursuant to macro drivers. Sure day-to-day stuff happens, a news story spooks the little guy or whatever. Individual stocks will jump to their own narratives, earnings, etc. What are the big drivers of stocks, commodities, currencies, bonds, etc. Then what are actually looking at and for, totally changes. You are looking for stretch points and contractions within a trend. Stretch points are sells or hedges, contraction points (what we just had) is an opportunity to really load up for a big move. Think 2009 - 2020. There were 3 or 4 contraction points along the way of that huge move to take profits and reload. My system (if you could call it that) is measured (usually) in years. The big macro drivers do not flippe-floppe: they are huge and powerful. The 'virus' is not the origination of a new world order, it is a transient blip. On the other hand, the China/US issue could, like the US/Soviet issue, impact the macro trend if it continues to escalate.

duc

Forever indebted to many of you for the constant inspiration, stimulation, and backside kicks from time to time. There's some gold just on this last page. Duc, those first 3 points (above) alone deserve gold-plating and being pinned for any new system trader (probably any new market investor in general). Don't think I would have even understook point 2 in the beginning, but now that I'm "less than clueless" from many years learning 1 and 2 are wonderful. 3 is scarey, but trying to keep up and learning in your "Trading the Trend" thread.

QFFrog, your post inspired me to run a few old strategies for just this year (all weekly), and the outcome was something like what Skate described - surprisingly good results for early versions of current strategy or older strategy systems. That doesn't mean I'd still like to be trading the older code - see below.

Skate your prolific thread continues to greatly inspire and during Covid downtime around Easter led to a few weeks of intensive testing of different ideas for GTFO, ROC filters with variable parameters, various buy filters. Ironically I had earlier this morning been playing the game of "what if I jump back 12 months and tell my self to sell everthing and wait a year". I'd have more capital right now, but I'm certain the pressure cooker environment this year has forced some valuable evolution in my trading/coding maturity:

1. Backtesting (god knows how I ever came up with anything decent years ago with the amount of crazy over-fitting, poor coding, almost no graphics, clueless about Monte-Carlo). Even the "number of positions" debate many pages ago helped highlight possible improvements in optimizing code
2. DD and crash defences - previously of the "less is better"camp, found that wasn't much fun in Feb/March this year - more "condoms" in place now.....
3. Returns - believe changes in systems this year very positive for returns, and slight improvement in av loss and maxDD

So, on the time travel scenario, actually feel like last 6 months may possibly have been most valuable for future earning since started many years ago. Difficult times can lead to positive change if you continue regualar dispassionate review and try to learn from what's happened. Even seeing Skate putting out new work (HappyCat etc) shows the value of hard work, versus "set and forget".

Don't want to affect new traders here too much until I have a better track record under my belt, but greatly appreciate the time and effort so many put in to sharing performance and learning here.
Never underestimate the benefit of frank benchmarking against your peers.

 

[Newt]

The ETF versus individual stocks debate is also interesting. I suspect Duc is right that ASX is each year becoming more "US-ified". If so, we'll get more mean-reversion, shorter or choppier long trends and systematic trend trading returns may tail off a bit. Diversification into mean reversion may make more sense.

Find it fascinating how Duc has matured into trading as her does (advanced discretionary systematic approach?) in US ETFs, but don't think I'm anywhere near competence or skillset to make that work currently.

 

[Skate]

Skate your prolific thread continues to greatly inspire

@Newt thanks for the kind words. At one stage I didn't think this thread would develop but as you have eluded to it's because on the ongoing quality contributions.

Even seeing Skate putting out new work (HappyCat etc) shows the value of hard work, versus "set and forget".

Back to protection & the "HappyCat Strategy"
The new "HappyCat Strategy" seems "new" but to me, it's quite old. The "HappyCat Strategy" has been a "long" work in progress & could develop into my best strategy to date. I've got into the habit of wearing 5 condoms when trading these days as I'm using the added protection a comfort blanket. I'm wondering if I could manage to peel back 4 condoms & get away with just wearing one.

Confidence
When trading you need the confidence to keep pulling the trigger & keep going even when all feels lost. The old momentum strategy had been tested to death before making it to "live trading" so with some update code it can only help rather than being a hindrance. Trading the lean mean old "Momentum Strategy" would still be a stretch. Having a lack of guts it's only fitting the old strategy needs to be resurrected & be given a second chance. It would be nice to paper trade the "Momentum Strategy" alongside the "HappyCat Strategy" for a comparison.

What I'm thinking
After looking at the backtest results I'm inclined to give it a go & paper trade it.
























Skate.

 

[frugal.rock]

Nice work Skate.
Have restarted my Uni course, and having another go at Fundamental Programming using JavaScript.

Will have to stay off the forum a bit to get it done.

It may be the last subject I do.... depending on Mr Market and the soldiers commander making sound, wise plans and decisions to avert disaster.

Thanks for your time and consideration.

 

[Skate]

What I'm thinking
After looking at the backtest results I'm inclined to give it a go & paper trade the "Momentum Strategy".

I'm going to paper trade both strategy for a comparison of apples to oranges
I've decided to paper trade both strategies (1) The "HappyCat Strategy" & (2) My old updated "Momentum Strategy". Both strategies enjoy the same momentum base code but that's where the similarities end. In the short term, I'd expect them to crack off returning similar results but over time they should start to drift apart. Paper trading adds confidence & confidence to trade either one & that's paper trading will decide.

I'm wondering
Really, can an old lean momentum strategy be given a new lease of life by swapping out a momentum indicator with fresh modern momentum coding - only time will tell.

The old Momentum strategy has been renamed
The "HappyCat Strategy" remains unchanged but the old "Momentum Strategy" will be renamed as the "HappyCat v2 Strategy". They both enjoy a similar logo with slight differences to define them.

Comparisons have started
The next few captures are a snapshot of how they currently compare.



1. "The HappyCat Strategy"











2. The HappyCat v2 Strategy (the old "Momentum Strategy, revamped & renamed)





Skate.

 

[Newt]

I'm just not sure about Mr Happycat 2 - his gold skin is a bit too flashy, and there's something about his pose that makes him look like he's hiding something?

V1 however looks much more trustworthy

 

[Skate]

I'm just not sure about Mr Happycat 2 - his gold skin is a bit too flashy, and there's something about his pose that makes him look like he's hiding something? V1 however looks much more trustworthy

@Newt you might be on to something there
What HappyCat v2 hides the exit strategy that relies on "momentum" to switch off. Admittedly it will be excruciating waiting for momentum to switch especially when you are looking at the share price is in freefall. Nerves of steel will be required to trade this strategy as designed.

It's happened to me
In saying this, we have all experienced selling a position only to see the position reverse & take back off as soon as we sell. HappyCat v2 alleviates this issue by hanging onto positions "when" the buy condition is still "true".

FYI
The exit condition is the reverse of the entry condition, meaning the momentum is either on or off - there is no in-between.

Skate.

 

[Skate]

Chart Example - it can be excruciating to watch
@Newt let me show you the "dark side" of the shiny "Happycat".

The momentum ribbon is for individual stock (not applied to an index)
A position is sold when the momentum ribbon turns red. The strategy buy when the ribbon is green & sell when the ribbon turns red, simple & normally effective. (look at the hold period & imagine the feeling of missing the "huge" move) Momentum is a mathematical formula applied to individual stock & it's not a filter applied to an index.

It pays to remember this is the exception "NOT" the rule
The chart is posted to display a negative side trading momentum (entering & exiting on momentum only).



Skate.

 

[Newt]

Holy cow, that would be very hard to stomach Skate.

Hard to believe it pulls that much profit when a trade like that can be holding up the portfolio. Very strange business at times this trading....

 

[Trav.]

The exit condition is the reverse of the entry condition, meaning the momentum is either on or off - there is no in-between.

@Skate I took some inspiration from your post above as I did some further education on momentum and decided to try a few different things in my code which showed some interesting results.

The there is no in-between had me think about RSI and Stochastic's again and the ability to set levels for overbought / oversold signals and obviously anywhere in between.....

So I ended up running some tests to see if a better entry 'up trend' detection could be defined instead of using a ROC filter on the relevant sector, and yes you can define a better entry condition.

Thanks for the inspiration.

Cheers
Trav

 

[Skate]

The there is no in-between had me think about RSI and Stochastic's again and the ability to set levels for overbought / oversold signals and obviously anywhere in between.....

@Trav. I'm glad you mentioned RSI & Stochastic a momentum oscillator giving me the opportunity to expand further about momentum & momentum oscillators. These oscillators are leading indicators that can "signal a possible trend change" that is yet to start helping to gauge the strength of a trend. The adjustable time period means these indicators are user friendly & can be finely tuned.

How does a momentum indicator work?
A Momentum Oscillator moves between an upper & lower bands. When a "Momentum Oscillator" reaches the upper & lower boundaries "indicate" that the stock may be overbought or oversold.

The “HappyCat Strategy” uses a different way to calculate momentum
The HappyCat Strategy calculates momentum over a period from bar to bar. The mathematical calculations use two values, such as two points that measure the rate & speed of the average change in price. The entry signal uses a smoothing effect "using the average" of the price bar. Using the average of the price makes it easier to pick a confirmed trend. Using this method of entry allows multiple entries positions rather than just one. The exit signal is the reverse of the entry condition. It's not so much when you buy within the trend but where you buy eliminates most of the false signals.

The adjustable time period of the “HappyCat Strategy”
Optimising the lookback period helps gauge the strength & momentum of a trend but also typically signal if a market is overbought or oversold, which can point to a stalling or reversal of the trend. The hardest part trading the “HappyCat Strategy” is getting past the notion that there is a right & wrong time to enter a trade. There isn't a "right & wrong" time as the price is random. System trading relies heavily on trade management ensuring that you are "never right or wrong". With the “HappyCat Strategy” parameters & settings there is only "get in & get out" alleviating any grey areas of interpretation.

Skate.

 

[Skate]

Holy cow, that would be very hard to stomach. Hard to believe it pulls that much profit when a trade like that can be holding up the portfolio. Very strange business at times this trading....

There is a premium to be paid for added security
The "HappyCat Strategy", (version 1) the "original" has the protection level I'm comfortable trading. With trading, it's always a compromise between results & stomachable results. Each layer of protection ultimately has a cost to be paid, meaning it's a tradeoff that varies in degrees between traders as risk tolerances vary. (Version 2) of the "HappyCat Strategy" peels back the layers of protection & the results are pleasing but sometimes it can be a horror show to watch. The "HappyCat Strategy" (version 3) is really (version 2) with one layer of protection added.

The simplicity of the extra layer of protection in (version 3)
The added layer of protection is quite simple. If the position fails to keep going "up" after a slight pullback - "exit" it's that simple.

Examples of the three versions of the "HappyCat Strategy" is below (NEA)
This is the original version, the version I would be happy to trade if "paper trading" mimics my backtest results.






The "HappyCat Strategy" (version 2) has no protection levels (the strategy buys & exits on momentum)





The "HappyCat Strategy" (version 3) has "one level of protection" added (If the position fails to keep going "up" after a slight pullback - "exit" )




In the next post
I'll post a comparison of the last financial year backtest results for the three versions (1st July 2019 to 30th June 2020) to visually gauge the "price" when you add layers of protection.

Skate.

 

[Skate]

The "HappyCat Strategy" comparisons between the versions



Skate.

 

[Skate]

Hi everyone, been lurking here for a few weeks - reading the phenomenal amount of material. I have not traded shares directly before, but have always been drawn to it. My plan is to research and learn, specifically systematic trading approaches. I have a background in software development and like the methodolical approach, so I think it will work well for me, but we'll see. I'd like to then papertrade one or two methods that appeal to my personality and stage in life / risk appetite. Following that, I will trade the system that has ticked as many boxes as possible. Cheers

Members are joining the "Aussie Stock Forums" to learn how to trade & trade profitably
@TraderJimmy is typical of new members seeking the help of others without fully releasing the effort required to learn. It’s frustrating & time-consuming trying to figuring out what trading is all about. Posting information that I found relevant hopefully shortens their learning curve. Learning & bouncing ideas off each other is the very essence of the "Dump it here" thread.

The "Dump it here thread" keeps growing
3593 posts makes it near impossible to set the time aside to read the complete thread & absorb its content.

Reading the entire "Dump it here" thread
As Sweet Brown famously said "ain't nobody got time for that" & admittedly it's a big ask.

I would like to talk about Kimberly Wilkins
Who? Kimberly Wilkins is better known to the world as Sweet Brown. Kimberly was in the right place at the right time to become an internet sensation. Her declaration of, "Ain't nobody got time for that!" went viral, garnering millions and millions of views on YouTube that lifted her lifestyle.

The original


Better version remix


Skate.

 

[Skate]

As mentioned over in the Dump It Here thread I thought about the use of momentum So after reading this I decided to run a few tests and found some improvement in my back test results - (at) what level you want to enter the market with (typical Oversold = 20 and Overbought = 80) a level that confirms an uptrend.

Stimulating ideas
@Trav. is posting some stimulating ideas at the moment. His last thought-provoking idea is "at what level do you enter a trend" using the typical Oversold & Overbought levels. @peter2 has referenced "Swing Trading" in over 180 posts talking about "trend trading", "swing trading" & "momentum trading" all in the same breath so it's only fitting that I try to combine them into a tradable strategy - that's the initial brief I gave myself.

My trading methodology is very similar to that documented by @Skate . We need prices to move higher on all time frames. You may call it trend, swing or momentum trading.

I get the feeling Peter enjoys swing trading taking quick profits
Taking quick profits the gains might be smaller, but when he is doing it consistently I'm sure trading this way compounds into excellent system returns. Swing Trading & taking quick profits is totally foreign to how I trade.

The trades based on the daily charts will be much shorter in duration as they're designed to capture smaller price swings.

I've got an idea
What if we incorporate a strategy that can take advantage of both worlds using Trav's Oversold & Overbought idea & Peter's Swing Trading methodology. By combining these two ideas using a defined level between "Oversold & Overbought" to enter the trend. Sharpen the parameters so the "Swing Trades" can be held longer.

Long posts can get boring to read
I'll post my latest research over a few post to try & hold your interest with a more detailed explanation of my current thinking.

Skate.

 

[Skate]

How do you identify the beginning & end of a price swing
When markets exhibit extreme volatility as with the recent COVID-19 flash crash, traders move away from longer-term trading methods because of their devastating drawdowns & start to gravitate towards more nimble strategies trying to catch shorter-term price moves. Among the many short-term trading methods, swing trading offers a good middle ground that makes trading more manageable.

Traders may use a variety of methods to identify the beginnings & endings of a price swing
These methods usually include patterns formed by support/resistance, (oversold & overbought) or a multitude of momentum indicators. The approach I want to talk about the humble "relative strength index" (RSI), a momentum indicator.

The (RSI) identifies potentially overbought & oversold levels
However, an (RSI) with values that Trav quoted (80 & 20) will have to be discarded if we are to take advantage of entering a confirmed trend sooner rather than later. The (RSI) provides signals about bullish & bearish price, something I can work with.

The (RSI) is displayed as an oscillator
An oscillator is just a "line graph" that moves between two extremes from "0 to 100". My research confirms using the halfway point of the oscillator (50) to identify short-term price swing has the ability to catch fairly large trends before reaching the overbought level.

Using a short lookback period to increase the sensitivity of the indicator
Combining a short lookback period with an (RSI) & entering a position when the close is greater than the (50) level allow you to remain in an uptrend for a longer period. Using a level higher than (50) is counterproductive with a short lookback period.

Examples to follow
I'll post up some backtest results as examples of how the humble (RSI) can be utilised to our advantage.

Skate.

 

[Skate]

These are the Backtest results over three different time periods
1. 2019 Calendar Year
2. 2019- 2020 Financial Year
3. 2020 Year to date



Skate.

 

[Skate]

Swing trading can still deliver larger gains
A stock with enough initial strength can be held for a bigger gain giving the position room to run. Taking profits when they come can string together a number of gains back-to-back & your portfolio can turn a great result. The usual swing trading holding period is short but stronger gains mean longer hold periods. You'll often see stocks go up just after you sell & kick yourself for selling too early. Technical analysis is well suited to swing trading as it allows you to focus on buying positions with good momentum.

Chart & Swing Indicator displayed


Skate.

 

[Skate]

I always appreciate when others share
I'm like most members & get my share of private messages sharing & bouncing ideas around. Being a solidarity keyboard trader you tend to wonder what others are trading. I was also wondering if there is a trading benchmark out there that you can use to compare your trading results against? - Sometimes the simplest word or sentence can be the catalysis to make you re-evaluate your trading style or methodology always looking for improvements. Learning is always evolving.

I have a few strategies under evaluation at the moment (That I would like to share)
I'd like to share some backtest captures of 3 strategies (with 3 different time periods) that are under evaluation at the moment.

The three strategies
1. The KingFisher Strategy
2. The PocketPivot Strategy
3. The Swing Trading Strategy

# Backtest period
1st January 2019 to 30th December 2019 (2019 Calendar Year)






# Backtest period
1st July 2019 to 30th June 2020 (2019/2020 Financial Year)






# Backtest period (YTD)
1st July 2020 to 6th August 2019 (2020/2021 Calendar Year-to-Date)




Skate.

 

[Skate]

I have a fair few strategies that are completed, tradable & parked at the moment (There are three that I would like to share)
I'd like to share a few backtest captures of 3 strategies (with 3 different time periods) that have gone full circle of evaluation & are parked for future use.

The three strategies
1. The Bollinger Band Strategy
2. The Darvas Strategy
3. The Donchian Strategy

# Backtest period
1st January 2019 to 30th December 2019 (2019 Calendar Year)






# Backtest period
1st July 2019 to 30th June 2020 (2019/2020 Financial Year)






# Backtest period (YTD)
1st July 2020 to 6th August 2019 (2020/2021 Calendar Year-to-Date)




Skate.