Tech/a Techtrader - my questions

Skate said:

The ONE tweak that I've found that made an improvement was to change the sell criteria - By changing the sell criteria to:

Sell= C < MA(C,50);

skate

Click to expand...

The critical information you left out in your assumption is the test period. The best part of this trend trader was catching one of the strongest bull runs in ASX history. Simple as that.

Change the test period = change the result.

Wysiwyg said:

The critical information you left out in your assumption is the test period. The best part of this trend trader was catching one of the strongest bull runs in ASX history. Simple as that.

Change the test period = change the result.

Click to expand...

Yeh that's what it was designed for a bull market.

Well and truly out performed the market.
Strangely many people lose money in bull markets.

Actually its pretty consistent.

tech/a said:

Much much more here and coming up in the future for those interested.

http://lightning.he.net/cgi-bin/suid/~reefcap/ultimatebb.cgi?ubb=get_topic;f=74;t=000040

You'll have to register if not a member.

Click to expand...

I realise this is very old post now but does this link supposed to go somewhere different now or is there another link. Just a query.

Woosha said:

I realise this is very old post now but does this link supposed to go somewhere different now or is there another link. Just a query.

Click to expand...

Wow that is very old.
No that's lost to cyber space.

I have seen this claimed about the techtrader thread multiple times by tech, but I have never been able to verify.

From my understanding, it was $30k plus $70k margin loan, so the return from trades is actually more like 260% return over 7 years (~15% per annum, after margin costs I assume, otherwise not much better than market return) than the 1200% return that "turning $30k into $360k" seems to imply.

Still an impressive return of course, but should be considered in light of the overall market regime for those 7 years as well.

Click to expand...

There is quote a bit of info in this thread but specifically.

The universe of stocks traded was the current (Then) BT Margin List.
Stocks under $10---My thinking was that if it was in the list then BT
would have done the Fundamentals to allow its inclusion (Amateur thinking).

I was Trading a BT margin account.

The position sizing was 10% of Funds available with a stop of 10% of the purchase price.

Radge has built into me that Margin should be used only when a stocks full parcel
cannot be funded with the position sizing being adopted by the method.

This happened very rarely (Due to the position sizing and stop model) but it did happen.
The method and any other method for that matter wasn't/shouldn't use the full margin
available. As you are aware this alters the risk to initial capital dramatically.

The system was very pedestrian for 2 yrs as the average winning trade was held for just under a year
so initially new trades were being funded from stopped out trades and those that swung away from
profit early.
But it really started to crank up as longer term trades triggered exits.
The increased funds were then used in the next trade/s

The results we achieved on Radges site (He's taken it off when he did a new revamp only a year ago)
were at the higher end of the Monte Carlo analysis results we got when running tests. (Dumb luck).

My own portfolio which was a little different as I started it a little after we got started (On The Chartist) had slightly lower results.

I posted Entries and Exits and another Member Darryl ran a spread sheet with the portfolio results published every week.

This was in Radges words one of the most scrutinized methods made available to the public.
There were 1000s of posts and questions over the years. I don't know if Nick kept all the stuff from his server but Ill ask him.

If you know Radge he wouldn't put anything on his site or publish anything in one of his books if it didn't stand up to the most stringent scrutiny---HIS---and 100s of others over the years.
He IS a professional educator and Trader/Investor.

If you want all the info available in print Un Holy Grails Pages 103-109

I didn't design and test/trade it to be a big deal---I wanted to see if a builder (Read amateur) could design and trade profitably a trading system.

To be honest it far surpassed any expectation I had.(Which wasn't much!)

I know of 4 people who still use it for their Super---as of 18 mths ago.

tech/a said:

The position sizing was 10% of Funds available with a stop of 10% of the purchase price.

Click to expand...

So was the "funds available" for the initial 10 positions $30k or $100k?

If you turn $30k into $360k does it really matter if margin was used or not? It's still $360k....

Roller_1 said:

If you turn $30k into $360k does it really matter if margin was used or not? It's still $360k....

Click to expand...

I think it matters in the context of the previous discussion on returns...of course, no one would dispute that.

Roller_1 said:

If you turn $30k into $360k does it really matter if margin was used or not? It's still $360k....

Click to expand...

From the perspective of understanding the returns from trades taken, it obviously matters, otherwise the claim should be "I used techtrader to generate a gross profit of $260k from a $100k capital base over 7 years". See the difference?

Thought this was clear.

The first trades were $3000 each.
As the account was a margin account there were 2 X Initial funds available.

They were not used unless there was an instance where the 1% stop (10% of the purchase price) on 1 of 10 trades and the Purchase price then the margin was used.

This hardly ever happened as the stop was so wide it didn't cause a problem with the trade.
All initial trades were under $10

Taking a $10 trade (the worst case) Available capital $3000
Shares we could purchase 3000 Capital required $30,000
$5---same.
Most trades were well under that $1/$2/3 and less the universe didn't have a lot under $10
About 70 I think.

There were many many times when we were only trading 5 or 6 trades as we waited for other trades to qualify.
Often there were not enough funds and we had to wait.

But as Profit grew it was often not needed and by 3rd year was self supporting it had enough of its own profit to finance what Margin did from time to time.
Darryl always included dividends and costs.(Transaction and interest---Interest was very minor).

Anyway it was what it was---The full $100,000 was NEVER used at the one time.
Crap on it as much as you like many have done so in the past.

You guys are bringing up % return.
When I bought a house and put nothing down and used 100% banks money and it doubled in 3 yrs then I sold it
what's my return on my initial capital???

Frankly to start with $30K and end up with over $300k I was happy as a ---Duck.

I think it proved the question I was looking for an answer to for me anyway.

Radge ran extensive testing of his own all in the book.
He asked if he could include it I didn't and couldn't twist his arm, I initially refused---ask him!

Roller_1 said:

If you turn $30k into $360k does it really matter if margin was used or not? It's still $360k....

Click to expand...

Exactly was and is my take as well.

My own money and the use of others at times meant I had a great return on capital risked.

So What??

I do it all the time.
In housing Business and trading the DAX.
What I shouldn't be doing that??

what's my return on my initial capital???

Click to expand...

That's irrelevant, my goal was simply to try and clarify the return of the trades the system took in light of comments you had made about the margin loan.

It's the difference between understanding the return of house prices increasing threefold, and what is essentially an "infinity" return on capital if you put none down.

Hopefully you understand the desire for others to make sure they grasp the source of returns.

tech/a said:

What I shouldn't be doing that??

Click to expand...

You can do whatever you want. This defensiveness in light of any sort of questioning is toxic.

I just wanted clarification on the thing I asked, which you provided.

Thankyou for clarifying.

I'm sure someone like minwa is using leverage in some form to gain his returns, but his return on initial capital is still awesome.

If someone invests 10k into a CFD account and has access to 100k of tradeable funds and makes 5k profit it would be counted as a 50% return in the end wouldn't it? not 5%

You can do whatever you want. This defensiveness in light of any sort of questioning is toxic.

I just wanted clarification on the thing I asked, which you provided.

Thankyou for clarifying.

Click to expand...

Radge has extensive figures and tables on returns in his book---un leveraged.
I think around 20-25% in there somewhere.

Thought that was OK and the power of leverage and Compounding is clear from return on capital invested.

Your questioning is designed to discredit.

From my understanding, it was $30k plus $70k margin loan, so the return from trades is actually more like 260% return over 7 years (~15% per annum, after margin costs I assume, otherwise not much better than market return) than the 1200% return that "turning $30k into $360k" seems to imply.

Click to expand...

The only people I know that can turn things on others as well as this are politicians and My wife!

Perhaps you should be less confrontational!
I and others may then become less toxic

Roller_1 said:

I'm sure someone like minwa is using leverage in some form to gain his returns, but his return on initial capital is still awesome.

If someone invests 10k into a CFD account and has access to 100k of tradeable funds and makes 5k profit it would be counted as a 50% return in the end wouldn't it? not 5%

Click to expand...

Yeah, if you buy a $1000 put option on SPY and tomorrow you wake up and the put is now worth $1500 then it's a 50% return but it doesn't entitle you to say "I can make 50% returns in a day".

Nobody is disputing that return on capital is return on capital. The issue raised was only to clarify the source of those returns, which has been done. If you don't care about the source, that is fine, but some people do.

tech/a said:

Your questioning is designed to discredit.

Click to expand...

Dude, you are actually crazy if you think the motivation for people to post things is to discredit you. Perhaps try to think about other reasons why people might post things.

Perhaps you should be less confrontational!
I and others may then become less toxic

Click to expand...

How in the hell is making a comment that begins with the words "from my understanding" and includes the statement "still an impressive return", confrontational?

Someone messaged me in private to let me know this discussion has already occurred in another thread, and after checking I can see tech saying he bought $90k of stock...which implies the full margin was used (my mistake it was $60k not $70k).

The starting equity is also shown as $90k, not $30k.

https://www.aussiestockforums.com/forums/showthread.php?t=26473&p=761907&viewfull=1#post761907

Not really sure how the charts presented in the above thread make sense if the initial position size was $3000.

sinner said:

Yeah, if you buy a $1000 put option on SPY and tomorrow you wake up and the put is now worth $1500 then it's a 50% return but it doesn't entitle you to say "I can make 50% returns in a day".Nobody is disputing that return on capital is return on capital. The issue raised was only to clarify the source of those returns, which has been done. If you don't care about the source, that is fine, but some people do.

Click to expand...

But you did that day---you can say that that day.
Just like I did in 7 yrs.----I can say that over that 7 yr period.

I have never said or implied that I have done that every seven year period or indeed in any future seven year period. Only in THAT 7 yrs period.

My apologies that it doesn't fit with you.
It is what it is/was

sinner said:

Dude, you are actually crazy if you think the motivation for people to post things is to discredit you. Perhaps try to think about other reasons why people might post things.



How in the hell is making a comment that begins with the words "from my understanding" and includes the statement "still an impressive return", confrontational?

Click to expand...

The bit in the middle you don't quote
In particular the Blue bit.

so the return from trades is actually more like 260% return over 7 years (~15% per annum, after margin costs I assume, otherwise not much better than market return) than the 1200% return that "turning $30k into $360k" seems to imply.

Click to expand...

tech/a said:

// here we define our buy conditions
cond1=Cross(H,Ref(HHV(H,10),-1)); //todays high crosses last highest high over the last 10 periods
cond2=H > EMA(C,40); // todays high is greater than the 40 day Exp MA of closes
cond3=HHVBars(H,70) == 0; // todays high is the highest for 70 periods
cond4=EMA(V*C,21) > 500000; // ensure at least $500k of money flow
cond5=C < 10.00; // only trading in stocks less than $10
cond6=C > O; // todays close higher than open

// initial stop aim to never lose more than 10%
ApplyStop( stopTypeLoss, stopModePercent, amount=10 );

//our exit conditions if not stopped out
Sellsig= Cross(Ref(EMA(L,180),-1),C); // close crosses below yesterdays EMA of the low

Click to expand...

Morning everyone, longtime no hear (from me anyway)!

I had a renewed interest in a system I could use for super and thought I would backtest TT. However, I think I would have to modify slightly.

Data: using yahoo finance. This gives you adjusted OHLC which accounts for all splits and dividends. However, if I use this, it is a little hard to compare to XAO unless I can easily split out the dividend yield.
For this reason I think I will use the unadjusted values and see if I can adjust only for splits. Dividends would then be extra return

Position size.
I am a little more conservative. I would feel more comfortable with position sizes of $15k now. I intend to work out what that translates to at the beggining of the data set and have a year-by-year escallation that equates to $15k now.

Stock universe.
If I could get the BT margin list each year for the last 30years I could use that. It might present data problems for me though. For now I intend to rely on the EMA of the volume*close to select on liquidity instead.

The buy conditions.
cond1 - i think this will always be true when cond3 is true. cond1 ignored.
cond4 - (also see Stock universe). In two minds. For liquidity checking I would prefer to use the MA rather than EMA, however EMA would allow selection of stocks with smaller traded $ which have an uptick in the traded $. Will leave as EMA.
cond5: if I use data that is backadjusted for splits and dividends, this interferes with this condition. I do have unadjusted data. Will try with and without this condition.

EMA generally - assume smoothing = 2 as per https://www.investopedia.com/terms/e/ema.asp

The sell conditions.
I will change the stop slightly - if any of o/h/l/c <= 0.9 * buy price, then sell next open. This is realistically the best I could acheive in real life. I prefer not to have automated stop losses via the broker system - as you wouldn't know what price you got for the sale if it dumps to market, and it would be hard to backtest without knowing sell price. Assuming the liquidity is selected suitably, you have better chance buying/selling on the open/close without influencing the market.

Monte carlo analysis: will do this.

Any thoughts or comments?


I'll send analysis when I get it all working.

markrmau said:

Morning everyone, longtime no hear (from me anyway)!

I had a renewed interest in a system I could use for super and thought I would backtest TT. However, I think I would have to modify slightly.

Data: using yahoo finance. This gives you adjusted OHLC which accounts for all splits and dividends. However, if I use this, it is a little hard to compare to XAO unless I can easily split out the dividend yield.
For this reason I think I will use the unadjusted values and see if I can adjust only for splits. Dividends would then be extra return

Position size.
I am a little more conservative. I would feel more comfortable with position sizes of $15k now. I intend to work out what that translates to at the beggining of the data set and have a year-by-year escallation that equates to $15k now.

Stock universe.
If I could get the BT margin list each year for the last 30years I could use that. It might present data problems for me though. For now I intend to rely on the EMA of the volume*close to select on liquidity instead.

The buy conditions.
cond1 - i think this will always be true when cond3 is true. cond1 ignored.
cond4 - (also see Stock universe). In two minds. For liquidity checking I would prefer to use the MA rather than EMA, however EMA would allow selection of stocks with smaller traded $ which have an uptick in the traded $. Will leave as EMA.
cond5: if I use data that is backadjusted for splits and dividends, this interferes with this condition. I do have unadjusted data. Will try with and without this condition.

EMA generally - assume smoothing = 2 as per https://www.investopedia.com/terms/e/ema.asp

The sell conditions.
I will change the stop slightly - if any of o/h/l/c <= 0.9 * buy price, then sell next open. This is realistically the best I could acheive in real life. I prefer not to have automated stop losses via the broker system - as you wouldn't know what price you got for the sale if it dumps to market, and it would be hard to backtest without knowing sell price. Assuming the liquidity is selected suitably, you have better chance buying/selling on the open/close without influencing the market.

Monte carlo analysis: will do this.

Any thoughts or comments?


I'll send analysis when I get it all working.

Click to expand...

Hi @markrmau,

markrmau said:

Morning everyone, longtime no hear (from me anyway)!

I had a renewed interest in a system I could use for super and thought I would backtest TT. However, I think I would have to modify slightly.

Data: using yahoo finance. This gives you adjusted OHLC which accounts for all splits and dividends. However, if I use this, it is a little hard to compare to XAO unless I can easily split out the dividend yield.
For this reason I think I will use the unadjusted values and see if I can adjust only for splits. Dividends would then be extra return

Position size.
I am a little more conservative. I would feel more comfortable with position sizes of $15k now. I intend to work out what that translates to at the beggining of the data set and have a year-by-year escallation that equates to $15k now.

Stock universe.
If I could get the BT margin list each year for the last 30years I could use that. It might present data problems for me though. For now I intend to rely on the EMA of the volume*close to select on liquidity instead.

The buy conditions.
cond1 - i think this will always be true when cond3 is true. cond1 ignored.
cond4 - (also see Stock universe). In two minds. For liquidity checking I would prefer to use the MA rather than EMA, however EMA would allow selection of stocks with smaller traded $ which have an uptick in the traded $. Will leave as EMA.
cond5: if I use data that is backadjusted for splits and dividends, this interferes with this condition. I do have unadjusted data. Will try with and without this condition.

EMA generally - assume smoothing = 2 as per https://www.investopedia.com/terms/e/ema.asp

The sell conditions.
I will change the stop slightly - if any of o/h/l/c <= 0.9 * buy price, then sell next open. This is realistically the best I could acheive in real life. I prefer not to have automated stop losses via the broker system - as you wouldn't know what price you got for the sale if it dumps to market, and it would be hard to backtest without knowing sell price. Assuming the liquidity is selected suitably, you have better chance buying/selling on the open/close without influencing the market.

Monte carlo analysis: will do this.

Any thoughts or comments?


I'll send analysis when I get it all working.

Click to expand...

Hi @markrmau,

cond1 - i think this will always be true when cond3 is true. cond1 ignored.

Rather than ignoring this condition immediately, why not leave it in initially and run a test. You could then remove the condition and run another test so that you can verify the effect of removing that condition.

Cheers, Rob