Nick Radge's Weekend Trend Trader

[RowanC]

Sorry getting this old thread alive, but there's something that's needs to be spoken out. About 1 year ago he updated his performance page, removed the actual performance numbers and instead it's only some nice looking charts.

E.g. The US momentum strategy chart, one barely sees SP500 down there, the momentum strategy looks awesome. However this year SP500 is at new highs wihle the momentum strategy it's not. The actual reality is not even close to what that chart is saying.

I queried Nick in the past about this unfortunate change, he said he changed it because the actual numbers were confusing. I fail to see how the actual numbers were confusing and this performance chart not...

Anyway in the past years seems that the chartist is investing a lot in marketing. Not that it's a bad thing, but sometimes makes one wonder.

View attachment 67684

I tend to agree with you here. Charts and graphs are a good way to mask the unflattering truths that lurk behind many trading strategies. It would be nice to get a clearer picture of things by seeing actual figures like the Sharpe or Max DD.

That said I think if you asked Nick he would be happy to disclose any info as he does seem like a legit guy.

 

[skc]

I tend to agree with you here. Charts and graphs are a good way to mask the unflattering truths that lurk behind many trading strategies. It would be nice to get a clearer picture of things by seeing actual figures like the Sharpe or Max DD.

That said I think if you asked Nick he would be happy to disclose any info as he does seem like a legit guy.

Agree.

When you show a long term chart (this one >10 years) you are illustrating the long term performance. If you want shorter term month by month performance then a monthly performance table against the benchmark is a much clearer way to show that.

With long term charts, once you have a few good periods of outperformance, any underperformance in subsequent periods are difficult to see. The system could be treading water for the last 5 years if it beat the index by 50% in year one, and the chart would still show a long term outperformance.

Trading someone else's system is hard for this reason alone...

 

[Roller_1]

I think you will find it so people with less experience can get a feel for what to expect.

Anyway, is anyone trading The WTT?

 

[soso]

That said I think if you asked Nick he would be happy to disclose any info as he does seem like a legit guy.

I don't doubt Nick's integrity, I'm pretty sure he will provide any available performance data if a (potential) client asks for it. My main complaint was the usefulness of the data that is readily available. Adding actual numbers in the performance page would be helpful imo, e.g. I may constantly check the performance page every few months to see how it's doing, why send an email each time for the actual numbers?

Lastly fwiw, he sent me an email explaining the reason for current performance page layout - "We use this data because it is constantly requested by current and prospective clients".

 

[soso]

Anyway, is anyone trading The WTT?

davetapson was trading it, but last report from him was about 1.5 years ago.

 

[lftrader]

There are much bbetter traders out there. You must look for yourself. No one will talk about a service that makes money. All the talk is about losers using curvefitted systems.

 

[Porper]

There are much bbetter traders out there. You must look for yourself. No one will talk about a service that makes money. All the talk is about losers using curvefitted systems.

Happy for you to point us in the direction of a great trader doing no advertising. Also, can you explain the "curve fitting" remark. Proof would be good for you too...on all counts. Waiting with baited breath

 

[CanOz]

Trading any system is difficult, as trading itself is difficult. At least with Nick's systems they're very robust. They'll stand the test of time. You only need to to understand HOW they work to understand how robust they are.

The only question is can the trader stand the test of time, regardless of who's system it is.

I plan on trading a portfolio of Nick's systems and my own on my SMSF. Nick's systems add the low volatility element i need for when most of my systems perform poorly. Who knows what in store for the market in the years to come, but i'm pretty confident Nick's systems will outlast my own. Maybe the world goes as Howard says and our systems are our least worries....

 

[2Finn5]

I've been trading WTT for over two years now. Well, it's my code based on the ebook, with one intentional difference and a universe of ASX stocks I spent some time coming up with.

Performance since the start of the year (including brokerage but not including dividends):

 

[CanOz]

I've been trading WTT for over two years now. Well, it's my code based on the ebook, with one intentional difference and a universe of ASX stocks I spent some time coming up with.

Performance since the start of the year (including brokerage but not including dividends):

View attachment 67698

Interesting Finn, how exactly does the WTT work? I have run tests on the 20% flipper as a weekly system and it gets a win rate of 60%. Pretty encouraging. I'm planning on running it as a weekly and the MR system which is a daily.

 

[2Finn5]

WTT is a weekly system, buying on open on the first trading day of the new week if a 20 week high is made on the previous week's close. Add an index filter which stops entries during downtrends. Stops are set to risk no more than 2% of total equity to start with and tightened if the index filter shows a downtrend. Enter twenty positions as they come up and hold them until they indicate a sell.

I've added some position scoring to the system to try to pick better candidates if the number of buy signals exceeds the amount of cash on hand. It probably doesn't make much difference overall.

Nick also posted (when his forums were public) stating that he avoids any stocks under takeover even if they trigger a buy signal, and to sell any held stocks two weeks after a takeover offer being made public. I've been doing this and it avoided a few stocks that just stagnated after a buy signal, so running a system backtest since I started trading gives worse results than I have received by trading the signals but ignoring those stocks under takeover. So it should mean that the historical backtests I ran were also a bit pessimistic.

I'm curious you managed to get the 20% flipper to work - I coded that one up too, and couldn't get decent results from the ASX no matter what I did with it. Well, the backtests were profitable, but not by enough to justify the time and effort and paperwork to actually trade it.

 

[CanOz]

WTT is a weekly system, buying on open on the first trading day of the new week if a 20 week high is made on the previous week's close. Add an index filter which stops entries during downtrends. Stops are set to risk no more than 2% of total equity to start with and tightened if the index filter shows a downtrend. Enter twenty positions as they come up and hold them until they indicate a sell.

I've added some position scoring to the system to try to pick better candidates if the number of buy signals exceeds the amount of cash on hand. It probably doesn't make much difference overall.

Nick also posted (when his forums were public) stating that he avoids any stocks under takeover even if they trigger a buy signal, and to sell any held stocks two weeks after a takeover offer being made public. I've been doing this and it avoided a few stocks that just stagnated after a buy signal, so running a system backtest since I started trading gives worse results than I have received by trading the signals but ignoring those stocks under takeover. So it should mean that the historical backtests I ran were also a bit pessimistic.

I'm curious you managed to get the 20% flipper to work - I coded that one up too, and couldn't get decent results from the ASX no matter what I did with it. Well, the backtests were profitable, but not by enough to justify the time and effort and paperwork to actually trade it.

I think the big difference is i use it on the Russel 3000. Because the principle is to catch multi baggers at 20% you get allot more good ones in the US. The bigger the universe the better i reckon.

 

[Lone Wolf]

I'm planning on running it as a weekly and the MR system which is a daily.

Is the MR system the same as what is used for the US high frequency strategy? The US HFT strat potentially puts out a large number of orders. You need a way to manage how many trades actually get taken, ie. only fill the first x number of trades, kill the rest.

 

[Roller_1]

Is the MR system the same as what is used for the US high frequency strategy? The US HFT strat potentially puts out a large number of orders. You need a way to manage how many trades actually get taken, ie. only fill the first x number of trades, kill the rest.

You will need a API to trade it as designed, which he has. It also allows bulk upload of trades into IB's TWS too. It would take <15mins a day to run the system

 

[Lone Wolf]

You will need a API to trade it as designed, which he has. It also allows bulk upload of trades into IB's TWS too. It would take <15mins a day to run the system

Yes, if you don't have the coding ability to make an API, The Chartist has one for sale.

You really do need to use IB too, as brokerage would kill you on any Australian broker. I'm trying to picture how a HFT strat would be affected by the lack of marign with IB and having to wait T+3 for the sale funds to clear... Wrong thread for this really, thanks for your input though.

 

[Trendnomics]

I've been trading WTT for over two years now. Well, it's my code based on the ebook, with one intentional difference and a universe of ASX stocks I spent some time coming up with.

Performance since the start of the year (including brokerage but not including dividends):

View attachment 67698

Interesting thanks for posting!

What is your annual return in %? Would be good for comparison purposes (I'm also a systematic trend-follower).

 

[Lone Wolf]

I've been trading WTT for over two years now. Well, it's my code based on the ebook, with one intentional difference and a universe of ASX stocks I spent some time coming up with.

Performance since the start of the year (including brokerage but not including dividends):

View attachment 67698

The profit/loss per trade seems very small for the first couple months then suddenly changes. I don't want to pry into exact details, but did you modify your position size?

 

[CanOz]

Here is an example of how running two portfolios can smooth out the returns. I only went back three years in this example. Its best to only ever test a minimum of 2 years as i believe EOD systems need two years of results to be fair to the systems capability.

The first one is the Flipper, followed by the combined results and the last one is the MR system on its own.

 

[Roller_1]

Here is an example of how running two portfolios can smooth out the returns. I only went back three years in this example. Its best to only ever test a minimum of 2 years as i believe EOD systems need two years of results to be fair to the systems capability.

The first one is the Flipper, followed by the combined results and the last one is the MR system on its own.

That's a good way to see the results of combining portfolios Can, looks good.

 

[Trendnomics]

Here is an example of how running two portfolios can smooth out the returns. I only went back three years in this example. Its best to only ever test a minimum of 2 years as i believe EOD systems need two years of results to be fair to the systems capability.

The first one is the Flipper, followed by the combined results and the last one is the MR system on its own.

I'm a little bit confused.

For the impressive combined returns, you are simply adding the Flipper and MR annual returns together. Generally, percentages are not to be added together (normally multiples of each other).

Do both systems run on the same allowed capital? Or are there two separate pots of capital? If separate (exactly same sized pots), you need to half each of the individual returns and then multiply by each other - for the overall return on funds.

This would reduce your largest combined annual return (69.30%), to 36.12% - still impressive.