Nick Radge – Daily Journal

[Nick Radge]

ASX MR has been frustrating over last 12 months…has happened before though…and almost identical time of year

[Nick Radge]

Seriously considering ceasing my ASX MR system.

This month has been a disaster. Of the 30 ‘possible’ trades I have only caught 20. Those 10 missed trades were equal to a 4% return, so this month I am +0.4% instead of +4.4%.

[ScottMcNab]

Due to not getting filled at low of day ?

[Nick Radge]

Yep – all lows of the day.

[JulianCohen]

This was also my concern although I haven’t had it occur in the last few weeks. I’m not sure what to do about it either.

If the brokerage drain on the system wasn’t so high then it might make a difference. That was one of my reasons for stopping the MOC system in the ASX. However there’s not a lot can be done about that.

[SaidBitar]

recently i modified the code for testing purpose and I changed the condition that says
L <= BuyPrice to L < BuyPrice - 0.05 so i am making sure that there is at least 5 cents between the buyprice and the low of the day i think the trade frequency dropped by 20+% and the CAR dropped by 10%

[JulianCohen]

Said Bitar wrote:

recently i modified the code for testing purpose and I changed the condition that says
L <= BuyPrice to L < BuyPrice - 0.05 so i am making sure that there is at least 5 cents between the buyprice and the low of the day i think the trade frequency dropped by 20+% and the CAR dropped by 10%

I just tried the same thing.

I made the line in the Limit Buy Section

BuyLim = (BuyLimVal * TickLo) – 0.05;

It improved the MDD quite a lot, but halved the number of trades and nearly halved the CAGR from 24% to 13%.

[JulianCohen]

I just ran a test using 1 through to 5 ticks below the entry to see how the results are affected:

0 tics = 24% CAGR -18% MDD 1857 Trades
1 tics = 24% CAGR -16% MDD 1441 Trades
2 tics = 18.6% CAGR -16% MDD 1175 Trades
3 tics = 16.2% CAGR -10% MDD 987 Trades
4 tics = 15.5% CAGR -8.5% MDD 873 Trades
5 tics = 13% CAGR -8.6% MDD 746 Trades

I think the only thing this proves is that I have too much time on my hands and I need a hobby. However it does show the rate at which we can expect the results to fall away depending upon how we get filled.

[ScottMcNab]

Nick Radge wrote:

Yep – all lows of the day.

been thinking of possible options…may all be rubbish or not apply but am looking to do this myself for aussie MOC:

1. trade XJO rather than XAO…my xao system uses 20 positions at 10% but can get XJO to work if go to 20 positions at 20%..bit of extra volatility but liquidity may make it worth it

2. if trading odd lots (which IB tells me means I am not necessarily entitled to fill as on separate order book) then add rounding calculation to code to eliminate odd lots

3. raise buylimit price by a tick

[Nick Radge]

Quote:

3. raise buylimit price by a tick

Not sure this will help. My ‘guess’ is that my volume is being ‘lent’ against in the smaller stocks.

[Nick Radge]

Here’s how mine shapes up.

When I compare real time results since start date, the 0.05 is about right on…a little surprised it’s that much.

[TrentRothall]

When you say ‘Lent on” Nick, do you mean people can see you order sitting there and that has a effect?

I have been thinking this maybe a problem, couldn’t you use “Stop Limit” orders, i think I holds the orders then sends them once the price is hit. Could this help?

[Nick Radge]

Yes.

You can’t use a stop limit – different order type entirely.

You could use Limit if Touched (LIT) which means a LMT order is held on the server and only released when it trades at your price. Could miss gap opens though.

https://www.interactivebrokers.com.au/en/index.php?f=592

[TrentRothall]

Oh yeah, that’s what i meant. Can the new API handle them?

[Nick Radge]

I’m just testing it out.

[Author]

Posts