Scott’s Journal

[JulianCohen]

I can see it happening in the cheaper universe as the volume doesn’t always allow getting fills at open. I have noticed missed trades where it would trade very close to the low of the day on open, but if you look at the 30sec bars in TWS there is only a thousand shares traded in the first 30s, then it goes against you and the next 30s has 50,000 volume.

[ScottMcNab]

No problems with RUI or RUA….my fault was I was targeting some cheaper stocks in “entire USA” in the SM system in particular…would seem I did not have the turnover filters high enough…were 4 or 5 stocks where it hurt me for a couple of %…next month will clarify I hope

[JulianCohen]

Yes the smaller stocks will do that. To avoid that issue my strategy on the All stocks outside of RUA is set as C > 20 and EMA(V,10) > 200,000

That seems to keep things honest….I think it’s the C > 20 that makes the most difference as there’s less jiggery pokery with those. Probably less of those altogether….I never did a check to see how many there were.

Edit: I make it 530 with C > 20 and 320 with C > 20 and EMA > 200000

[Nick Radge]

Yes, I use C > 20 as well.

[ScottMcNab]

I was having a crack at sub $1 stocks in that one system …knew were risks…had volume and turnover in the millions but not (yet) enough so will keep experimenting

[ScottMcNab]

Jun 2021

MRV RUA 2.0%
MRV RUI 3.5%
MRV USA RB 5.3%
MRV USA AK 6.0%
MRV USA PB -5.8%
MRV USA SM 4.8%

[JulianCohen]

Noice…no leverage either right?

[GlenPeake]

Nice work Scott!

[ScottMcNab]

No Leverage Julian..I got a bit of a kicking in both April and May so performance was due to revert …was a relief tbh

[ScottMcNab]

[ScottMcNab]

[ScottMcNab]

July 2021

MRV RUA 1.8%
MRV RUI 3.9%
MRV USA RB 1.4%
MRV USA AK 2.4%
MRV PB -12.6%
MRV SM -6.1%

Continue to experiment with the MRV systems added in 2021. 70% of portfolio equity remains with the stalwarts from previous years (RUA and RUI systems). I have increased filters with USA systems so that there was no longer any slippage (when sell mkt on open) in July but in doing so the returns are now no better than other systems. As such, I will ditch the worst (PB has been fired) and will restart the MOC long system I was trading earlier in the year as it gives a bit better diversity in returns than if only trading MRV systems. Considering also stopping MRV USA RB and using a second MOC I have waiting in the wings.

All in all…would have been better if had just kept the MOC going from earlier in the year but the MRV USA looked so much better in testing :whistle: …but assumptions on how much liquidity was present were wrong…nothing ventured nothing gained

[ScottMcNab]

Aug 2021

MRV RUA 1.5%
MRV RUI 4.0%
MRV USA RB -0.7%
MRV USA AK 6.0%
MRV USA SM 5.4%
MOC USA 0.7%

[ScottMcNab]

Have been inspired by Glen’s post to do some research….so I thought I would also update my journal. First half of 2021 was meant to be the straightforward launch of new MRV and MOC systems…was a bit of a cluster but now into second half of Sep these are finally tracking. Second half of this year was meant to be launch of longer-term systems but only now getting to the paper trading stage so hopefully live Jan 2022. Three types are in development: daily trend following (breakout/trailing stop loss), weekly rotation and end of month rotation (with ability to exit immediately if index fails on any day). My goal has always been 15% cagr/-25% maxDD of long term portfolio with under 30% maxDD for any individual system. I am not confident that I will be able to achieve this last aim (actually I am fairly sure I won’t)….something always seems to sneak over the 30% DD mark somewhere in the back testing on some market…time for me to accept this and move on.

Of these three system types, the daily TF system is the only one I am happy with at this stage and I will start paper trading this weekend (actually my daughters will… they prob wont muck things up as often as I do). I find myself surprising comfortable with this daily trend following system. It is remarkably robust across time frames and I can alter ranking criteria or break out / look back time frames with relatively little impact on the results. The only thing I would like is a larger sample size but the systems only do 20 or so trades a year. For this reason, I have tested it on all data I have, just to try and exclude any complete disasters (not to add/remove or tweak any conditions or filters etc). The simplicity of this system is something I have not yet been able to achieve with my other long term (rotational) systems and gives me confidence (hope) that it is not over fitted.

I did some testing of the XAO excluding XTO based on Glen’s post and got a major improvement, particularly in the 2010-2021 period. While Glen was able to base his findings on a large number of trades, I have quite a small sample size. The improvements are enough that I will pursue this idea further.

edit: typo…should be XAOnotXTO

[JulianCohen]

Looks good Scott

[Author]

Posts