Jump Off

If you’ve been reading my posts for a while now you’ve come to realise that I’m a tad obsessed with the SPY. Mainly because, with the proper set of tools, it can be easily timed. Much in the same way that similar tools are used for folks that try to do practical things with statistics and probability – anything that has nothing to do with attempting to make a prediction as to what a certain number is going to probably be at a given time.

I’m about to show you a spreadsheet based system that, traded since November of 1993 would have made approximately $88.05/share over the last nearly sixteen years. So, let’s get to it.

Rules

1. If the SPY Adjusted Close is more than the 200 Day moving average and you do not have a share, then buy one share and hold it for the next 50 days.

That’s it.

Funny isn’t it? After all the smart math – all of the goofy work being done trying to fit quantum physics on top of equity prices and what not. I’m as much a fan of smart math as the next guy, and probably more than a lot, but that has always struck me as being well – silly.

Using Brownian motion formulae to scope out the suggested price chart still makes my head hurt while giving me a chuckle. You see, it’s all a bit like trying to figure out where a car is going by examining and projecting every single particle.

Why not just sit back and look at the whole thing in aggregate? It’s a whole lot easier I think.

The same goes true for the SPY and the DJI and well, a whack of the indices as a whole. And I’m not alone in this regard. Anyone who has read Curtis Faith’s Way of the Turtle will remember Chapter 10 and the performance his Donchian Time Based exit system traded. (hint: It was second only to the Dual MA system with a CAGR% of 57.2%)

He says “When I first tested the time-based exits, I was floored. They performed far better than I had imagined they would, better even than the breakout-based exits. So much for the idea that it is the exit that makes a system profitable. This shows that an entry that has an edge can account for the entire profitability of a system”

Note: Curtis Faith uses an 80 day based exit. I’m using a 50 Day window in this example but, there are probably others more suitable.

The Results

• Average Trade: $1.47
• Win Percentage: 73%
• Average Win: $3.77
• Average Loss: -$4.54
• Expectancy: $1.55

Method of Discovery

I took a spreadsheet and created a column for the date, the adjusted close and then started filling in the moving averages. I knew that I wanted to create a simple system that would take advantage of the primary difference (as I see it) between a “Bull” market and a “Bear” market. Essentially, the difference I have noticed is that the market falls faster and rebounds sharper during a bear market than it does in a bull regime.

That fact alone, is one of the reasons that the 357SPY is doing so well – it is taking advantage of the short term reversion to mean of the index and plays things very carefully by exiting as soon as any of the moving averages are crossed at the close.

In a bull market regime – the experiments that I’ve conducted has led me to conclude that the 357SPY does not take advantage of the slower running run that is made.

So, I created a set of columns for each of the well known moving averages (3, 5, 7, 11, 20, 50, 100, 200) and began setting up filters and formulae that would subtract the current days price from a date at some point in the future.

Example: I tested the 50 Day MA originally as “if the SPY is over the 50DMA then buy a share and hold it for 50 days”.

That did not fair so well, so I decided that I would consider the SPY trading over the 200 MA as a “bull regime” and then tested the expectancy of the various holding periods until it was above .5

Once that was done, I back tested the system by hand with a spreadsheet and am typing it up for you all today.

Conclusion

This system is not designed for the majority of the readership herein for the simple fact that it’s movements are glacial compared to the turn around times that most traders expect for a system to be profitable. It is designed for the folks who have 401k/RRSP that allow them only one or two trades a month.

Yeah, the folks that have basically been having their retirements destroyed by idiots that manage mllions of dollars and don’t really consider what one account is worth to the person that holds that account.

I wrote back in December that my focus has become aimed at developing systems that can be managed by spreadsheets for the simple same reason that a pinhole camera photograph taken by Ansel Adams looks a whole lot better than your neighbour’s holiday snaps.