All things entail rising and falling timing. You must be able to discern this. In strategy there are various timing considerations. From the outset you must know the applicable timing and the inapplicable timing, and from among the large and small things and the fast and slow timings find the relevant timing, first seeing the distance timing and the background timing. This is the main thing in strategy. It is especially important to know the background timing, otherwise your strategy will become uncertain.
Book of Five Rings by Miyamoto Musashi
From Samurai.com
I’ve become a tad OCD with staring at the 5MA chart for the 4Qs and everytime I stop and stare at it for a prolonged period of time it says something new.
Oddly enough, Wood’s recent RSI(2) post came a a most fortuitous time because the more I looked at the numbers and that chart, the more I wanted to tweak the results.
Rationale AKA Motivation
Please consider that if outside were inside, I’d be Al Pacino in the CBS office – Cadence Weapon
Now, I’ll be honest with you – I’m looking to design a no thinking swing trade programme for what is called the “retail investor” AKA “The Sucker”.
I’m not building a race car – I’m thinking a station wagon albeit with armour plating and turrets on the roof. Remember that.
Why?
Because someone needs to. The fact of the matter is that a good many swing trading plans are designed for trending markets where it’s obvious what’s going to happen over the long term. Generally, the best ones are the bull markets because it reduces the chances of the retail sucker getting, well suckered.
True story: a couple of Christmasses back one of my relatives told me about an office mate of theirs that was going on repeatedly about how they were killing the market with their 401k money, how they were going to retire early blah blah blah (kind of reminds some of us of a certain commentator that started with the 10th letter of the alphabet and his fixation with a certain shiny metal).
Anyways, my relative did some quick math and realized that annualized this braggart’s profits were about the same if they would have been if they had been put in a money market account.
Besides, it’s supposedly impossible to time the market and I like impossible things. Especially impossible things that help the schleprocks make money in this loopy market.
The Qsinator version 4
note: This plan is to be used in the last 30 minutes of the trading day.
The usual WTF period.
0. Start with $5000 in capital
1. If for the last 3 days the high has been lower than the 5 Day Moving average AND you have no shares THEN buy 100 shares of QQQQ at market.
2. If you have 100 shares AND the close will be above the 5 Day Moving average THEN sell them at market.
3. If you aren’t in either position, then (to quote Senor Tropicana) “Eat a Sandwich”
The problem that I had with the previous versions that I blogged about in December was that they were in and out of the market very quickly. One had a whole lot of movement and if you were using a “standard” broker then the commissions would kill any profit you might have made.
The flip side was that there were some very oblique moves it would have made to take trades that turned over the next day.
The other problem was that since it was using a 5MA, there were plenty of times where the profits were chewed up because the sell signal came after the 5MA had caught up with the pricing.
I call that the rising timing.
After staring at the chart again I thought I perceived a small edge and so I tested it using a spreadsheet and found that only 12 trades would have been made between Jan 23 – October 28, 2008.
This means that it was not flailing around like a drunken monkey trying to take any and all trades. It was waiting for the right moment and then striking.
I submit to you as well that the new version does not seem to signal during a sideways market which, aside from the dip in November, the Qs have been mostly between 28 and 32.
The Results
Of the twelve trades that would have been made in 2008, ten of them were profits for an average of $99.88 with a standard deviation of 63.20 and the two losses for an average of $85 with a standard deviation of 108.89.
The largest profit was $235 and the largest drawdown was $162.
Because I don’t like to get all happy about things just because they seem to work, I decided to check use the ChiSquare over at graphpad and find out how ten profitable trades out of twelve stack up.
I am testing this against a coin toss, being right 50% of the time.
ChiSquare Says
P value and statistical significance:
Chi squared equals 5.333 with 1 degrees of freedom.
The two-tailed P value equals 0.0209
By conventional criteria, this difference is considered to be statistically significant.
To be honest, I would have preferred this to have been extremely statistically significant but I’ll take what I can get.
| Date | High | Adj Close | 5D Mov Avg | Recomendation | Purchase/Sale | Gain/Loss | ||
| 23/01/08 | $44.15 | $43.92 | $44.90 | BUY | -$4,392.00 | |||
| 24/01/08 | $44.93 | $44.81 | $44.68 | SELL | $4,481.00 | $89.00 | ||
| 07/02/08 | $43.79 | $43.02 | $43.93 | BUY | -$4,302.00 | |||
| 11/02/08 | $44.22 | $43.98 | $43.36 | SELL | $4,398.00 | $96.00 | ||
| 04/03/08 | $43.04 | $42.82 | $43.30 | BUY | -$4,282.00 | |||
| 11/03/08 | $42.85 | $42.74 | $42.20 | SELL | $4,274.00 | -$8.00 | ||
| 15/04/08 | $44.35 | $44.08 | $44.53 | BUY | -$4,408.00 | |||
| 16/04/08 | $45.48 | $45.31 | $44.62 | SELL | $4,531.00 | $123.00 | ||
| 11/06/08 | $48.52 | $47.31 | $48.78 | BUY | -$4,731.00 | |||
| 13/06/08 | $48.47 | $48.30 | $48.03 | SELL | $4,830.00 | $99.00 | ||
| 30/06/08 | $45.83 | $45.14 | $46.14 | BUY | -$4,514.00 | |||
| 08/07/08 | $46.01 | $45.94 | $45.17 | SELL | $4,594.00 | $80.00 | ||
| 21/08/08 | $47.08 | $46.84 | $47.32 | BUY | -$4,684.00 | |||
| 22/08/08 | $47.68 | $47.46 | $47.18 | SELL | $4,746.00 | $62.00 | ||
| 05/09/08 | $43.78 | $43.42 | $44.75 | BUY | -$4,342.00 | |||
| 11/09/08 | $43.69 | $43.57 | $43.10 | SELL | $4,357.00 | $15.00 | ||
| 02/10/08 | $38.25 | $36.75 | $38.61 | BUY | -$3,675.00 | |||
| 13/10/08 | $35.25 | $35.13 | $32.60 | SELL | $3,513.00 | -$162.00 | ||
| 24/10/08 | $30.25 | $29.51 | $31.06 | BUY | -$2,951.00 | |||
| 28/10/08 | $31.88 | $31.86 | $30.23 | SELL | $3,186.00 | $235.00 |
Takeaway
I don’t think it’s quite ready yet though I’m waiting for the first signal since Oct 28 to let go and start tracking it for 2009. If you are interested in this idea and you’re not a piker, then take a look at the excellent work that Wood has done with the 5MA.
Theme Song



(6 votes, average: 4.33 out of 5)

You mathematicians/thinkers/trading quants crack me up. Stuff never works out like it does on paper.
Nicely done though.