Mechanical Bull Portfolio

In this post I present the current status of my Mechanical Bull portfolio which I set up on 24 May 2013 with an initial investment of £30,000. This is not my real-life portfolio. I am constantly adding cash to my real-life portfolio, which complicates measurements of performance. However, I use the same buy and sell principles and I have interests in 12 out of the 15 stocks currently held in the portfolio below. The only stocks I don't currently have an interest in are Keller, Matchtech and Microfocus International. The reasons are mostly just down to timing. For simplicity's sake I exclude trading costs, but I also exclude dividend payments. I believe this to be a conservative approach, which if anything should slightly underestimate actual performance.

The portfolio has increased in value by 29% since inception, which is annualised rate of over 52.5%. The FTSE AllShare has fallen by 2.3% over the same period. I see this level of performance as exceptional and I do not expect it to continue. Still, it appears pretty solid with 75% of past and current stock showing gains. The overall overperformance has not been driven by just a couple of stocks, which gives me a fair degree of confidence in the approach. I will update this spreadsheet every month around the 24th of the month when I will usually make my trades.

Creston and Microfocus International are currently in line for the chop. Creston had a disappointing set of results last week while Microfocus have just drifted down in the MB scoring stakes. Interserve and Kentz are looking likely purchases in the lead up to Christmas. I have bought and sold Interserve twice (in real life) since early 2011, making a small profit each time. The second time I sold was in August this year as a result of following my current strategy. So, there is a natural resistance to buying in again when the price has jumped up further. I also currently own Kentz in my real-life portfolio. It has done pretty well is now my second most valuable holding.
 

Mechanical Bull Portfolio as at 12 December 2013

The Mechanical Bull Method

An idea rooted in Stockopedia

Until now, this blog has been laying the groundwork for my investment philosophy. In this post I will provide the nuts and bolts of my investment method. In essence it is very simple. However, it will be very difficult to understand without reference to some particular features of the Stockopedia website. So if you haven't already please read the following post.

As I have mentioned, there are two features about the Stockopedia website that I find particularly intriguing. The first is their idea of the 'Screen of screens'. This takes the 60 or so 'guru' screens and counts the number of times a particular stock appears in each screen. The idea is that a stock that appears across multiple screens is displaying strengths across a number of criteria. These stocks are interesting because they may be considered by investors with very different investment approaches.

The second feature is their recently introduced StockRanksTM. This is calculated by first ranking stocks in terms of quality, value, and momentum using a number of different metrics. These ranks are then compiled into a composite rank of these three factors. As I have mentioned previously, I believe that this kind of approach is better than screening, because a screen will sift out stocks that are weak on only one criteria, whereas rankings are more forgiving of any particular weakness.

The 'Superscreen'

It seems to me that both approaches are using different methods to achieve the same basic objective, that is, to find stocks that are attractive from a number of different angles. This got me thinking about an approach for combining the Screen of screens with the Stockranks to produce some kind of 'Superscreen'. Where you get a similar answer following two different methods, you should have more confidence in your results than if you only follow one.

My method for doing this is childishly simple. The Screen of screens is a simple count with the highest count typically between 8 and 10. StockRanks give each stock a composite value of between 0 and 100. There is only stock with a value of 100, about 20 with 99, about 20 with 98 and so on. My method for combining the two approaches is to to simply add the two value together. Here are the results all those with a 'Superscreen' score of over 100 as at 8 December 2013:

Ticker	Name	# Screens (Long)	Stock Rank™	Superscreen score
RM.	RM	6	99	105
RNWH	Renew Holdings	6	99	105
DTG	Dart	7	98	105
TT.	TUI Travel	7	97	104
FRP	Fairpoint	4	99	103
CHRT	Cohort	4	99	103
VP.	VP	5	98	103
JD.	JD Sports Fashion	5	98	103
ATK	WS Atkins	5	98	103
KENZ	Kentz	6	97	103
TRI	Trifast	3	99	102
NWF	NWF	3	99	102
STCM	Steppe Cement	3	99	102
THG	Terrace Hill	3	99	102
CAMB	Cambria Automobiles	3	99	102
MTEC	Matchtech	4	98	102
HFD	Halfords	4	98	102
RCDO	Ricardo	5	97	102
IRV	Interserve	6	96	102
CDY	Casdon	2	99	101
HYDG	Hydrogen	2	99	101
MCM	Motivcom	2	99	101
NWS	Smiths News	3	98	101
TOT	Total Produce	3	98	101
STAF	Staffline	3	98	101
CSG	Sweett	5	96	101
BKG	Berkeley	5	96	101
TNI	Trinity Mirror	6	95	101

Effectively, what I am doing is using the Screen of screens to sift out the most promising Stockrank candidates in the high 90s.

Trading Rules

In terms of translating this screen into an executable strategy. The buy./sell rules are as follows:

1. New buys are restricted to the top half dozen or so stocks. If I were starting form scratch at this moment I would buy equal amounts of all stocks with scores of over 103.
2. Accumulate for any existing stocks with a scores of over 100.
3. Hold for any stocks with scores 90-100.
4. Sell when score drops below 90.
5. Look to hold between 15 and 20 stocks.

I generally look to make all my trades on the same day about once a month. The precise day varies for practical reasons that have to do with taking advantage of reduces trading commissions. I do not try to time my trades with any trends in either particular stocks or the broader market. This is all part of a deliberate strategy of 'strategic ignorance'.

Does it work?

I have been implementing this strategy since 24 May 2013, that is, for just over six months. The strategy has returned 27.6% during this period compared with 19.5% for the standard Screen of screens. The FTSE All share is virtually unchanged over this period (down 0.2%) although the FTSE250 is up 6.4%. This is not long enough to provide definitive proof of this being an effective strategy. However, the results to date are certainly very encouraging. I will aim to do a bit more analysis of the performance of the strategy portfolio in future posts.