On 22 November last year, we sketched out a mechanical stock picking strategy that comprised of two valuation metrics (P/E and Price/Cashflow), one quality metric (Return on Capital) and a momentum metric (12-month % change) when selecting the top 10 shares to buy on the JSE.
Based on some comprehensive local research, these metrics were found to be the most powerful predictors of a share’s likely outperformance since 1985, provided they are all used together.
It has been almost three months now, and we have decided to execute the quarterly rebalance today, based on our assessment posted to clients on Friday that there is a 81% probability the JSE bottomed last Thursday, after a peak-to-trough drawdown of 6.6%.
- ALVIVA (previously PINNACLE), KUMBAIO & RAUBEX remain.
- ROLFES, SAPPI, PAN-AF, WBHO, MRHLD, GROUP-5 & CLOVER are sold.
- New entrants are ASSORE, ANGLO and five others to fill positions of the sold shares.
The equally weighted portfolio gained 10.87% versus the JSE’s 1.75% for the period. There were only two losers in the ten selections, for an 80% win rate. More importantly, gains exceeded losses by a factor of 5-to-1. This means that for every 1% we lost, we clawed back 5% on the winners. Normally you like to see this gain-to-loss ratio exceed 3-to-1, so this is a very satisfactory result. With regard to today’s seven disposals, the average gain was 7.95% with 2-in-7 losers (a 71% win rate) and a gain-to-loss ratio of 3-to-1.
Remember you can build many of these trading portfolios quarterly and leave them individually to run in parallel for 12 to 18 months, or you can run a single portfolio and rebalance quarterly as we have done above.
The performance tracking of the portfolio constituted on 23rd November 2016 appears below. It is typical of this style of share-selection strategy, namely you have two to three shares that shoot the lights out, two to three shares modestly outperforming the JSE index, one or two shares that closely track the index itself, and maybe two or three losers, with one being outsized. This demonstrates the importance of the "shotgun" approach of buying all ten shares in the selection and not "cherry picking" one or two if you are trying to execute the strategy.
(Click the image to enlarge)
The chart shows that an end-of-day stop of 10% would have ejected PAN (-23.3%) from the portfolio early on. This kind of wide stop is important with this strategy (as it must allow the horses to "gallop") so it means you need to adjust your position-sizing for each share accordingly.
If you take your available portfolio amount and divide it by 10 to dictate how much of each share you can buy, and configure all 10 shares with 10% stops, then theoretically your maximum loss if every share should tank 10% in the quarter is no more than 10% of your entire portfolio. Alternatively to cap your worst-case loss to 5% of your entire portfolio, divide it by 20 to dictate how much of each share you can buy, whilst retaining the wide 10% stops.
If you have the Sharenet Advanced Charts add-on, then you can track the progress of the above chart by loading the PWS-MSNOV16 chart that is being made available to PowerStocks subscribers and which is updated in real time:
Dwaine van Vuuren
RecessionAlert, PowerStocks Research
Dwaine van Vuuren is a full-time trader, global investor and stock-market researcher. His passion for numbers and keen research & analytic ability has helped grow RecessionALERT.com (US based) and PowerStocks Research into companies used by hundreds of hedge funds, brokerage firms, financial advisers and private investors around the world. An enthusiastic educator, he will have you trading and investing with confidence & discipline.