Premium Model
Defensive Sectors (Canada)
Stocks estimated to deliver strong growth are the first to be penalized when they don’t meet expectations or when the market starts getting nervous, so market timing rules are enabled to mitigate market risks associated with a market crash due to recession.
This is a low-turnover model, and rebalance is done weekly. It holds up to 7 stocks. It only contain stocks from TSX (so no Venture Exchange stocks).
This model has been revised in November 2023 to include additional sectors in order to provide a balance between a defensive portfolio while seeking growth of the companies on that sector.
The idea is to initially filter stocks by sectors that are considered defensive or recession proof, and from that smaller universe, screen further for strong metrics regarding quality and value, and that has exceeded market expectations recently. The stocks are screened and ranked according to Greenblatt’s and growth principles, and then their distribution weight are assigned taking growth and momentum into account. Rebalancing on weight distribution is done weekly.
This is a mechanical model that chooses what to buy and sell based on a set of rules. Therefore, there will be losing trades from time to time. By no means it reflects a broken strategy. No model can outperform at all times, so it’s paramount to have the proper temperament to stick what a strategy that is aligned to your goals and risk tolerance.
See backtest information below to evaluate how these rules would have worked since 1999:
Model summary, including variable slippage (starting capital was $50,000) and fixed commission of $4.95:
Performance summary:
Detailed performance summary which provides max drawdown and % stock invested periods (log scale):
Stats information below, with winners, losers and how long typically these stocks are held for:
Below are various risk indicators as well:
Since the model typically holds positions for 1 month on average, the following histogram was run, to evaluate how consistently the model would perform if it started at different periods (roling offset was set to 1 week, while the performance period was set to 1 month); below is the histogram excess performance when compared to the benchmark (TSX):
And below is the same histogram, but showing the actual model portfolio performance, instead of the excess when compared to the benchmark (TSX);
The first screening criteria is related to the market timing, to either be in equities or cash. This market timing rules use data from micro-elements (earnings and price movement trends for the whole market) as well as macro-elements, such as economic indicators (which might drive prices lower, regardless of solid fundamentals) to issue a sell all signal (move to cash) or buy according to the rules of this model.
The next screening criteria (buying as per the rules of this model) rules narrows down the Universe to stocks from defensive sectors, with metrics aligned to typical recession proof companies. Then, further screening takes place to ensure that:
- Companies demonstrate good value, by having attractive fundamental metrics that indicates strong value (with lower price to earnings, book, while estimated to grow operational results), and strong quality (with healthy returns on margin, ROIC and ROE and strong balance sheet);
- Companies have demonstrated better than estimated results, exceeding expectations in one or more quarters for the last fiscal year;
Once a list of stocks are found, then they are ranked from best to worse, where only the top ranked are chosen. The ranking is based on weighted metrics for Greenblatt’s “magic formula”.
Defensive sectors being considered: Commercial Electronics, Healthcare Equipment, Bio-pharmaceuticals, Consumer Non-cyclicals, Utilities, Plastic and Rubber Products, Construction Materials, Downstream and Midstream Energy.
A stock will be sold if these metrics deteriorate for that stock or if market timing rules activates to switch to fixed income.
Subscribe to this model
Interested in subscribing to multiple models? Discounts are available when subscribing to 2 or more models. Simply subscribe to your first model and then let us know which model you want to subscribe next, and we’ll send you a discount coupon.
Discount rates:
| 1st model | full price |
| 2nd model | 10% discount (applied to the 2nd model) |
| 3rd model | 20% discount (applied to the 3rd model) |
| 4th model | 30% discount (applied to the 4th model) |
| 5th model | 40% discount (applied to the 5th model) |
| 6th model | 50% discount (applied to the 6th model) |
| 7th model | 60% discount (applied to the 7th model) |
| 8th model | 70% discount (applied to the 8th model) |
Got questions? Check our Frequently Asked Questions for Premium models or contact us.
Already a member?
Cost of this model: $28 / month. Cancel anytime.
Interested in subscribing to multiple models? Discounts are available when subscribing to 2 or more models. Simply subscribe to your first model and then let us know which model you want to subscribe next, and we’ll send you a discount coupon.
Discount rates:
| 1st model | full price |
| 2nd model | 10% discount (applied to the 2nd model) |
| 3rd model | 20% discount (applied to the 3rd model) |
| 4th model | 30% discount (applied to the 4th model) |
| 5th model | 40% discount (applied to the 5th model) |
| 6th model | 50% discount (applied to the 6th model) |
| 7th model | 60% discount (applied to the 7th model) |
| 8th model | 70% discount (applied to the 8th model) |
Got questions? Check our Frequently Asked Questions for Premium models or contact us.
Hi Rod,
This model being a defensive one, did it meet your expectations on how it should have behaved in the recent COVID-19 market crash? I suppose a crash is different from recession so perhaps that explained the beating it took.
This is not a defensive model, it’s a growth model based on Defensive sectors. Every sector got affected during the crash, and the way that market timing was implemented, it slowed recovery further. There were no data to contemplate this scenario (the closest setup is from 1929, but no data to simulate back then), and all models were revised to incorporate reversion to the mean moving forward, so they don’t get affected by flash crashes like this. Once back to be invested again, they are performing as per what I expected, according to the model rules.
Please let me know if you have any questions.
Right, silly me…got too focused on the word ‘defensive’