I know that you have talked about the medical devices sector in the past. Would you kindly explain your sorting method?
Start with VL
Select entire sector (well, both sectors, invasive and non-invasive medical devices)
Remove dividend payers
Then I’m unclear how you are narrowing down to the top 15 holdings.
Do you sort by those with the most cash? Those with the largest one-year (or five-year?) sales growth rate? Something else?
Thanks in advance for any clarification that you might provide
Start with VL Select entire sector (well, both sectors, invasive and non-invasive medical devices) Remove dividend payers Then I’m unclear how you are narrowing down to the top 15 holdings.
Do you sort by those with the most cash? Those with the largest one-year (or five-year?) sales growth rate? Something else?
Couple of variations.
Sort by 5-year annualized rates of sales growth. (Higher is better)
Sort by cash to market cap ratio. (Higher is better)
He has mentioned doing a Screen-of-screens by doing half of the portfolio each way.
================ … rebalance and reconstituting the portfolio periodically, but it doesn’t mean much trading as firms don’t change industries quickly. You could do it annually and get much the same result.
Since 1986, that set has averaged about 55 stocks. If that’s too much typing, limit it to (say) 20 or 30 stocks by choosing the ones with the highest fraction of cash in their market cap. … * Value Line stocks, standard edition (average 1658 stocks) * Stock has any valid Timeliness rating (average 1512 stocks) * in those two industries (average 59.4 stocks) * No dividend (average 32.4 stocks) * Sales-growth 5 year greater than zero (average 23.4 stocks) * { Sort/ranking step, your choice. } * Equally weight them all, reconstitute and rebalance each two months
Start with VL Select entire sector (well, both sectors, invasive and non-invasive medical devices) Remove dividend payers Then I’m unclear how you are narrowing down to the top 15 holdings. Do you sort by those with the most cash? Those with the largest one-year (or five-year?) sales growth rate? Something else?
Excluding dividend payers is optional.
It seems to improve long run returns a hair in backtest, but is hardly the definitive secret to the strategy.
I think the reason it helps is that it causes you to skip a few giant mature conglomerates like JNJ
that aren’t fast growers or, for that matter, very concentrated in medical devices.
As for how to narrow it down to a smaller and constant number of stocks:
I’ve proposed a couple of ways.
The highest ratio of cash to market cap works very well.
The thinking is that the sector might include some dodgy startups; the ones with a big cash pile are less likely to go pop, whether they have any profitable sales yet or not.
Another good alternative is to pick the ones with the highest 5-year sales growth figures.
(geeky people like me could rank them both ways separately, sum the two rank numbers for each stock, and pick the lowest sum of ranks…)
And of course you could simply buy them all. It’s not that much typing.
My preferred version goes like this:
those two industries, average 59 stocks through the years
ensure the stock has a timeliness rating (not too new, no huge M&A lately). This chops off 3-4 stocks on average.
Note, you’re not looking for a GOOD ranking, just ensuring that it’s ranked.
No dividend. This is the high flyer version! Down to 32 stocks.
Highest 15 stocks sorted by cash to market cap ratio
Buy equal dollar amounts of the top 15, hold a month.
Every month find the fresh picks, sell anything no longer ranked in the top 25 and replace it with the highest ranked fresh pick you don’t already own.
This eliminates a lot of needless trading, since the 15th ranked isn’t likely to do meaningfully better than (say) the 18th ranked.
The estimated trading costs are 0.4% per round trip trade, adding up to only 0.16%/year for the portfolio.
It seems best to rebalance to equal weight periodically, but it could be monthly up to twice a year with no visible difference to returns.
Some brokers now offer a one-button “rebalance” feature.
This particular industry lagged pretty badly in the tech boom, 7%/year while the S&P left it in the dust.
But even so, results May 1997 through March 2022 inclusive, this would have beat the S&P by 15.03%/year after estimated trading costs.
In that test it beat the S&P in 81% of 288 rolling years (checking monthly), not just one big rally.
Nothing is that good in real life, but it’s a reasonable omen that there might be a bit of an edge to be had in future.