When you look at performance of the StockRank system in the USA (under Ranks - Performance), the S&P 500 has outperformed all Deciles of StockRanks since 2015. Is there any reason for this?
For the UK and Euro zone, the results are ideal for using the rankings system.
Also to Stockopedia management, I'd love to see this performance measure provided for Australia.
I would hazard a guess that part of the reason is the FAANG stocks' dominance of the S&P500.
I agree with Nick!
At first sight it seems anomalous that all of the StockRank deciles could underperform the index.
But one has to remember that the S&P500 is a market cap weighted index and so is hugely dominated by a few of the very biggest market cap companies. (According to Wikipedia for example the top 2% of companies account for over 20% of the value).
So the over-performance of the S&P vs every decile of the StockRanks suggests to me (I am no expert on the US market) that the strong performance has been largely driven by a small number of these very biggest companies.
If there was available (and maybe there is?) information on the performance of an equal weighted portfolio of the S&P500 members (eg a $1m portfolio with $2,000 invested in each S&P member) then I suspect that the index would have hugely outperformed this equal weighted portfolio.
I expect that this equal weighted portfolio would probably sit somewhere in the middle of the performance table of the different StockRank deciles.
Assuming that performance of the SR deciles is broadly in the right order (90+ being above 80-90 etc) - I don't subscribe the US data, but last time I saw published info it was broadly true.
Then I would argue that the StockRanks are "working"; in the sense of discriminating between individual companies that are likely to be good investments vs those that are less likely to be so.
However, it seems that there has been a much more powerful discriminator in play for the last few years "Buy the FAANGS" (actually it may not just be the FAANGS as there are a few other significant majors at the top of the index that may have contributed - $MSFT lead the pack and I think they are up c. 140% over three years).
Over-time though we MIGHT expect this trend to reverse, if one or more of the majors in the S&P stumble the S&P's future performance could fall below the median. (Although the way the indexes rebalance, tends to mean that they are always backing the winners so momentum being what it is indexes have a strong performance to outperform equal weighted portfolios.
This struck a chord with me as I have been researching recently for an article elsewhere on the dot-com boom and bust and very provisionally it looks to me that if one had been following an equal-weighted index, then you might have missed most of the excitement.
In fact it is a key result of Stochastic Portfolio Theory that an equal-weighted portfolio will out-perform a market-weighted portfolio of the same set of stocks provided that the market is "diverse".
This is a rather arcane property which you can read about in the linked paper, but the basic idea is that the capital in the market stays distributed in roughly the same way amongst the stocks over time, rather than gradually becoming concentrated in only one or a small number of companies. It is generally thought that market diversity is a property seen in all real markets.
But, with the FAANG situation, the S&P is far less diverse (in this mathematical sense) than maybe it should be and this is what is causing the S&P to out-perform equal-weighted portfolios. Arguably, this is not a healthy situation and will eventually correct itself - possibly painfully.
(Curiously this concept of "market diversity" was a driving factor in the development of SPT in the first place and a reaction to the limitations of Modern Portfolio Theory. In MPT it is assumed that stocks have a constant underlying growth (with some volatility) and so at T=infinity whichever stock has the largest growth would completely dominate the market - something which is assumed to not be true of real markets and caused MPT to make some incorrect predictions over long time periods.)
We actually stripped the FAANG stocks out of the US market and found precisely this - QVM does beat the market without them. I haven't found that paper before Nick - printing off now and it actually might trigger a blog from me.
So it's interesting that the QVM Stockranks 'work' if one's strips out the FANG stocks from the S&P-500. But for completeness, do they still work when one strips out the dominant stocks from ALL indices?
Facebook, Amazon, Netflix, Alphabet & Microsoft account for 13% of an index of 500 stocks.
HSBC, BP, AstraZeneca & Royal Dutch account for 28% of an index of 100 stocks.
My doubtless dodgy arithmetic suggests that if FANAM were part of an index of 100 stocks they would account for up to 65% of it. I know it's not that simple, there are other very large companies in the S&P500, but you are not comparing like with like.
You can do the experiment yourself on the stockrank performance page by "hacking" the url, as recommended by Ed in the past.
Set the performance filter to UK, qvm stockrank, >350 , quarterly, quintiles (simplifies the graph) and submit. Then replace cap=350 in the url with cap=5000 (roughly the FT100). Compare that to the range cap=50-5000 (everything except the FT100)
I had a quick look and your estimate is incorrect, I only went down to DIS and I was already more than the FANAM share of the index, DIS is 18th largest company on the S&P500, so I guess those 5 are only about 30% of the top 100. They account for about 41% of the Nasdaq 100.
I don't really see how you can exclude some of the more successful shares to say that Stockranks works. FANAM is not FAANG anyway. I do find it odd that the index does beat every Stockrank level on the performance table. Will be interested to see Ed's figures. MSFT is the largest share and has a SR of 91 at present, APPL is 2nd largest and has SR of 85, I don't subscribe to the US SR so I don't know what they are earlier this year, my question is why is Stockopedia not picking the FAANG is it's top 20? Is it a timing issue?
