StockRanks Performance - one year on

Friday, Aug 15 2014 by
StockRanks Performance  one year on

The recent credit crunch illustrated to hundreds of thousands of investors that even blue chips can’t be relied on to preserve and grow capital. And yet so many still invest first in stocks they are familiar with and think they know best.  Buying what one knows may feel safe, but they can be over-followed, over-owned and over-priced.  A good company may not make a good stock.   So what makes a good stock?

Behavioural Finance to the rescue?

Much academic research shows that oddly enough, what makes a good stock is often what investors don’t like buying.  The typical investor is scared of buying cheap stocks and even more scared of buying stocks at new highs.  It’s these behavioural biases that are behind the ‘value effect’ and the ‘momentum effect’ - two of the driving forces behind market beating portfolios.

More recently, research has illustrated the power of safe, stable, profitable businesses to beat the market (a ‘quality effect’). In fact a team investigated Berkshire Hathaway’s performance (in a paper titled “Buffett’s Alpha”) and found that all Warren Buffett had done was figure out before everyone else that ‘good, cheap’ stocks beat the market - more importantly Buffett has always had the tenacity to stick to at it while everyone else loses their head.

Stockopedia StockRanks

What we’ve been doing at Stockopedia over the last few years is diligently putting rule based investing to work to find those shares that might be most exposed to these forces.  We began our journey in 2011 by modelling and tracking the performance of 65 classic strategies from the stock picking literature - the GuruScreens.  But while stock screening is a great way to filter for high quality candidate stocks in the market, it doesn’t help anyone to understand or compare the whole market.  So we set about designing a ranking system that could.

Inspired by the work of academics like Robert Novy Marx and popular stock market authors like James O’Shaughnessy and Joel Greenblatt - we sought to quantify every company’s exposure to the factors we have been discussing - quality, value and momentum.  Our goal was to highlight ‘good, cheap, improving’ stocks while shunning ‘risky, expensive, worsening’ stocks.  

The Stockopedia  ‘StockRanks’ - which score companies from 0 (worst) to 100 (best) - have an impressive record in the UK since launch in April…

Unlock this article instantly by logging into your account

Don’t have an account? Register for free and we’ll get out your way


As per our Terms of Use, Stockopedia is a financial news & data site, discussion forum and content aggregator. Our site should be used for educational & informational purposes only. We do not provide investment advice, recommendations or views as to whether an investment or strategy is suited to the investment needs of a specific individual. You should make your own decisions and seek independent professional advice before doing so. Remember: Shares can go down as well as up. Past performance is not a guide to future performance & investors may not get back the amount invested.

Do you like this Post?
24 thumbs up
0 thumbs down
Share this post with friends

21 Comments on this Article show/hide all

Edward Croft 16th Aug '14 2 of 21

In reply to Sully8786, post #1

Hi Sully.

We are working hard on scaling up our backend architecture to handle the data explosion that's going on behind the scenes here. In 2 years we've moved from computing a few hundred ratios and aggregates for 2000 UK stocks to a few thousand ratios and aggregates on 18,000 stocks, sectors and markets.

Being mildly ambitious with all this the goal is to extend the coverage to about 45,000 global stocks over the next 18 months, deepen the coverage to geographic and business segments of companies, deep screenable technical analysis, all the ownership data (Smart Money in progress), and extend the ranking system to a much deeper nested hierarchy of ranks.

This is all going to take time. I've just this afternoon been reviewing some work we've been putting in on the StockRanks hierarchy. It's shaping up to become a 4 level deep set of ranks with the StockRank and Quality, Value, Growth and Momentum Ranks as the first 2 levels.

When we can release this StockRanks work I'm not yet sure. We have a v. busy pipeline and also a complete redesign of the website under way from the ground up, but we are inching forwards on all fronts. If I'm realistic, it's probably 12 months away. That might seem a long time but it will come round very quickly. The great news is it will allow much more granular screening across all kinds of company rankings.

Thanks for your interest and patience on this.



Blog: Follow @edcroft on Twitter
| Link | Share
Mechanical Bull 17th Aug '14 3 of 21

Ed, Thanks for showing the updated Stockranks chart. It shows a nice fanning out pattern for each decile from May 2013 to February 2014. However, it is worth pointing out that since the early part of this year, this pattern seems to have dissipated almost completely. All deciles appear to have declined at more or less the same rate. In other words, StockRanks seem to have lost nearly all their  predictive power.

This is not to say that the StockRanks are not useful, they certainly are and I am sure that over the long term, they will prove their worth However, it reaffirms your warning that past performance is not necessarily a predictor of future performance. Progress is not a straight line.

A more interesting question for me is why these performance cycles occur. I really have no idea, but if anyone has any ideas, please share!

Blog: Mechanical Bull Blog
| Link | Share | 2 replies
ajsmck 18th Aug '14 4 of 21

In reply to Mechanical Bull, post #3

One "Massively Reductionist" answer to the question of why the performance has not continued (bearing in mind I'm just a user, not a developer), is that 2013 was an exceptionally good year for the market. Alongside real signs that the recovery was finally on, risk was also most definitely "On" for much of the year and this led to a very bullish mood across the board.

This makes investors less risk averse and thus more prepared to take punts on smaller "growth" companies like aim stocks, which rose massively as an index for that reason - as well as others such as the government making AIM stocks holdable within an ISA for the first time, and subsequently getting rid of stamp duty too, all to encourage investment in small businesses.

If you think back to the premise of the stock rank system, it is looking for (paraphrasing) good, cheap, improving stocks and then ranks them according to those criteria. Thus a smaller company is more likely to rank high than a larger one because 1) Usually Cheaper - illiquidity and mkt cap minimums for funds along with a lack of analyst coverage and less news lead to a lack of interest in these stocks in a "risk off" environment (which we largely had up to 2012-13 since the crisis), therefore you will be looking at some very cheap valuations in small companies and thus high ranks in that category 2) Usually more growth - these smaller companies you've found with cheap valuations tend to be less mature businesses than larger ones and you probably already know that smaller companies tend to outperform the market because of the increased proportional value of any earnings upside. After these points you are then left with Quality, and hey, a company that makes money is better than one which doesn't right? (again, reductionist but the point holds), this criteria is more likely to be met in an improving economy racing out of recession (2013).

And there you have it, because of the profile of a highly ranked company as designated by the stockrank system, and the pace of the recovery in 2013, as well as certain government regulatory changes that took place in the period being one-offs, we are unlikely to see similar outperformance in 2014 (which so far looks like a year of moderate growth and consolidation, but could yet turn out to be dominated by geo-poilitical risk and end up down on the year - who knows, no-one has the proverbial crystal ball!).

Having said all that, I still use the ranks as an integral part of my investment process, as well as lots of research of course (DYOR and all that!).

You can't expect to win'em all, but you can do everything in your power to tilt the odds in your favour enough towin more than you lose.

Best of luck,


| Link | Share
Edward Croft 18th Aug '14 5 of 21

In reply to Mechanical Bull, post #3

Hi MB/ajsmck,

Thanks for comments.  Firstly - no system is going to work all the time !   We haven't made that claim for the QVM ranks but, given the market's historic tendencies, we are optimistic they will continue to work over the long term.  If they don't then the market will have either become completely manipulated or completely efficient - if returns don't continue to accrue to good cheap stocks then everyone should pack up their bags and go home. 

Joel Greenblatt's Magic Formula just focuses on 'good, cheap' companies. His studies suggested a 30% annualised return (which nobody has been able to replicate), but other studies have shown an up to 17% p.a. annualised return for his system. Of course he made it plainly clear that it didn't work all the time - in fact there were 2 year periods where it fell lame.  That's the nature of value investing.

There are plenty of studies that show that often when value ceases to work, momentum takes over and vice versa. Value & Momentum are slightly uncorrelated, which means that incorporating momentum in portfolio construction should lead to shorter underperformance periods in systems like this. Certainly that's what academics like Robert Novy-Marx and Cliff Asness have found.

I'm hopeful that the QVM deciles will continue to fan nicely over every 1-2 year period - what happens in the interim is anyone's guess. We've been extremely lucky that in the first year since launch they behaved so well.

A couple of points to make. Firstly, the performance YTD has been less remarkable, but we've still seen an 8% spread between the top 20% ranked stocks and the bottom - it's less noticeable but still a clear outperformance. See below:


Also - I wouldn't say that last year's performance was entirely small/micro cap driven.  Here's the £100m+ rank performances held since inception (I unfortunately can't generate the six-monthly rebalanced portfolio for this test at the moment).


And here's the £1bn+


It should be noted that there are fewer and fewer stocks in the bottom deciles as you move up the Market Cap (Size) spectrum.  In that last test there are only 3 stocks in the lowest ranked group, vs 77 in the top set. Larger cap companies in general tend to be higher quality stocks.  Most of the bottom ranking stocks are falling in price & low quality  - bargain basement microcaps make up the majority.

PS - we will be providing the opportunity to generate these charts yourself on the site at some point in the future.  Not quite ready yet !

Blog: Follow @edcroft on Twitter
| Link | Share
dangersimpson 18th Aug '14 6 of 21


As you may recall I'm a big fan of stockranks and the idea of having sub ranks. It would be good to see where you are with the stock rank hierarchy and get feedback from subscribers when you are ready with the ideas but before implementation.

I think the Quality Rank is something that would benefit greatly from sub-ranks - for me a quality measure like Sustainable Competitive Advantage (profit margin stability and growth, gross profit to assets, long term ROCE etc) is quite different to one that measures Earnings Quality (Beneish, Accrual ratio, SNOA) or Financial Strength (Z-Score, CHS, F-Score, Volatility). The sub-ranks gives me the ability to exclude say the lowest ranks on Earnings Quality and Distress, say bottom 20% (since the majority of companies are not fraudulent, manipulating earnings or bankrupt) then I could then filter for companies that have the highest sustainable competitive advantage, say greater than 80%. At the moment a company fraudulently reporting great margins & ROCE may appear high on the combined quality rank.



| Link | Share | 1 reply
yellowdog 18th Aug '14 7 of 21


I am becoming a believer in your system and enjoy the SR reviews when you publish them.

Is it possible for subscribers to find out which stocks are going up or down in the SR and if so how do we do it.

For instance on the 5th Aug. review You said LTHM SR had risen by 19 to 97- that 's good info. but how can we find the same raisers and faller ourselves.


| Link | Share | 2 replies
JulianD 18th Aug '14 8 of 21

Hi Ed,
I was on the webinar when you presented this material and I was very impressed with the performance of the stockranks especially how well they performed through the range with such accuracy. This must show that with sufficient exposure across a range of stocks, that an above average performance should be attained. As a recent convert to a self select ISA and SIPP, I still need time to work on my strategy but the stockranks are a great help. I really enjoyed your book 'How to make money in value stocks' as well.

| Link | Share | 1 reply
whitepjs 18th Aug '14 9 of 21

In reply to yellowdog, post #7


Until the site gives that option (I haven't seen that it currently does), there is a way of doing it yourself. You can't do it looking back, but you can do it going forward.

Simply create yourself a screen say for any company with a StockRank (or whatever rank interests you) e.g. >50. Then download the CSV file of the screen results (see Download at bottom of screen). Save the file with a date in the name.

In a week or month's time (depending on what period you want to compare), do the same thing and again give the file a new name. You then open both files using MS Excel and copy and paste the two lists into a new Excel spreadsheet. You then compare the two lists to see which companies have entered or left your screen during that time. Here's a tutorial on how to compare the two lists using Excel.:

If you know how to use MS Access, it's even easier to do and enables you to keep a running score for any given company.

Just keep in mind that the site won't let you download more than the top 200 results, so you need to make your screen fairly specific and not just try to dump lots of data.



| Link | Share
Edward Croft 18th Aug '14 10 of 21

In reply to dangersimpson, post #6

Hi Mark, Yes this is exactly what we are thinking. I know you've shared your own hierarchy before and our work in progress hierarchy is not so dissimilar.

Regarding potentially distressed/manipulating stocks ranking highly - I had a chat last week about Wincanton with Paul Scott - it's ranking 99 on the combined rank but is clearly massively levered - the Altman Z Score is in the distress zone. It's this kind of stock that many might like to exclude from their own portfolios using additional filters. I'm aware of the great research in the book Quantitative Value which showed the benefit of using these kinds of pre-filters on portfolio results - so yes very much on my mind.

I think you are right that we should throw this open to discussion. At some point in the coming months we may start a thread - though given how many other sites are becoming prone to copying our work here it might be worth us waiting till we are closer to launch :-)

Blog: Follow @edcroft on Twitter
| Link | Share
Edward Croft 18th Aug '14 11 of 21

In reply to JulianD, post #8

Julian, thanks for the compliment on the Value book - it's certainly overdue an update - we were discussing it last Friday - time for a second edition I think.

On the ebook note - we'll be publishing a StockRanks eGuide a bit later this year - I hope in a few months - which will divulge a lot more of the background and thinking on the subject.

In the meantime, aside from our own knowledgebase and blogs, it's definitely worth reading up further on James O'Shaughnessy (What Works on Wall St 4th Edition) and Joel Greenblatt (Little Book that Beats the Market) - there is a lot of accessible material in those books that were strong background influences on the StockRanks. There's lots of academic material (Novy-Marx, Haugen, Lakonishok) that we've based our work on too - but a lot of that is over many people's heads. We'll put a lot of summaries and more further reading references in the eGuide.

PS Yellowdog - there is a stockranks changes URL - but I'm not giving it out yet. It's what we base our fortnightly article upon. We'll be releasing it as a feature in time, you'll be able to get email alerts for new candidate stocks. Can't promise when yet - lots going on behind the scenes right now.

PPS - always remain skeptical ! Stocks go up as well as down and we've been lucky to date that the StockRanks system has behaved well. At some point the deciles aren't going to behave themselves at all well !

Blog: Follow @edcroft on Twitter
| Link | Share
davidtalbo 18th Aug '14 12 of 21

Hi Ed

An interesting original article and interesting subsequent comment from Mechanical Bull and your response. I would like to add a few observations which I hope are constructive.


From the graphs you have supplied it appears that the percentile StockRanks outperformance does break down but not, as Mechanical Bull has suggested, from the start of 2014, but from around the start of March 2014. Therefore it may be helpful to split your original analysis into pre and post the start of March 2014. Up to March 2014 we had a generally rising UK stockmarket, and from March 2014 the overall UK stockmarket has been trending downwards. Overall it would be desirable to do an analysis over a least one economic / market cycle, so I am not am not suggesting one can draw definite conclusions from such a limited time period. However, my observations suggest that StockRanks, as currently configured, may perform better in a rising market than a falling one.


I have always been puzzled why you do not have an income rank.

Firstly income stocks are generally known to perform well. If one looks at the average annualised performance of your Guru screens since inception (as a rough measure of the relative performance of stock type) then the results are:
Momentum 22.2%
Value 22.0%
Income 19.2%
Growth 17.8%
Quality 16.7%
Bargain 13.2%.
Although the performance of income stocks above is impressive, it understates the position, since Stockopedia performance metrics take no account of dividend income, and hence the higher dividend income of income stocks is ignored..

Secondly, income stocks are known to exhibit relative outperformance is falling markets, since the relatively higher dividend payments tend to support the price. If one looks at the average performance of the Guru screens over the last 6 months (that is approximately from the start of March 2014 when the UK market started falling) this outperformance tendency is reinforced:
Income -0.98%
Quality -4.27%
Growth -4.34%
Value -4.63%
Momentum -5.15%
Bargain -6.55%.
The above figures again understate the income stocks impressive relative performance, since Stockopedia metrics exclude dividend income.

To summarise income stocks appear to perform relatively well, particularly in falling markets.

Also David Dreman's research has shown that certainly US income stocks (high dividend) exhibit the essential percentile out performance characteristic, to form an effective means of stock ranking. Dreman has also shown that income stock rankings are particularly effective in bear markets.

I think Stockopedia is an outstanding, visionary product. However, I have thought that StockRanks, since inception, should have included an income rank (possibly replacing the qualitiy rank) and this is worthy of further investigation. For my own investments, I use StockRanks but with an income rank add-on and a quality rank partial reduction, and this certainly has worked well for me.

Kind regards

David Talbot

| Link | Share | 1 reply
davidtalbo 18th Aug '14 13 of 21

Apologies Mechanical Bull

You suggest the StockRanks percentile outperformance breaks down at around the same time as suggested by me, and not from the start of 2014, as I incorrectly stated in my previous post. Pleased we agree!

David Talbot

| Link | Share
Munday 19th Aug '14 14 of 21

Very interesting discussions. I was particularly interested in the idea of an income rank section,as this would be suitable in preparing for a falling market,which must come at sometime in the near future.

| Link | Share
pka 19th Aug '14 15 of 21

Ed, Thanks for all the graphs you've shown in these comments for the performance of the QVM StockRank. Could you please show similar graphs for the combined QV Rank? This is because I'm considering using the QV Rank to select my portfolio, as I think it might require less frequent changes than one based on the QVM StockRank, so I'm very interested in how well it has performed relative to the QVM StockRank over the period that both have been in existence.

| Link | Share
dangersimpson 19th Aug '14 16 of 21


Income is already part of the Value rank. An argument can definitely be made for having an income rank which is a sub-component of the value rank (maybe dividend yield, dividend streak and 5yr avg dividend growth?) however I don't think the case can be made for a separate income rank since you are really using income as an indicator for capturing a value premium (i.e. you are interested in the total excess return not just the income.)

A word of caution that if you are building an investment strategy on income IIRC the Dremen study showed that dividend yield was the least effective source of excess return with approximately half that of P/E or P/B, although as you mention a reduced draw down. I would also expect this excess return to be lower today than when Dremen did his study as companies have favoured buybacks more and more. For this reason Meb Faber uses shareholder yield = dividend yield + buyback yield (net of equity issuance) + debt pay down yield. This is so similar to FCF yield for the majority of companies I'm not sure its worth the extra effort but the principle is good. In the book 'Quantitive Value' Gray et al found that Earnings Yield = EBIT/EV was the most effective value indicator. Overall to me this is an argument for a Value Rank of different proven sources of excess return of which income is one component as we have today.


Makes sense that you don't want to show your hand too early. As all investors with a good historical perspective are aware the true beneficiaries of innovation are usually the consumers not the companies who innovate since competitors can copy any innovation that doesn't have a big moat. IMO one of your sustainable competitive advantages is your willingness to both listen to your customers and to keep up with the latest industry research and then to react quickly to those ideas. Competitors can copy your ideas but while you do that you will always stay one step ahead.



| Link | Share
Edward Croft 19th Aug '14 17 of 21

In reply to davidtalbo, post #12


We are actively considering a set of Income/Dividend Rankings as they'd clearly be popular. I would though say the following:

Firstly, we've tried to build the ranking system along the lines of the key academic drivers or market returns - i.e. along the lines of so-called stock market 'anomalies'. The "QVM" acronym we use incorporates the Value effect, the Momentum effect, the Quality effect. Of course there are many more strong anomalies - the 'size' effect, the 'liquidity' effect, the 'volatility' effect etc etc... BUT we wanted to build a ranking system that was applicable to ALL stocks in the market rather than zeroing in on those smaller size/liquidity/volatility segments. 

We know there are ranking systems that can be generated that can perform 'better', but that wasn't the point of our system. We hoped it would perform well (and hope it will continue to) while providing anyone with a point of reference for analysing any capitalisation of stock.

High dividend yielding stocks as dangersimpson rightly points out, are effectively a subset of the Value category. High yielding stocks do outperform the market, but (arguably) no more so than low PE, low PB etc stocks. After all a high yield is just the reciprocal of the P/D which sits in that family.  So in our StockRanks framework a Dividend Rank would be an outlier... standing on it's own outside the QVM framework. 

On the other hand, the investment community splits down the line into 'capital growth' investors, and 'income' investors. Why? Due primarily to tax structures/incentives - there are differing tax rates on income and CGT. Also, frankly, most of our subscribers are either nearing, or in retirement so income is v. important. So I can 100% understand why so many investors would like us to launch a dividend rank.

But what is or should be a dividend rank? It's not as simple as a:

  • Value Rank  - where you just want to find cheap stocks
  • or a Quality rank  - which aims to find good stocks
  • or a Momentum rank  - improving stocks...

A good dividend rank is really a blend of all 3 really. Investors would probably want a dividend rank that incorporated the following:

  • Dividend Yield
  • Dividend Growth
  • Dividend Stability
  • Dividend Safety  (Cover, F-Score)

I could make the argument that in many ways that looks like a QVM+Yield strategy to me.  Isn't that really a stock screen rather than a ranking ?  What would a high Dividend rank really be highlighting?  I would imagine that you'd get as many safe companies ranking highly with low 1% yield as you would high yielding companies with more moderate quality aspects.  

Also, to date, we've tried to make the StockRanks entirely inclusive to date - where every single stock that trades in the market receives a ranking.  A dividend rank, by definition, would only rank about 50% of the market (as most don't pay dividends).  So what would zero or 100 mean in that context?  Is zero bad even if those stocks are  paying dividends - surely they are better (from a dividend perspective) than non paying stocks.   ?     

These puzzles we haven't quite fully worked out - so remain undecided, but I will say that Alex Naamani who works with me is super-eager to add a Dividend Rank so it's not something that is undiscussed in the office.  We'll add this chat to our Ranking discussion documents.

I hope my late night,  whisky fuelled, ramble makes sense to those reading.   We have a lot of work to do all round.... I do wish every idea we had we could just build overnight, but there's something about a steady cautious pace of development that makes you really think every feature through fully before releasing them. 

It's easy to add features, but very hard to take them away without getting into a lot of trouble from subscribers. Thanks for listening and please do keep the comments coming.


Blog: Follow @edcroft on Twitter
| Link | Share
Mechanical Bull 22nd Aug '14 18 of 21

If anyone is interested I've posted a entry on my blog, which explores some of these themes:

It describes a ten year backtest of QVM metrics from Sharelock Holmes. The main conclusions:
- The top decile QVM ranked stocks clearly outperformed
- Momentum works well in the short term but returns decline the longer you hold
- Quality gives good and consistent returns no matter how long the holding period
- Value will tend to disappoint (although it will still beat the market).

Blog: Mechanical Bull Blog
| Link | Share | 1 reply
iwright7 23rd Aug '14 19 of 21

In reply to Mechanical Bull, post #18


A very interesting piece of back testing QVM type data. The linear Momentum decline over 5 years surprises me - Has this set been rebalenced quarterly, or is it a factor of stock market weakness going further back over time. Does appear that quality shines.


| Link | Share
Mechanical Bull 23rd Aug '14 20 of 21


Thanks for the feedback. The graphs show the the average return by quarter over different holding periods from 3 months to 5 years. For example, the momentum figure for 5 years is calculated like this:

The average of:

2003 q3: Top decile momentum stocks bought and held for 5 years and
2003 q4: Top decile momentum stocks bought and held for 5 years and
2009 q2 Top decile momentum stocks bought and held for 5 years

This is far as you can go for now because any stocks bought in 2009 q3 (30 September) will not have been held for 5 years until 30 September 2014.

For momentum you would expect an inverse relationship between holding period and time. Not many stocks keep going up for five years.


Blog: Mechanical Bull Blog
| Link | Share | 1 reply
Edward Croft 23rd Aug '14 21 of 21

In reply to Mechanical Bull, post #20

Actually there's an inverse momentum relationship after between 3 - 5 years of out or underperformance. i.e. when stocks bomb for 3 years, they often recover, while stocks that rock for 3 years tend to pause. There's a bunch of research on this phenomenon - and it's highly correlated with the value effect. We've blogged on it somewhere but don't have time to dig it out just now.

Blog: Follow @edcroft on Twitter
| Link | Share

What's your view on this article? Log In to Comment Now

You can track all @StockoChat comments via Twitter

About Edward Croft

Edward Croft

CEO at Stockopedia where I weave code, prose and investing strategies to help investors beat the stock markets. I've a background in the City and asset management but now am more interested in building great stock selection tools for the use of investors online.   Traditionally investors online have had very poor access to the best statistics, analytics and strategies for the stock market and our aim is to set that straight.  High Quality fundamental information has been prohibitively expensive in the past and often annoyingly dull. People these days don't just want to know the PE Ratio and look at a balance sheet. They expect a layer of interpretation over data, signal from noise and the ability to know at a glance whether a stock is worth investigating or not. All this is possible using great design and the insights gleaned from quantitative research.  Stockopedia is where we try to make it happen ! more »


Stock Picking Tutorial Centre

Let’s get you setup so you get the most out of our service
Done, Let's add some stocks
Brilliant - You've created a folio! Now let's add some stocks to it.

  • Apple (AAPL)

  • Shell (RDSA)

  • Twitter (TWTR)

  • Volkswagon AG (VOK)

  • McDonalds (MCD)

  • Vodafone (VOD)

  • Barratt Homes (BDEV)

  • Microsoft (MSFT)

  • Tesco (TSCO)
Save and show me my analysis