Rapid ranking changes

-->

26th Jun '19 by John Brown 23

9 comments 1870 reads

Stockopedia ranking provides a wealth of great information, but I am concerned at the rapid changes in the ranking of some shares recently, e.g. Babcock down from 72 in March to 43 today; Easyjet from 73 to 39, and BT from 98 to 54. It is hard to believe that the fundamental position of such large companies can change so much in so short a time. This would suggest that the algorithims on which the ratings are based are too short term and do not look closely enough at the long term fundamentals - not a conclusion with which I am comfortable.

Unlock with Email

Unlock with Google

Already have an account?
Login here

About the Author

John Brown 23

Premium Member

9 comments

Nick Ray

26th Jun '19

With BT and Babcock I think the fall has been largely due to falls in Broker forecasts for EPS and DPS over the last few months.

For Easyjet, the fall is linked to a drop in Piotroski F score from 7 to 4 which has dragged the Quality Rank down. I think this might be related to changes in long term debt/assets recently.

John Brown 23

27th Jun '19

Nick, thank you, that is a helpful explanation regarding these three. At the same time, it is worrying when such large companies are being highly rated (and it's not so long since they were rated even higher), despite poor news on the relatively short term horizon.

Edward Croft

27th Jun '19

The StockRanks are based on fundamentals (quality), valuation (value) and sentiment (momentum). As these change in the normal course of results and trading, the ranks will shift. Many British companies report on an interim basis at the end and half way through the year. The end of year results come through at the end of March - or often through April for the year end, and that's a big trigger in terms of raw fundamentals and then the impact on valuation. Once that's all been processed - you start seeing the brokerage community updating their forecasts - again a big shift. So you often see these drifts in ranking.

For example - BT released new results and they were processed in April.

The Quality Rank has taken a big dive from the 80s to the 50s - partly because the data is based on 5 year averages and BT was doing far better 5 years ago than it is today... as the old data has moved out of the average range, the 5 year ROCE and 5 year Net Margins have dropped. Those are key components of the franchise factor of the Quality Rank.
The Momentum Rank has plummeted down to 27 from 81. This can be the most flighty of the ranks as it assesses price and earnings sentiment. Since the results the brokerage community have cut forecasts by 5% and the price has pulled back so that 6 month relative strength is very poor against the market.

These are factors that matter for future stock returns. BT may do well in future, but there are solid reasons why it should be ranked averagely against the rest of the market. You can see the last couple of months of StockRank drift below.

5d14b70b57750Image_2019-06-27_at_1.30.59

This is a picture taken from the new app - to which everyone will receive an invitation in due course.

Gromley

27th Jun '19

That's an interesting observation.

I had noted that for both BT Group and Easyjet (LON:EZJ) the biggest movement was in the Quality metric.

As you mention some of the metrics making up the QR are five year views and here as you indicate the difference between Year minus 1 and Year minus 6 can make quite a difference, but of course only impacting 1 year out of 5 of the average. So I'm surprised if the 5 year elements made such a big impact?

Just at first glance it's not easy to see which of the quality measures for BT Group will have fallen significantly.

For Easyjet (LON:EZJ) it is perhaps a little clearer in that the Piotroski F-Score falling from 7 in March to only 4 now will have had a big impact.

I would have to dig quite a bit further to establish what element of the F-Score have worsened although I code hazard a guess given that most of the F-Score measures compare the year just gone with the prior year. IE it is to a large degree a one year measure.

This is the puzzle for me, conceptually i have the view that Quality should be quite persistent (barring genuinely dramatic changes to a business) but the Quality Rank and most particularly the F-score seem to be pretty volatile measures.

Now I know that you have gone through rigorous testing in developing the SR system and I'm not questioning the effectiveness of the Ranks as such, but for me the Quality Rank often does not seem to measure what I would personally want it to measure. I've been looking at the possibility of creating my own variant (although if that degree of customisation were available on Stocko that would be even better!!

One other 'issue' I have encountered with the F-Score is the use of statutory rather than adjusted results. I know there is a massive issue of subjectivity when it comes to adjusted numbers, but ignoring all of the adjustments can give some "interesting" results.

One of my holdings £RCH has an F-score of 4 or 5 (There is a long standing discrepancy between the headline score of 4 and the 5 actual "passes" in the detail).

Of the 4 (underlying) F-score fails.

One I would say is real, "reducing margins".

Two are the direct result of taking a significant impairment charge on goodwill in the last FY. As a result the company "is not making a profit" and "is less profitable than last year". Both assertions I would argue are misleading - even if you accept the impairment as a "real" cost, it is only accounting convention that chooses to apply it all to one year rather than spreading it across all of the years to which it truly applies.
The final fail is "Is the company's long term debt reducing or stable?" Technically this is true, but the fact is that Reach (LON:RCH) raised new debt last year to make an acquisition (agreed acquisitions can more often than not be destructive) but the cash-flow is strong that this debt will be eliminated in very short order.

The fact here is (barring anything left field) that once Reach (LON:RCH) publish their next results (not sure if it is the FY or HY figures that drive the calculation here) the F-Score will leap from 4(5) to 7 (or maybe even the golden 8 if mix changes improve the Gross Margin) but in reality the business will not have changed much at all.

Now obviously you have lots of evidence that the Quality Rank is meaningful and predictive of prospects, but I really wonder if there is something even more powerful lurking just below the surface?

Nick Ray

27th Jun '19

Now I know that you have gone through rigorous testing in developing the SR system and I'm not questioning the effectiveness of the Ranks as such, but for me the Quality Rank often does not seem to measure what I would personally want it to measure. I've been looking at the possibility of creating my own variant (although if that degree of customisation were available on Stocko that would be even better!!

I agree with this sentiment. The Quality Rank combines some metrics from the "Profitability" category with some from the "Financial Strength" category thereby slightly confusing what QR might be telling you. This matters somewhat because it is averaging the ranking of these metrics. So if the metrics are measuring different things then we are averaging apples and oranges.

Having said that, averaging ranks does work very well in giving a blended view of what some underlying probability distribution might look like and handles the odd misbehaving metric better than a simply yes/no screener. So I would love to see a DIY version of the ranks system where we choose the metrics to be blended (and maybe weight them too). Then we have no-one to blame but ourselves on the choices we make!

Edward Croft

27th Jun '19

Hi Gromley - really great comments.

Now I know that you have gone through rigorous testing in developing the SR system and I'm not questioning the effectiveness of the Ranks as such, but for me the Quality Rank often does not seem to measure what I would personally want it to measure. I've been looking at the possibility of creating my own variant (although if that degree of customisation were available on Stocko that would be even better!!

Yes, I do understand this. The "Franchise Factor" - which is the more "Buffett-esque" part of the Quality Rank is perhaps 50-60% of the overall weighting. The fundamental trend component (F-Score) is 30-40% which is more volatile year to year. There are many who just want access to the Franchise Factor Rank - even though it's actually less predictive of returns - it's perhaps a better proxy for finding long term buy and hold compounders.

Out of interest, we've spent considerable time building a new data system that is set to go live later in the year. It allows the generation of custom ranking systems, and also the absolute look through auditability of all rank components both today and historically. I can't promise when this will be 'productised' - we have to consider how it fits in with the offer, and at what price etc... but good things coming here.

First things first - we'll be shipping the new interface (which is mobile and fresh) and then we'll be shipping the new data systems later in the year. Lots to look forwards to.

mmarkkj777

27th Jun '19

Great comments on this thread (and great replies).

My time has been very limited recently (for non-market related reasons) but I'm especially interested in Franchise/Brand factor for long term holding.

I'm also very interested in the app Ed. Looking forward to that going live.

Hopeing to be more active again shortly.

Gromley

27th Jun '19

Thanks Ed,

I certainly look forward to seeing the new developments. (Although changes to the UI would be pretty low down on my priority list and I almost universally loathe those sites that have become more "mobile friendly" - so I'll be a tough sell on the UI changes!)

But back to the QR : I wasn't so much wanting just to home in on the franchise factors I do actually agree and value the idea of the fundamental momentum element , but my perception is that the "F-Score" is too volatile (as opposed to fast moving).

I Don't know whether I am objecting to : JP's original work, your rendering of it, or reporting standards changes that have developed the way it should apply. I just seem to see too many exceptions where it does not (imho) make sense. Exceptions of course do not necessarily invalidate it's statistical effectiveness , but as I say I cannot get away from feeling that there is something better lurking a small way below the surface. Any tool therefore that gives the ability to riff against the underlying Piotroski melody would therefore be great.

To continue the musical allusion, I have to confess to never being able to resist the temptation to take the precious things we love apart to see how they work.

Looking forwards to the new toys in due course :-)

John Brown 23

2nd Jul '19

Ed, Nick and Gromley, many thanks for your valuable comments and this very interesting discussion. I am clearly not so knowledgeable on the background to the rankings, but I particularly share Gromley's concern at rapid movement of the QR in the absence of a significant fundamental change in the position of a company; similarly a drop in MR from 81 to 27 following a fall in forecast price of just 5%. Thinking in Warren Buffet terms, either the quality of a company is good, and subject to value and/or momentum it is worth investing in, or the quality is poor and you stay clear. It does not suddenly change. As pointed out, the difference between year -1 and year -5 is only 1/5 of the average, and for a company like BT to go from 98 super stock or high flyer, (well worth investing in) to 54 (stay clear) in less than 3 months is problematic. Perhaps I am being too simplistic.

Unlock the metrics for all these stocks and more...

Take a free trial

Facebook Twitter Youtube Instagram RSS

This site cannot substitute for professional investment advice or independent factual verification. To use it, you must accept our Terms of Use, Privacy and Disclaimer policies.

Unlock the metrics for all these stocks and more...

Take a free trial

Rapid ranking changes

Unlock the rest of this article with a 14 day trial

About the Author

John Brown 23

9 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...