The New StockRanks Portal Is Now LIVE

10th Feb '15 by Edward Croft

43 comments 24779 reads

Thanks to everyone who listened into the webinar just now - this is a quick announcement to say the StockRanks tab is now live... please have a look in the main site navigation bar where you'll see a 'ranks' tab... there's a super StockRank browser, a 'movers' page and a whole performance history portal.

We'll be improving these features over the next couple of weeks... for now we want as much feedback as possible so please do let us know any issues using the feedback button to the right. We'll try and get the webinar live as soon as we can for anyone who missed it, but here's a quick summary of the new features....

StockRanks Browser

We've made it incredibly easy to filter the market by our growing variety of ranking factors... across quality, momentum, value and even growth.

You can add one of ten sector filters or drill down using a range of size filters. With about 20,000 shares in the system it's easier than ever to find candidate stocks without fiddling around in the screener.

There's even an option to load your settings in the screener for further filtering - for example if you'd like to add an additional rule (like PEG, Yield etc).

StockRanks Performance

The StockRank Performance Portal allows the building of a range of stockrank tests across European and UK shares (US data is coming shortly). You can check out how different buckets of stock ranks have performed against each other across the different ranks - with different rebalancing periods, with different market cap filters and so on. We are still working on speed, but as an example here's the European StockRank performance since launch in June 2014 rebalanced quarterly. A stunning 46% gain for the top StockRank bucket.

54da0fcc61d35StockRanks_QVM_Performance_

StockRanks Movers

You can also find the stocks moving the most across the StockRanks in recent weeks using the 'movers' tab. We've tried to make it really simple to find shares that are rising up in the ranks (or down). Someone asked in the webinar if we could send alerts based on the rank changes - not yet, but we are indeed working on it. One thing at a time !

54da104c276cbStockRank_Movers_-_QVM_upgr

I think these features make a massive improvement to the toolbox on the website. For too long it's been…

Unlock with Email

Unlock with Google

Already have an account?
Login here

PhilHofGrowYourDough

12th Feb '15

Hi Ed,

My QMPlus screen (where the Plus is the growth parameters I require) is currently highlighting (in order of increasing risk adjusted PEG):

EPA:INF - a French IT software/service provider
HEL:SUY1V - A Finnish paper wipe manufacturer
Data Modul AG Produktion und Vertrieb von Elektronischen Systemen (ETR:DAM) - A German flatbed TV manufacturer/distributor (which I hold but has disappointingly been bought out)
Cabot Microelectronics (NSQ:CCMP) - A US polishing pad supplier for microelectronics industry - although with a high PE and a high PEGR (0.94) I'm less attracted to this one

QMPlus type stocks I currently hold include:

Lam Research (NSQ:LRCX) - A US provider of wafer manufacturing equipment (-4%)
Character (LON:CCT) - A UK based toy distributor (+11%)
CellaVision AB (STO:CEVI) - A Swedish based medical technology company (+132%)
Sprue Aegis (LON:SPRP) - A UK based smoke/CO2 alarm manufacturer and distributor (+310%)
Cancom SE (ETR:COK) - A German provider of IT software/services (+0%)
ETR:GFT - A German provider of IT solutions (+93%)
Southwest Airlines Co (NYQ:LUV) - A US airline (+5%)
Adesso AG (ETR:ADN1) - A German provider of IT solutions (-12%)

You can see all of my current holdings including my other SuperStocks in my fantasy fund ... http://www.stockopedia.com/fantasy-funds/philhs-fund-4060/

Regards
Phil

pka

Ed, Is it meaningful to calculate Value Ranks for infrastructure investment trusts? I ask because HICL Infrastructure Fund has a Value Rank of 77 at present, which suggests it is good value, but it is on a 15% premium to its Net Asset Value at present according to Hargreaves Lansdowne's web site:

http://www.hl.co.uk/shares/shares-search-results/h/hicl-infrastructure-co-ord-gbp-0.0001

Traditionally investments trusts on significant premiums to their Net Asset Values are usually considered to be poor value.

Mechanical Bull

Ed

I wonder if Paul may have a point. Both Wincanton and Hogg Robinson have negative book values, and their price to book value shows up as "n/a". PBV is one of the six value factors underpinning the stock rank so how is a negative PBV stock treated exactly? Is is possible that there is some kind of mathematical anomaly that is inflating the value rank of negative PBV stocks?

MB

Edward Croft

If companies have NULLs for any ratio the ratio is excluded from the Value Rank calculation. This is pretty standard practice amongst value ranking methodologies. We don't publish negative PBV so this won't impact it.

I do understand that using the Value Rank in isolation could be hairy. I certainly wouldn't!

shanklin100

13th Feb '15

Hi Ed

Very difficult to understand why you convert a negative TBV to null and then ignore it for value metric calculation purposes. Surely you could use the true -ve number in some way when calculating the V metric? As it is IMHO you are calculating excessively high V's for all such stocks.

HTH. Cheers, Martin

AnonymousUser24675

I have been a member for a few days, so forgive me if the following has already been discussed.

In your help texts related to the rankings, it is explained which factors are calculated into a weighted averaged composite score. Are the mentioned factors all weighted equally? If not, did you disclose the different weights somewhere for review?

More specifically, it could be that the Altman-Z score plays its role adequately in the Quality Ranking. But its influence may be diluted too much when it comes to the QVM Ranking.

Q+V+M each equally weighted in the StockRank.

For the Value Rank everything is equally weighted where available. Quality Rank is quite a bit more complex. We can't disclose absolutely everything as the only way we could would be to open source our code.

Frankly - I'm all for transparency and this may be possible with the product goals we have here. The goal is to eventually let anybody create their own ranking systems and 'fork' our own - with our rankings being literally a starting point for anybody who wants to build on them. That might take a bit of time, but I'd be thrilled if we get to that point and some of our subscribers start generating more powerful ranking systems on subsets of our data.

I don't like 'proprietary' / opaque systems like Value Line in the US where you have no idea of the inputs and they rely on the mystique of 50 years performance history rather than full disclosure.

Hi Ed

The transparency is helpful because it means we can discuss the metrics.

However, it is extremely unhelpful to ignoreP/ TBV for any company for which it is negative as this means you give an excessive value rank to all such companies and by implication you give an excessively low value rank to all companies with positive P/TBV.

I can see no possible justification for the approach of ignoring -ve P/TBV. Whilst it would perhaps only have a modest impact on, for example, WIN's value metric, it would have the effect of reducing it slightly.

Thoughts welcome. Why is it ignored?

Cheers, Martin

P.S. Obviously in my earlier post, TBV should be replaced by P/TBV.

Can we clarify, does ignoring a n/a PBV mean either that ValueRank is:

1) balanced equally over the six factors but with 0 being input for PBV, meaning that the ValueScore is throttled by a sixth

OR

2) PBV is totally ignored and the Value Score only considers five inputs

If it's the former then the ValueRank is penalised for a negative PBV.

Thanks in advance
Phil

peterg

Can we clarify, does ignoring a n/a PBV mean either that ValueRank is:

1) balanced equally over the six factors but with 0 being input for PBV, meaning that the ValueScore is throttled by a sixth

2) PBV is totally ignored and the Value Score only considers five inputs

If it's the former then the ValueRank is penalised for a negative PBV

No, low price to book is a good thing in value screens, so if you use the former, a zero in rankings for stocks with negative PBV, you are effectively giving companies whose PBV is infinity an impossibly good, low, score.

Peter

Hi Peter,

You misunderstand.I believe that each parameter is given a score and then these scores are combined to create a ValueScore which is then ranked across the set to give a ValueRank.

So companies with a negative PBV would receive a score of 0 for that parameter, which would reduce the ValueScore and hence the ValueRank

I hope that helps
Phil

Low +ve P/TBV is better than high +ve P/TBV is better than -ve P/TBV.

High -ve P/TBV is better than low -ve P/TBV. For example, if xxx's SP is £1.00 and:
- P/TBV is -100, xxx has TBV of -1p/share
- P/TBV is -1, xxx has TBV of -£1.00/share

Everything else being equal, there is a 99p/share difference in how one would value company xxx. On this basis, I can see no justification for setting TBV to "Null" and just ignoring it.

Cheers, Martin

Phil

That is not how I read Ed's description of how a -ve TBV is treated in the V calculation. Perhaps wrongly, my understanding is your (2).

Cheers, Martin

Hi Martin,

Let's see if he clarifies it then.

Cheers
Phil

Reading the ValueRank help literature it says this ...

"Each company in the market is ranked from 1 to 100 for each of these value ratios and a composite score is calculated as the weighted average of all valid values."

So it'd appear to be option two as you suggested.

Hope that helps
Phil

vaticinator

It seems to me that the problem of using PBV in a calculation is that as the Book Value approaches zero, the PBV ratio approaches infinity?

Why not invert the ratio, Book Value/Price or BVP so we have a continuous metric which falls to zero and goes negative as book value falls through zero. Similar to earnings yield vs PE ratios.

Sorry if I have missed the point!

Interesting discussion chaps.

"Low +ve P/TBV is better than high +ve P/TBV is better than -ve P/TBV."

Is negative book value really an indication that a stock is 'expensive' ? There's really no justification for that if you study the research. A -ve BV is actually an indication of years of accrued losses and therefore a high distress risk. If you believe that the value premium is a payoff for taking distress risk then -ve BV stocks can actually enhance the value effect.

There's plenty of research that shows that Negative Book Value stocks with high distress risk actually significantlyoutperform - which is precisely the scenario we have seen with Wincanton. e.g. http://world-finance-conference.com/papers_wfc2/35... and in summary http://www.cfapubs.org/doi/pdf/10.2469/dig.v39.n2....

They observe that negative BE stocks with the highest default risk have higher returns than all positive BE stocks.

In the paper they actually add -ve BE stocks back into different valuation quantiles - rather than sticking them all in the highest growth (expensive) segment - classified by their historic return similarity.

They provide evidence that suggests that including these stocks in a value group enhances the value premium.

We've followed a lot of literature in our construction of the Value Rank and many will exclude all negative book value stocks from rankings - e.g. and most famously Fama & French who treat -ve BV stocks the same way as delistings. Our approach, being a blended scoring system, doesn't throw the baby out with the bathwater like Fama & French, but falls back on other value factors for classification. I think it's a more moderate and fair approach.

Obviously others will disagree, but that's what makes a market ! Hope that clarifies.

PS - we originally published -ve P/B on the stock reports... but removed it as about half a dozen subscribers were up in arms about it with very good arguments - it's very hard to please everybody!

Hi Ed

I look forward to hearing Paul's reply on this front to you :-)

I find it quite difficult to accept any suggestion that -ve TBV should be a positive when considering how Stockopedia calculates its Value metric. Whatever the studies you refer to, I think many of us probably take more of a PYAD stance when considering value.

Will look at your links next week and reply on those then.

Best wishes, Martin

P.S. As highlighted by vaticinator, perhaps it would be more helpful to present TBV/share divided by the share price, rather than P/TBV, as that would enable an easy visual comparison across all stocks, however you wish to use this number in the Value metric!!

Shanklin - I'm merely relaying the academic studies that have been done on the subject and the standards that have been applied elsewhere by greater minds than myself.

Once again, I really don't feel that we should lose ourselves in the semantics of what is or is not 'value' - I've already been over that debate extensively above. It's just a Value Rank - it's not the entirety of Value Investing in a single number - it's pure and simple a better measure of value than basic ratios like the P/E, the P/B and all the others.

Ben Graham started all this off with his Graham Multiplier - when he merged the P/B and P/E. We've merely done a more extensive job with a 6 factor model... there's exactly the same issues with this measure of 'cheapness' as you'd get with a simple P/E but it's got a better track record and is far more understandable for lay investors.

And of course... nobody needs to use it if they don't wish... but it's proved to be a remarkably good component of our StockRank system to date... even on it's own the cheapest 20% of the market has dramatically outperformed the most expensive quite consistently as can be shown in this chart.

54de3cc3f2bb2StockRanks_QVM_Performance_

Used in conjunction with other metrics it's done far, far better... the real message is - one shouldn't use any single factor on it's own... it's better to use multi-factor models.

red

re: negative P/B, this is a useful perspective
http://www.gurufocus.com/news/176639/is-negative-book-value-bad

The new StockRanks Portal is now LIVE

StockRanks Browser

StockRanks Performance

StockRanks Movers

Unlock the rest of this article with a 14 day trial

About the Author

Edward Croft

43 comments

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

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