Using Momentum, Quality, Value and Risk to evaluate UK stocks

The four composite scores described in this article are calculated for over 600 UK stocks across this site, all for FREE. I am currently expanding the UK stock universe over the next week to include 150 more AIM stocks.

Stock evaluation

When trying to evaluate a stock, one of the first available sources of information is the company’s own published reports containing all the reported metrics and derived fundamentals such as P/E etc. These are backwards looking performance indicators and as such are not always a good guide to how the stock will perform in the future. But it’s quick, cheap and for an established large company the best guide we have to make a value judgement about the core business. When we combine this with more forward looking company news, analyst predictions, director dealings, short positions and macro economy trends we can start to make a more informed decision about where our hard earned cash will find a good home.

Why a scoring system

Reading large amounts of company information and holding core fundamental values in our heads for multiple companies is not trivial. For scanning a large cohort of companies we need a short cut to enable us to quickly evaluate a company from a distance before getting into the details. By condensing the information into just four scores we can quickly determine where the strengths and weaknesses potentially lie within a company. We can also combine them with additional summary metrics like Price Earnings (P/E) and Debt to Equity (D/E) ratios to give a richer picture. Then if we deem it to be still of interest we have a heads up on where our attention should be directed for further investigation. A low score does not immediately mean that we should dismiss a share. For example a low ‘value’ score may mean that the business is rapidly expanding with nailed on future growth, and is priced accordingly. A good example at the moment would be Rolls Royce. You probably don’t want to pay top price, but you may want to place it on your buy list and then wait until it goes on sale.

Why composite scores beat single measures

Financial ratios are great except when they are not. They all have blind spots and by combining several ratios and producing a composite score we are not fooled by outlier data. An example would be a P/E score where the company books a one off £50M gain and their P/E drops from 12 to 4. A composite is harder to fool — a stock has to look cheap on several independent bases to earn a high Value score, and pairing it with Quality and Risk filters out the cheap-for-a-reason names.

Where you will see the scores

Most of this article talks about screening, but the four scores are not confined to the screener page. The same Momentum, Quality, Value and Risk scores appear on every company page, next to each stock on your watchlist, and alongside the stories on the High Impact RNS page. So a little time spent getting a feel for what the numbers mean pays off across the whole site, whether you arrive at a company through a screen, a news story or a search.

Complementary measures

The four scores condense a great deal of information, and no summary is perfect, so it is worth reading them alongside a few standalone measures. The Piotroski F-Score (0-9) simply counts how many of nine year on year fundamental improvements a company has managed, which makes it a good “is this business actually getting better or worse?” check on anything the Value score flags up. Altman Z is the raw financial distress number that sits behind the Risk score, and it is useful when you want to see the actual margin of safety rather than a ranking. Revenue growth separates the cheap-and-shrinking companies from the cheap-and-growing ones. Dividend yield read together with free cash flow cover tells you whether an income stock can actually afford the payout it is promising. And market cap is worth watching in its own right, because below the small cap line the spreads widen and liquidity becomes a real risk. All of these are available as columns on the screener — market cap and dividend yield are shown by default, and the rest (Piotroski, Altman Z and more) can be switched on from the Settings page.

Settings for different stock types

On the screener you can set floors and ceilings on each score, and different styles of investing call for quite different settings. Here are some sensible starting points.

• Quality compounders: try Quality of 7 or more with Risk of 4 or less, and ignore Value altogether. You will pay up for these companies; the bet is that their consistency persists for longer than the market has priced in.

• Recovery and deep value: try Value of 7 or more, a Piotroski score of 6 or more, and Risk no higher than 6. The F-Score is doing the real work here because it separates the businesses that are genuinely recovering from the ones that are quietly melting away.

• Momentum with a safety net: try Momentum of 8 or more with Risk of 5 or less. A pure momentum list quickly fills up with speculative small caps, and the risk cap removes most of them.

• Growth at a reasonable price (GARP): try Quality of 6 or more, Value of 5 or more and Momentum of 6 or more. This is the “nothing terrible anywhere” screen. You get shorter lists, but fewer nasty surprises.

• Income: try a dividend yield of 4% or more with Quality of at least 5 and Risk of 5 or less. The quality and risk floors are what keep you out of the yield traps.

Pitfalls to watch

There are a few traps worth knowing about. The first is sector blindness. Banks and insurers look “cheap” on P/E and price-to-book almost by definition, so compare financials with financials and read the Value score alongside the sector column rather than on its own. The second is missing data. A blank score is not a bad score; small caps and recent listings often simply lack the history needed to calculate momentum or multi year medians, so don’t let a sorted column quietly bury them. The third is that everything here is backward looking. The scores describe the company that filed the last set of accounts, not the one that issued a profit warning this morning, and a screen run the day after results can be a year out of date in spirit. Finally there is over-filtering. Stack six strict filters on top of each other and you will end up with three stocks, all of them flattered by some accounting quirk. Looser filters plus your own judgement beats a screen tightened until only the anomalies survive.

After the screen: the next layer of checks

A screen produces candidates, not decisions. For each stock that makes your shortlist it is worth reading the last two or three RNS statements (especially the outlook paragraphs), checking the recent news flow for anything the accounts haven’t caught up with yet, looking at the fair value comparison against sector peers, glancing at the analyst coverage and how the estimates have been moving, and checking the short interest, because a heavily shorted “bargain” deserves some extra scepticism. All of these checks can be done from the company’s own page on the site. Then read the actual report and accounts of anything that survives. The screener’s job is simply to make sure that the two hours you spend reading are spent on the right company.

Other ways in

Screening is only one funnel among several. Regulatory news itself is another; unexpected earnings upgrades and contract wins often appear in the news months before they show up in any screen, and that is exactly what the High Impact RNS page is for. The trending and most viewed lists show you where other people’s attention is flowing, and the sector heatmap shows where the money is. The best ideas often come from crossing two funnels. A stock that pops up on a news driven list and already screens well is telling you the same story twice.

Appendix: how the summaries are calculated

This is the technical bit and probably not that important to understand in detail. Each summary score is often made of several smaller scores each measuring an aspect of the variable.

Momentum (1-10)

Momentum score is quite simple, it ignores the last three months of price data. It chops off that data and then calculates the change over the previous 12 month period. So its asking how did this stock perform 12 months ago compared with 3 months ago. Skipping the most recent period avoids scoring a stock highly just before the mean reversion kicks in. This is one of the most studied anomalies in finance. This exact “12-month minus recent month” momentum factor is what Carhart (1997) added to the Fama-French model as UMD (“up minus down”), and it’s still one of the four standard factors used in academic asset pricing today. The price changes are calculated as percentages and then ranked into 10 bins, bin 10 containing those shares with greatest price percentage changes in that period.

Value (0-10)

The Value score asks a simple question: how cheap does this stock look? Rather than relying on a single ratio it looks through up to six different lenses: forward price to earnings, price to book, price to sales, free cash flow yield, dividend yield and a growth adjusted measure (PEGY). Each lens that is available for a company is scored and the results are averaged, so a loss-maker with no P/E is not unfairly punished. More importantly, a stock can’t look cheap just because one distorted ratio says it is. The forward P/E uses analyst estimates rather than the statutory earnings, so a one off write-down can’t manufacture phantom cheapness.

Quality (0-10)

The Quality score rewards businesses that earn high returns on the money they employ, and have done so consistently. It scores return on capital, return on equity, and the gross, operating and cash margins. Points are awarded both for clearing an absolute bar (for example ROIC above 10%) and for being at or above the company’s own multi year median. That second test matters because plenty of companies post one great year at the top of a cycle. Quality is trying to find the companies for whom great years are normal.

Risk (1-10, higher = riskier)

The Risk score works the other way around: a higher number means a riskier share. It is trying to estimate the chance of a serious permanent loss of your capital, and it blends balance sheet strength (Altman Z type distress measures), leverage and interest cover, and the volatility of the share price itself. Importantly it is sector aware. Banks, insurers, investment trusts and asset heavy businesses like utilities are scored with models suited to their balance sheets, because a bank measured like a manufacturer always looks bankrupt.