Systematic & Quantitative Valuation
Quantitative Value: A Practitioner's Guide to Automating Intelligent Investment and Eliminating Behavioral Errors
Gray and Carlisle turn value investing into a fixed pipeline a computer can run: first remove companies that may be cooking the books or heading for distress, then keep the cheapest on EBIT to enterprise value, then pick the best quality among those.
The big picture
The book starts from a finding that runs through decades of research: simple statistical models tend to beat expert judgment, including the judgment of experts who are given the model's output as a starting point. The authors, then a finance PhD (Gray) and a former corporate lawyer turned fund manager (Carlisle), apply that finding to value investing. Their answer is a process with no room for a story: every step is a rule, every rule is tested on historical US data, and the investor's job is to follow the rules through the years when they lag.
The process has an order. Step one avoids permanent loss of capital: companies with signs of earnings manipulation, fraud or financial distress are removed before price is even looked at. Step two finds the cheapest names, and after testing a range of price ratios the authors settle on EBIT divided by total enterprise value. Step three sorts the cheap survivors by quality, split into franchise power (high, stable returns and margins over many years) and financial strength (a checklist adapted from Joseph Piotroski's F-score). Why it matters now: in a market where AI narratives move prices faster than filings, a pipeline that reads the filings first and the price second is a useful counterweight.
The 3 strategic pillars
-
Avoid permanent loss first
The worst outcome for a value investor is not a cheap stock that stays cheap; it is a cheap stock whose accounts were not real or whose balance sheet fails. Screen those out before ranking anything.
The book removes names with high accruals (reported profit far ahead of cash flow), a high probability of manipulation from the Beneish M-score (converted to a probability with the normal distribution), and a high probability of financial distress. Its distress model is the Campbell-Hilscher-Szilagyi measure; the classic Altman Z-score is the accessible stand-in used in the pack.
-
Cheapness as EBIT / TEV
Of the price ratios the authors test, operating profit against the price of the whole enterprise sorts stocks best in their sample. Debt and cash are inside the denominator, so leverage cannot make a stock look cheaper than it is.
TEV = market cap + total debt + preferred stock + minority interest − cash. Rank the screened universe by EBIT / TEV and keep the top slice (the book works with the cheapest decile). In the authors' tests, adding Greenblatt's return-on-capital rank to cheapness did not improve on cheapness alone.
-
Quality: franchise plus financial strength
Among cheap stocks, prefer those with evidence of a durable business and a balance sheet that is getting stronger, not weaker.
Franchise power uses multi-year averages: long-run return on assets and on capital, margin growth and margin stability. Financial strength is a ten-point binary checklist on current profitability, balance-sheet stability and recent operating improvement, built on Piotroski's nine signals. The two halves are combined into one quality rank that decides the final basket.
What a Closelooknet reader does with it
The working use is a fixed order of operations for any list of candidates: forensic screens, then price, then quality, and never in reverse. The main mistake it prevents is the value trap with a hidden cause, where a stock looks cheap because the market has already priced in accounting problems or a debt problem that the reader has not yet found. A second mistake is the override: seeing a model's output and adjusting it by feel. The book's argument, backed by the forecasting research it cites, is that most overrides reduce accuracy, so the pipeline is written down in advance and the checklist output is recorded rather than second-guessed.
The bridge to the Closelooknet approach
The book sits next to two entries already on the shelf. The Little Book That Still Beats the Market ranks on earnings yield and return on capital; Gray and Carlisle keep the same cheapness measure, test the quality half, and put a forensic filter in front of both. Value Investing: From Graham to Buffett and Beyond explains why a high return on capital lasts only behind a barrier to entry, which is what the book's franchise-power measures try to detect from multi-year data. The behavioral case for a rules-based process is covered in Misbehaving. On Closelooknet, the Company Scoring System shows Altman Z, Piotroski F and Beneish M with their standard zones and keeps the M-score out of the composite, the same separation of forensic flags from ranking that the book uses. The accruals ratio and earnings yield entries cover the two inputs the pipeline leans on most.
Action-Kit — from theory to practice
Tooling & data
| What you need | Where to get it | Cost |
|---|---|---|
| Two years of full statements per company Receivables, sales, COGS, SG&A, depreciation, PP&E, current assets and liabilities, debt, net income and operating cash flow for the M-score, Z-score and F-score | SEC EDGAR full-text search and company filings Both years must come from the same reporting basis; restated prior years are the right comparison. | Free |
| Standardized fundamentals and market data Market cap, debt, cash, EBIT and the balance-sheet lines in one table for building the input CSV | stockanalysis.com (financials tab) The free tier covers most US and many non-US listings; Pro adds longer histories and exports. | Freemium |
| Screener with EV/EBIT and F-score filters Run the cheapness and quality steps over a whole market before pulling full statements for the shortlist | Stock Rover Piotroski F and Altman Z appear as metric columns (some on paid tiers); check them against the pack's own calculation on a few names. | Freemium |
The formulas
-
Altman Z-score (original, 1968, manufacturers)
Z = 1.2·X1 + 1.4·X2 + 3.3·X3 + 0.6·X4 + 1.0·X5- X1 = working capital / total assets
- X2 = retained earnings / total assets
- X3 = EBIT / total assets
- X4 = market value of equity / total liabilities
- X5 = sales / total assets
Zones: above 2.99 safe, 1.81–2.99 grey, below 1.81 distress. For non-manufacturers use Z'' = 6.56·X1 + 3.26·X2 + 6.72·X3 + 1.05·X4' (X4' = book equity / total liabilities; zones above 2.60 / 1.10–2.60 / below 1.10).
-
Beneish M-score (8 variables)
M = −4.84 + 0.920·DSRI + 0.528·GMI + 0.404·AQI + 0.892·SGI + 0.115·DEPI − 0.172·SGAI + 4.679·TATA − 0.327·LVGI- DSRI = (receivables/sales)_t ÷ (receivables/sales)_t−1
- GMI = gross margin_t−1 ÷ gross margin_t; SGI = sales_t ÷ sales_t−1
- AQI = [1 − (current assets + PP&E)/total assets]_t ÷ same_t−1
- DEPI = [dep/(dep + PP&E)]_t−1 ÷ [dep/(dep + PP&E)]_t; SGAI = (SG&A/sales)_t ÷ (SG&A/sales)_t−1
- LVGI = [(current liabilities + LT debt)/total assets]_t ÷ same_t−1; TATA = (net income − operating cash flow)_t / total assets_t
Above −1.78 (Beneish's cutoff) flags a manipulation profile; −2.22 is a stricter cutoff in common use. The book converts M to a probability: PROBM = N(M), the standard normal CDF.
-
EBIT / TEV
EBIT / TEV, TEV = market cap + total debt + preferred + minority interest − cash- EBIT — trailing operating income
- Market cap, debt, preferred stock, minority interest, cash
Higher is cheaper. Applied only to names that pass the forensic screens.
-
Piotroski F-score (the financial-strength base)
F = Σ of 9 binary signals (1 = pass, 0 = fail), range 0–9- Profitability: ROA > 0; CFO > 0; ΔROA > 0; CFO/TA > ROA (cash beats accruals)
- Balance sheet: Δ(LT debt/TA) < 0; Δcurrent ratio > 0; no new shares issued
- Operations: Δgross margin > 0; Δasset turnover > 0
8–9 is strong, 0–2 weak. The book's own FS score uses ten similar signals, including free cash flow and net equity issuance.
Applied Pack · free members
Quantitative Value Applied Pack
The book's three-step pipeline as a working Excel checker and a Python screener: manipulation and distress screens, EBIT/TEV cheapness, then financial strength, run on your own two-year fundamentals.
- QuantitativeValue_Checker.xlsx — Altman Z and Z'', Beneish M-score with PROBM, EBIT/TEV and Piotroski F-score per company from two years of inputs, with a pipeline sheet that applies the screens and ranks the survivors; live formulas, amber input cells
- graycarlisle_pipeline.py — stdlib-only script: two-year fundamentals CSV in, table out with Z, M, PROBM, EBIT/TEV, F-score, screen verdict and final rank
- fundamentals_sample.csv — example input with EXAMPLE_ rows showing the expected columns
- README.txt — inputs, thresholds, how to run, and the educational-use disclaimer
Pack security
Macro-free Excel · plain-text Python you can read before you run it · no installers, no network access — the code works only on files you provide. Served only from closelook.net; we never distribute through download portals or email attachments. How to verify in 30 seconds →
SHA-256 1941ba940d6d2a2576d356a8e91adf3d9e2946c81eada038a7c40c918bf4e064
The pack is a free-membership download — no card, free forever. Membership also delivers the Daily Pulse and Weekly Signal to your inbox.
Join the Look — free → Already a member? Sign in →
Educational templates — a research diary companion, not investment advice.
Closelooknet publishes a market diary, not investment advice. This condensed read restates the book's ideas in our own words for education — for the author's full argument, go to the source.