slop-stuff / Quantitative
Finance — value investing cheatsheet
Buy businesses, not tickers — with explicit math.
Statements, FCFF/FCFE, ratio herding, DCF, MoS, EDGAR workflow, and AI-slop kill-list.
Value investing means estimating intrinsic value and buying only with an explicit margin of safety. This page herds the formulas, ratios, and process — not vibes.
Quick reference
| Thing | Formula / rule |
|---|---|
MoS | 1 − Price/IV — buy vs bottom of IV range, not a feeling |
EV bridge | MktCap + Debt + Pref + NCI − Cash (− non-ops) |
FCFF | EBIT(1−t)+D&A−Capex−ΔNWC → discount @ WACC → EV |
FCFE | FCFF−Int(1−t)+NetBorrow → discount @ r_e → Equity |
WACC | (E/V)r_e + (D/V)r_d(1−t) — market weights |
TV (Gordon) | FCFF_{n+1}/(WACC−g) — need WACC>g; g≲2–4% |
TV (exit) | EBITDA_{n+1}×ExitMult — back out implied g |
Equity/sh | (EV − NetDebt − Pref − NCI) / diluted shares |
| Cash ≠ EBITDA | Prefer CFO / FCF / FCFF — EBITDA skips capex, WC, tax, SBC |
Eight multiples to herd (always sector + cycle + quality):
| Multiple | Cheap context | Watch |
|---|---|---|
P/E, fwd P/E | vs peers / history | cyclicals, GAAP vs non-GAAP |
PEG | <1 rough | define growth period |
P/B | banks/asset-heavy | needs ROE/ROTE |
P/S | early / trough | ignores margins |
EV/EBITDA, EV/EBIT | cap-structure neutral | leases, SBC, capex |
EV/FCF, FCF yield | label equity vs firm | define FCF |
Mindset
Four ideas. Price is what you pay; value is what you get.
- Intrinsic value — PV of future cash the business will produce.
- Margin of safety — explicit % below estimated IV (sized to uncertainty).
- Mr. Market — daily quote you may ignore or exploit.
- Circle of competence — only businesses you can understand and value.
KEY: Think in decades. A share is ownership in a business you’d hold for ten years — not a ticket to flip on the next print.
Statements map
Three reports + notes. Read together; normalize before ratios.
| Statement | What it says | Herd |
|---|---|---|
| Income | Rev → COGS → GP → OpInc → NI | mix, one-offs, SBC |
| Balance | Assets = Liab + Equity | cash, AR, inv, debt, goodwill, NCI |
| Cash flow | Op / Inv / Fin | CFO vs NI; capex; buybacks |
| Equity stmt | Shares, OCI, NCI | dilution, RSU, buybacks |
Normalize before you trust a number:
| Adjust | Why |
|---|---|
| One-offs | Strip (carefully) for run-rate |
| SBC | Real cost — expense or dilution |
| Leases | Add for IFRS16 peer comps |
| Op vs non-op | Value non-ops separately in EV bridge |
| Diluted shares | Options/RSU treasury method |
| NCI / Pref | In EV bridge; don’t orphan them |
✓: Cash is fact, earnings are opinion. Want
FCF ≈ NI. Rising AR/inventory vs sales → accruals / channel risk.
Ratio tables
Each row: formula · meaning · use · pitfall. Prefer 5–10y history + peers.
Profitability
| Metric | Formula | Use | Pitfall |
|---|---|---|---|
| Gross margin | (Rev−COGS)/Rev | pricing power vs peers | mix / freight / reseller |
| Op margin | EBIT/Rev | core ops | restructuring, lease/SBC class |
| Net margin | NI/Rev | quick only | financing distorts — prefer FCF margin |
| ROE | NI/Avg equity | quality if sustained | leverage & buybacks inflate |
| ROA | NI/Avg assets | asset-heavy / banks | intensity differs by model |
| ROIC | NOPAT/IC · NOPAT=EBIT(1−t) | ROIC>WACC = value create | avg IC; goodwill consistency |
| FCF margin | FCF/Rev | cash conversion | lumpy capex; maint vs growth |
Leverage / liquidity
| Metric | Formula | Use | Pitfall |
|---|---|---|---|
| Current | CA/CL | near-term solvency | inventory-heavy mirage |
| Quick | (Cash+ST+AR)/CL | stress | AR quality |
| D/E | IB debt / Equity | gearing screen | prefer Net debt; hybrids |
| Interest cover | EBIT/Interest | debt cushion | cyclical EBIT; leases |
| Net debt/EBITDA | (Debt−cash)/EBITDA | credit comps | EBITDA≠cash; add leases |
Efficiency
| Metric | Formula | Use | Pitfall |
|---|---|---|---|
| Asset turnover | Rev/Avg assets | DuPont | model intensity |
| Inventory days | (Avg inv/COGS)×365 | mfg/retail | seasonality, write-downs |
| DSO | (Avg AR/Rev)×365 | collections | ↑↑ vs sales → stuffing |
| CCC | Inv days + DSO − DPO | WC quality | negative CCC ≠ always good |
Valuation
| Metric | Formula | Use | Pitfall |
|---|---|---|---|
| P/E | Price/EPS (diluted TTM/NTM) | peer + history | cycle peak EPS |
| Fwd P/E | Price/FY1 | growth bet | stale estimates |
| PEG | (P/E)/g% | growth-adj | define g |
| P/B | Price/BVPS | banks/asset-heavy | intangibles; needs ROE |
| P/S | MktCap/Rev | early / trough | ignores margins |
| EV/EBITDA | EV/EBITDA | firm comps | capex/SBC/leases |
| EV/EBIT | EV/EBIT | after D&A | still pre-reinvestment |
| EV/FCF | EV/FCFF | cash firm mult | define FCF |
| FCF yield | FCF/MktCap or FCFF/EV | owner screen | label equity vs firm |
| Earn. yield | EPS/Price or EBIT(1−t)/EV | vs bonds | earnings quality |
Quality
| Metric | Idea | Use | Pitfall |
|---|---|---|---|
| Accruals | (NI−CFO)/Avg assets | earnings vs cash | single-year noise |
| Owner earnings | NI + non-cash − maint. capex (−ΔWC) | Buffett cash base | maint. capex is estimate |
| Rule of 40 | growth% + margin% ≥ 40 | SaaS only | margin definition varies |
Sector lens
| Sector | Prefer | Avoid / caution |
|---|---|---|
| Industrials | EV multiples, ROIC, FCF | absolute “cheap P/E” |
| Banks / insurers | P/B, ROE/ROTE, NIM, NPL, capital | EV/EBITDA as primary |
| SaaS | Rule of 40, NRR, FCF margin | industrial templates |
| Cyclicals | mid-cycle / normalized EPS | trough P/E as “expensive”, peak as “cheap” |
| Asset-heavy | EV/EBITDA, FCF, maint. capex | P/E alone |
⌁: Cyclicals look cheapest at the top (peak EPS). Low multiples can be a value trap.
DCF machine
Match cash to rate to value:
| Path | Cash | Rate | Result |
|---|---|---|---|
| Firm | FCFF | WACC | Enterprise value |
| Equity | FCFE | r_e | Equity value |
# FCFF (CFA-style)
FCFF = EBIT(1−t) + D&A − Capex − ΔNWC
= NI + NCC + Int(1−t) − FCInv − WCInv
= CFO + Int(1−t) − Capex # US GAAP: CFO after interest
FCFE = FCFF − Int(1−t) + NetBorrow
= NI + NCC − FCInv − WCInv + NetBorrow
WACC = (E/V)·r_e + (D/V)·r_d·(1−t) # V = market E+D
EV = Σ FCFF_t/(1+WACC)^t + TV_n/(1+WACC)^n
TV = FCFF_{n+1}/(WACC−g) # require WACC>g
or EBITDA_{n+1}×ExitMult # check implied g
Equity = EV − NetDebt − Pref − NCI (+ non-op assets)
Value/sh = Equity / diluted shares
Knobs: forecast margins/growth · WACC · g or exit mult · diluted count · TV share of EV (flag if >80%).
Tiny worked path (illustrative)
fcf = [100, 108, 117, 126, 136] # explicit FCFF
r,g = 0.10, 0.03
pv = Σ f/(1+r)**t
tv = fcf[-1]*(1+g)/(r−g)
EV = pv + tv/(1+r)**n
# then − net debt, ÷ diluted → IV/share; MoS vs bottom of range
!: Single-point DCF is theatre. Bear/base/bull; sensitivity on
g/WACC/margins; sanity-check exit mult vs peers; fade ROIC→WACC in the stable phase.
Margin of safety
MoS = 1 − Price / IntrinsicValue
# Buy when Price ≤ IV_low × (1 − required_MoS)
# e.g. IV range 80–120, require 30% → buy ≤ ~56–70 vs low end — pick a rule and stick to it
MoS is an explicit haircut to estimated IV, sized to how wrong you might be — not “feels cheap.”
Graham shortcuts
Graham number = √(22.5 × EPS × BVPS) # 15×P/E × 1.5×P/B — ceiling heuristic, NOT IV
Owner earnings = NI + D&A + other non-cash − maintenance capex (− ΔWC if needed)
Defensive screen (compressed): adequate size · current ≥2 · low LT debt · 10y positive EPS · long dividend record · EPS growth · P/E≤15 on avg EPS · P/B≤1.5 (or P/E×P/B≤22.5).
!: Owner earnings ≠ reported earnings. Maintenance capex is judgment. Graham number is a ceiling, not a DCF substitute.
Workflow
- EDGAR — 10-K / 10-Q / 20-F: Business, Risks, MD&A, statements + notes, auditor. IR decks = non-GAAP — reconcile.
- Normalize — one-offs, SBC, leases, op vs non-op, diluted shares, NCI.
- Ratios — 5–10y; DuPont on ROE; accruals & CCC.
- Comps — same model; EV multiples for industrials; P/B+ROE for banks.
- DCF — bear/base/bull; TV share; implied
g; ROIC fade. - MoS — vs bottom of credible IV range.
- Checklist — moat · capital allocation · dilution · cyclical peak · BS survival.
Data: SEC EDGAR + IR are primary. Aggregators (Macrotrends etc.): lag, TTM stitch errors, non-GAAP pollution — spot-check the 10-K.
Screening starters (hypothesis generators)
- Quality: ROE/ROIC sustained, FCF≈NI, low accruals
- Solvency: net debt/EBITDA sane, interest cover, BS survival
- Price: peer-relative multiples + MoS to IV range — not absolute bands alone
AI-slop kill-list
| Slop | Reality |
|---|---|
| EBITDA ≈ cash | Ignores capex/WC/tax/SBC/interest → CFO/FCF/FCFF |
| DCF off raw NI | Need FCFF/FCFE + reinvestment + interest shield |
| Ignore dilution/SBC | Per diluted share; SBC = cost or dilution |
| Mix TTM↔fwd, GAAP↔non-GAAP | Label everything; reconcile footnotes |
Gordon g≥WACC | TV blows up; cap g; exit-mult cross-check |
| MoS as “feels cheap” | Explicit % below estimated IV |
| Equity multiple on EV (or reverse) | P/E world ≠ EV/EBITDA world |
| Bank ratios = industrial | Skip EV/EBITDA; P/B, ROE/ROTE, NIM, NPL, capital |
FCF=CFO−capex as FCFF w/o +Int(1−t) | US GAAP CFO is after interest |
| Single-point DCF | Sensitivity; report a range |
| Absolute “cheap P/E” bands | Sector + cycle + ROIC-growth context |
| Graham number as IV | Ceiling heuristic only |
Also: value traps, channel stuffing, survivorship bias, anchoring, confirmation bias, recency bias — build the bear case before you buy.
Refs
- Damodaran
- Damodaran ch.15 (valuation)
- CFA — Free Cash Flow Valuation
- SEC — How to Read a 10-K
- SEC EDGAR
- SEC — Beginners’ Guide to Financial Statements
- Buffett 1986 — owner earnings
- Graham — The Intelligent Investor (Mr. Market, MoS)