The words, in plain English.

What one share of the company costs on the stock market right now.

How it's worked out: Quoted by the exchange.

How to read it: On its own it says nothing about cheap or expensive.

The total price of the whole company on the stock market.

How it's worked out: Share price x number of shares.

How to read it: Apple ~$5,000B is 'mega cap'; under $2B is 'small cap'.

What it would cost to buy the whole business, including paying off its debt.

How it's worked out: Market cap + debt - cash.

How to read it: Used for comparing companies with different amounts of debt.

All the money customers paid the company in a year (also 'sales', 'top line').

How it's worked out: From the income statement.

How to read it: Growing revenue = more customers or higher prices.

Profit from running the business, before interest and taxes.

How it's worked out: Revenue - cost of goods - operating expenses.

How to read it: Shows how profitable the core business is.

Operating profit before depreciation and amortisation, a rough cash-profit measure.

How it's worked out: EBIT + D&A.

How to read it: Popular with bankers; ignores the cost of replacing equipment.

The final profit for shareholders after interest and taxes ('bottom line').

How it's worked out: From the income statement.

How to read it: Basis of the P/E ratio.

Depreciation & amortisation: the yearly accounting cost of wearing out buildings, machines, software.

How it's worked out: From the cash flow statement.

How to read it: Not cash going out this year, so it is added back for cash flow.

Capital expenditures: cash spent on long-lived things (stores, factories, equipment).

How it's worked out: Cash flow statement, investing section.

How to read it: Capex above D&A usually means the company is growing its asset base.

Money tied up in running the business day to day: inventory and money owed by customers, minus bills not yet paid.

How it's worked out: (Current assets - cash) - (current liabilities - short-term debt).

How to read it: Growing companies usually need more of it, which uses cash.

Cash the business actually brought in from operations in a year.

How it's worked out: Cash flow statement.

How to read it: Should be close to or above net income for a healthy business.

Cash left after running AND maintaining/growing the business: what could go to owners.

How it's worked out: Operating cash flow - capex.

How to read it: The single most important number for valuation.

Free cash flow to the firm: cash available to ALL investors (lenders and shareholders).

How it's worked out: EBIT x (1 - tax) + D&A - capex - increase in working capital.

How to read it: What the main DCF discounts.

Free cash flow to equity: cash available to shareholders after paying lenders.

How it's worked out: FCFF - interest x (1 - tax) + net new borrowing.

How to read it: Used by the FCFE DCF.

Cash the company pays its shareholders, usually quarterly.

How it's worked out: Cash flow statement, financing section.

How to read it: Stable, growing dividends signal confidence.

The company buying its own shares, which shrinks the share count.

How it's worked out: Cash flow statement.

How to read it: Like a dividend, it returns cash to shareholders.

Stock-based compensation: paying employees in shares instead of cash.

How it's worked out: Cash flow statement.

How to read it: A real cost: it dilutes existing shareholders.

Shares outstanding including those that options and stock awards would create.

How it's worked out: From the 10-K / cover page.

How to read it: Per-share values divide by this.

Official filings to the SEC: annual report, quarterly report, and 'something important happened'.

How it's worked out: SEC EDGAR.

How to read it: Where every number in this app comes from.

The machine-readable tags inside SEC filings (e.g. 'Revenues').

How it's worked out: SEC API.

How to read it: Lets the app import a company automatically.

A basket that tracks a whole market: the S&P 500 (500 large US companies), the Nasdaq (tech-heavy) and the Russell 2000 (small companies).

How it's worked out: Weighted average of the members' prices.

How to read it: The yardstick: a stock pick only helps if it beats the index.

What the business is worth based on the cash it will produce, regardless of today's price.

How it's worked out: Weighted blend of the valuation methods.

How to read it: Price below fair value = possibly undervalued (but see Track record).

Discounted cash flow: add up all future cash flows, each shrunk ('discounted') because money later is worth less than money now.

How it's worked out: Sum of FCF_t / (1 + WACC)^t + terminal value.

How to read it: The core valuation method; very sensitive to growth and discount rate.

The value of all cash flows after the forecast years, 'forever'.

How it's worked out: Gordon: FCF x (1+g) / (WACC - g); or exit multiple x EBITDA.

How to read it: Often over half of a DCF's value.

How fast cash flows grow forever after the forecast.

How it's worked out: An assumption, usually 2-3%.

How to read it: Should sit between inflation and the 10-year yield.

The yearly return investors need before this investment is worth making. Future cash is 'discounted' at this rate: the higher it is, the less future cash is worth today. Pros call it WACC, the weighted average cost of capital.

How it's worked out: Equity weight x cost of equity + debt weight x cost of debt x (1 - tax).

How to read it: Higher = lower value. 1 point can move value 15-20%. Typically 7-10% for large US companies.

The return shareholders require for owning this stock.

How it's worked out: CAPM: risk-free rate + beta x equity risk premium.

How to read it: Typically 8-11% for large companies today.

The interest rate the company would pay to borrow today.

How it's worked out: Risk-free rate + credit spread (from its rating).

How to read it: Interest is tax-deductible, so after-tax cost is lower.

What a 'riskless' investment pays: the 10-year US Treasury yield.

How it's worked out: FRED DGS10.

How to read it: Every other return is compared to this.

The extra return investors demand for owning stocks instead of Treasuries.

How it's worked out: Damodaran's implied ERP (monthly).

How to read it: Usually 4-6%.

How much a stock moves with the overall market.

How it's worked out: Regression of the stock's returns on the S&P 500; blended with its industry's beta.

How to read it: 1.0 = moves with the market; 1.5 = 50% more; 0.5 = half.

A discount to fair value you insist on before buying, in case you're wrong.

How it's worked out: Buy-below price = fair value x (1 - margin).

How to read it: Buffett's core idea; 25-30% is common.

Running the DCF backwards: what growth does today's price assume?

How it's worked out: Solve for growth that makes DCF value = price.

How to read it: If it assumes far more growth than history, the market is optimistic.

Assumes the business is 'sold' at the end of the forecast at a multiple of EBITDA.

How it's worked out: TV = EBITDA_n x multiple.

How to read it: A cross-check on the Gordon terminal value.

Values a stock as all its future dividends, discounted.

How it's worked out: Sum of dividends / (1 + Ke)^t.

How to read it: Only meaningful for dividend payers.

Book value plus the profits earned above what shareholders require.

How it's worked out: Book value + sum of (NI - Ke x book) discounted.

How to read it: Good for companies with stable book values.

What the company is worth if it never grows: a floor.

How it's worked out: Normalised after-tax operating profit / WACC.

How to read it: Price far above EPV = paying a lot for growth.

Runs the valuation 10,000 times with random but realistic inputs to get a range, not one number.

How it's worked out: Random draws for growth, margins, rates, etc.

How to read it: Gives probabilities like 'P(undervalued)'.

Share of the 10,000 simulated outcomes where value > today's price.

How it's worked out: Monte Carlo.

How to read it: A MODEL probability; the lab found it not well calibrated.

Optimistic, expected and pessimistic versions of the future, each with a probability.

How it's worked out: Your inputs or history-based defaults.

How to read it: The probability-weighted value blends them.

How many dollars of revenue each dollar invested produces.

How it's worked out: Revenue / invested capital.

How to read it: Used to make sure forecast growth is paid for.

Chart of every method's value range side by side with the price.

How it's worked out: One bar per method.

How to read it: Shows where the methods agree or disagree.

Table of values for different discount rates and growth rates.

How it's worked out: DCF recalculated per cell.

How to read it: Shows how fragile the valuation is.

The app's one-line read of price against value: WELL BELOW VALUE, BELOW VALUE, NEAR FAIR VALUE, ABOVE VALUE or EXPENSIVE (other investor profiles use their own words). Banks, insurers and REITs get ROUGH GUIDE ONLY, because the methods don't fit them.

How it's worked out: Price vs the buy-below price, the fair value +/- the verdict threshold, and the top of the value range.

How to read it: A starting point for your own judgement, not an instruction. See Track record for how well it worked.

The low-to-high band of what the shares could be worth, rather than one number.

How it's worked out: Average of the middle half of the valuation methods and the middle half of the simulation.

How to read it: Price inside the band = fairly valued; below it = possibly cheap; above it = possibly expensive.

The price at which the shares are cheap enough to leave room for error.

How it's worked out: Fair value x (1 - margin of safety); expensive-above = the top of the value range or fair value + threshold.

How to read it: Below it = WELL BELOW VALUE: room for the estimate to be wrong. Not a buy instruction.

Yearly return if the price drifts to fair value over 5 years, on top of the return investors require.

How it's worked out: (1 + required return) x (value / price)^(1/5) - 1, across the simulated range.

How to read it: An estimate of what the cash flows justify, not a price forecast.

How much each valuation method counts towards the fair value.

How it's worked out: Defaults: cash-flow methods 1.5, peers 2-3, simulation 1, re-cuts of the same forecast 0.5, floor 0.

How to read it: Set your own under weights.<method>; 0 leaves a method out.

Net operating profit after tax: operating profit as if the company had no debt.

How it's worked out: EBIT x (1 - tax rate).

How to read it: The profit a business earns on all the money invested in it.

Cash put back into the business to grow it: new equipment beyond wear and tear, plus extra working capital.

How it's worked out: Capex - D&A + increase in working capital.

How to read it: Growth has to be paid for; a forecast that grows without reinvesting is too good to be true.

What a future amount of money is worth today.

How it's worked out: Future cash / (1 + required return)^years.

How to read it: $100 in 10 years at 8% is worth about $46 today.

Assumes a year's cash arrives through the year, not all on December 31.

How it's worked out: Discount at t - 0.5 instead of t.

How to read it: Bankers' standard; adds a few percent to value.

The return earned on each new dollar invested to grow.

How it's worked out: Extra after-tax profit / extra capital invested.

How to read it: Above the required return, growth adds value; below it, growth destroys value.

The first years of the forecast, before growth fades to the long-run rate.

How it's worked out: Default 5 years at the starting growth rate, then a straight-line fade.

How to read it: Longer high-growth periods raise value a lot: be sceptical.

The share of operating profit paid as tax in the forecast.

How it's worked out: Average past effective rate, 0-35%.

How to read it: US federal rate is 21%; state and foreign taxes add to it.

The steady growth rate, forever, that would justify today's price with no high-growth phase.

How it's worked out: g = (EV x WACC - FCF) / (EV + FCF).

How to read it: Above about 4% the price needs a long run of strong growth.

Price-to-earnings: how many years of current profit the price represents.

How it's worked out: Market cap / net income.

How to read it: S&P 500 ~ 20-25 today; higher = expensive or fast-growing.

Enterprise value over cash-like operating profit.

How it's worked out: EV / EBITDA.

How to read it: Debt-neutral; 8-15x typical for mature companies.

Enterprise value over revenue.

How it's worked out: EV / revenue.

How to read it: Used when profits are small or negative.

Price over free cash flow.

How it's worked out: Market cap / FCF.

How to read it: Inverse of FCF yield.

Free cash flow as a % of market cap: the cash 'interest' the stock earns.

How it's worked out: FCF / market cap.

How to read it: Compare with the 10-year Treasury yield.

Net income as a % of market cap (inverse of P/E).

How it's worked out: 1 / P/E.

How to read it: Above the Treasury yield = stocks pay more than bonds.

Yearly dividends as a % of the price.

How it's worked out: Dividends / market cap.

How to read it: S&P 500 ~1.3%.

Valuing a company by the multiples similar companies trade at.

How it's worked out: Peer median multiple x this company's metric.

How to read it: Banks' second main method after DCF.

The middle value among the comparable companies (half are above, half below).

How it's worked out: Median of the peers' multiples; negative or absurd (>100x) values are left out.

How to read it: Less distorted by one odd peer than an average.

Return on invested capital: profit per dollar put into the business.

How it's worked out: After-tax operating profit / (equity + debt - cash).

How to read it: Above the cost of capital (~9%) creates value; in the lab, high (book) ROIC leaned toward beating the market, but not reliably (t under 2 once overlapping return windows are counted).

Return on equity: profit per dollar of shareholders' book money.

How it's worked out: Net income / book equity.

How to read it: Can be inflated by debt or buybacks.

Cents of operating profit per dollar of sales.

How it's worked out: EBIT / revenue.

How to read it: Stable or rising margins = pricing power.

How much of reported profit turns into real cash.

How it's worked out: FCF / net income.

How to read it: Near or above 100% is healthy; far below can mean aggressive accounting.

Profit that hasn't turned into cash yet.

How it's worked out: (Net income - OCF) / assets.

How to read it: High accruals often precede weaker earnings.

9 yes/no tests of financial health (profit, cash, debt, efficiency).

How it's worked out: Count of passed tests.

How to read it: 7-9 strong, 0-3 weak.

Bankruptcy-risk score from 5 ratios.

How it's worked out: 1.2 WC/TA + 1.4 RE/TA + 3.3 EBIT/TA + 0.6 MV/TL + 1.0 Sales/TA.

How to read it: > 3 safe, 1.8-3 grey, < 1.8 distress.

Screen for earnings manipulation from 8 accounting ratios.

How it's worked out: Weighted sum of 8 indices.

How to read it: > -1.78 = possible manipulation.

How fast the share count grows (owners' slice shrinks).

How it's worked out: Yearly change in shares.

How to read it: Negative (buybacks) is shareholder-friendly.

Share of profit or free cash flow paid as dividends.

How it's worked out: Dividends / FCF (or / net income).

How to read it: Under ~60% of FCF is usually safe.

Profit with one-off items (impairments, restructuring, asset-sale gains) removed.

How it's worked out: EBIT + one-off charges - one-off gains.

How to read it: Better guide to ongoing profitability.

A/B/C grade for how well the business turns invested money into profit.

How it's worked out: 5-year average return on capital vs the required return: A = 10+ points above, B = above, C = below.

How to read it: The one quality signal that predicted returns in the app's own backtest.

A/B/C grade for how safe the balance sheet is.

How it's worked out: Points for low debt vs cash profit, high interest cover, little debt due soon and moderate price swings.

How to read it: C = check the debt before anything else.

A/B/C grade for how likely the dividend is to keep being paid.

How it's worked out: Points for paying out under 60% of free cash flow, modest debt, no past cuts, and dividend growth.

How to read it: C = the dividend may be at risk.

How much the price swings, per year.

How it's worked out: Standard deviation of returns x sqrt(12 or 252).

How to read it: S&P 500 ~15-20%; single stocks 20-40%.

Fall from the previous peak.

How it's worked out: Price / running maximum - 1.

How to read it: Max drawdown is the worst such fall.

Years of cash-like profit needed to repay net debt.

How it's worked out: (Debt - cash) / EBITDA.

How to read it: < 1.5x comfortable; > 3x stretched.

How many times operating profit covers interest payments.

How it's worked out: EBIT / interest expense.

How to read it: > 8x very safe; < 2x fragile.

How much the DCF value changes if the 10-year yield rises 1 point.

How it's worked out: DCF at Rf + 1pt vs base.

How to read it: Long-duration growth stocks are most sensitive.

How much two things move together, from -1 to +1.

How it's worked out: Pearson correlation of returns.

How to read it: Low correlation between holdings = real diversification.

How much holdings offset each other.

How it's worked out: Weighted avg stock vol / portfolio vol.

How to read it: 1.0 = no benefit; higher = more diversified.

The lowest and highest price of the past year.

How it's worked out: From daily prices.

How to read it: Near the low can mean bad news or a bargain; near the high, momentum or optimism.

The average closing price of the last 200 trading days (about 10 months).

How it's worked out: Mean of the last 200 closes.

How to read it: Price above it = uptrend; below it = downtrend.

Company executives and directors buying or selling their own company's shares (SEC Form 4).

How it's worked out: Open-market buys and sells from the last 12 months of Form 4 filings.

How to read it: Insiders sell for many reasons; several of them buying is the stronger signal.

Anyone who owns 5% or more of a company's shares must tell the SEC: Schedule 13G for a passive holder such as an index fund, 13D for one that may push for changes.

How it's worked out: Each holder's latest 13D or 13G filing about the company.

How to read it: Index funds owning 5-8% is normal. A founder or family with a big stake controls a lot; a new 13D often means an activist wants changes.

When the company's debt has to be repaid, year by year.

How it's worked out: From the 10-K debt footnote (XBRL).

How to read it: A lot due soon at higher interest rates squeezes profit.

How much of the portfolio to put in one stock.

How it's worked out: Volatility target (2% of portfolio risk per position, cap 10%) vs quarter-Kelly; the smaller wins.

How to read it: Size by how wrong you could be, not by how excited you are.

A formula for bet size that maximises long-run growth if your edge is known exactly.

How it's worked out: Weight = (expected return - risk-free) / variance; the app uses a quarter of it.

How to read it: Real edges are uncertain, so full Kelly is far too aggressive.

How many equally sized holdings your portfolio behaves like.

How it's worked out: 1 / sum of squared weights.

How to read it: 10 stocks with one at 60% act like about 2.

Return per unit of risk, above cash.

How it's worked out: (Return - T-bill) / volatility, per year.

How to read it: S&P 500 long-run ~0.4-0.7; above 1 is excellent.

Like Sharpe but only counts downside swings as risk.

How it's worked out: Excess return / downside volatility.

Return beyond what the market exposure (beta) explains: skill.

How it's worked out: Return - beta x market return.

How to read it: Positive and statistically significant alpha is rare.

How many 'standard errors' a result is from zero, i.e. how unlikely it is to be luck.

How it's worked out: Average / (standard deviation / sqrt(n)).

How to read it: |t| > 2 is the usual bar for 'probably real'.

Correlation between a signal's ranking of stocks and their later returns.

How it's worked out: Spearman rank correlation, per date, averaged.

How to read it: 0.05 is useful in practice; negative = signal points the wrong way.

Probability a strategy's edge is real after accounting for how many variants were tried and luck.

How it's worked out: Bailey & Lopez de Prado (2014).

How to read it: > 0.95 convincing; trying many strategies makes it harder.

Testing by choosing settings on past data only, then trading the next period, rolling forward.

How it's worked out: Fit on trailing 60 months, test next 12.

How to read it: Prevents fitting the test to the answer.

Results on data the model never saw while being built.

How to read it: The only results worth trusting.

Accidentally using information that wasn't available yet at the time.

How to read it: The app prevents it by using only filings/prices dated before each decision.

Testing only on companies that survived to today, which flatters results.

How to read it: Fixed by using the index as it was on each date, plus delisting returns.

Whether a stated probability matches reality (do '70%' calls come true 70% of the time?).

How it's worked out: Predicted vs realised by bucket.

How to read it: The journal tracks your own calibration.

How well a score separates two outcomes (e.g. before-recession vs not).

How to read it: 0.5 coin flip, 1.0 perfect.

Simulating a strategy on history to see how it would have done.

How to read it: Easy to fool yourself; see walk-forward.

Compound annual growth rate: the steady yearly return that gets the same end result.

How it's worked out: (End / start)^(1/years) - 1.

How much of the portfolio is traded per year.

How it's worked out: Sum of weight changes / 2, per year.

How to read it: High turnover = high trading costs.

One fifth of a ranked list: Q1 is the bottom 20%, Q5 the top 20%.

How it's worked out: Rank, then split into 5 groups.

How to read it: If a signal works, Q5 should beat Q1 steadily.

How well this app's own verdicts would have worked: it re-valued 59 large US companies every April since 2012 using only data known at the time, then checked the next 1, 3 and 5 years.

How it's worked out: Rank IC, quintile returns and calibration vs the S&P 500 (Track record tab).

How to read it: Finding so far: 'cheap vs fair value' did not predict returns; high return on capital did.

The overnight interest rate the Federal Reserve targets.

How it's worked out: FRED DFF.

How to read it: Raising it slows the economy and inflation.

Interest rates for different loan lengths (3 months ... 30 years).

How to read it: Normally longer = higher.

Short-term rates above long-term rates.

How it's worked out: 10y - 3m < 0.

How to read it: Has preceded every US recession since the 1960s (with a lag).

Extra interest risky companies pay over Treasuries.

How it's worked out: Corporate yield - Treasury yield.

How to read it: Widening fast = stress; very tight = complacency.

How fast prices rise. CPI and core PCE are the main measures (core excludes food and energy).

How it's worked out: Year-over-year % change.

How to read it: The Fed targets 2% core PCE.

Inflation the bond market expects (from inflation-protected bonds).

How it's worked out: FRED T10YIE.

Interest rate after inflation.

How it's worked out: Nominal rate - inflation.

How to read it: Positive and high = tight money.

Recession signal from rising unemployment.

How it's worked out: 3-month avg unemployment - its 12-month low.

How to read it: >= 0.5 points has signalled recessions.

Fed survey: share of banks tightening lending standards.

How it's worked out: Net %.

How to read it: High = credit drying up.

The economy's phases: early expansion, mid-cycle, late cycle, contraction.

How it's worked out: Scorecard of spreads, curve, lending, labour, policy.

How to read it: Shapes which investments tend to do well.

NY Fed model's chance of a recession in the next 12 months.

How it's worked out: Based on the 10y-3m curve.

How to read it: Above ~30% has preceded recent recessions.

Chicago Fed index of how easy it is to borrow and raise money.

How it's worked out: Weighted mix of 105 measures of risk, credit and leverage.

How to read it: Above 0 = tighter than average; below 0 = looser.

How fast the economy is growing after inflation.

How it's worked out: Quarter-on-quarter change, annualised.

How to read it: Below 0 for a while = recession; 2% is about the long-run US pace.

Share of people looking for work who can't find it.

How it's worked out: FRED UNRATE (monthly).

How to read it: A quick rise from a low is one of the most reliable recession signs.

The stock market's expected swings over the next month, from option prices.

How it's worked out: CBOE VIX index.

How to read it: Under 15 calm, 20-30 nervous, above 30 fearful.

A list of what an institution owns (assets) against what it owes (liabilities) plus its owners' stake (equity).

How it's worked out: Assets = liabilities + equity.

How to read it: Every transaction changes at least two lines, so the two sides always add up (double entry).

Money a bank keeps in its own account at the central bank. Only banks can hold it; they use it to pay each other.

How it's worked out: Created or removed by the central bank (bond buying, lending) and by demand for banknotes.

How to read it: More reserves do not mean more loans: US reserves rose about 62 times from 2008 to 2014 while bank loans rose 12%.

When a bank makes a loan it credits the borrower with a new deposit: new money. Repaying a loan deletes it.

How it's worked out: Bank: loan +X (owns), deposit +X (owes).

How to read it: Limited by borrowers, the bank's own risk and capital, and the central bank's interest rate, not by reserves (Bank of England, 2014).

How banks pay each other when a customer of one pays a customer of another: reserves move between their accounts at the central bank.

How it's worked out: Payer's bank: reserves -X, deposits -X. Payee's bank: reserves +X, deposits +X.

How to read it: Payments inside one bank need no reserves; payments both ways mostly cancel out.

The central bank buys bonds and pays with newly created reserves.

How it's worked out: Central bank: bonds +X / reserves +X. The seller's bank: reserves +X / deposit +X.

How to read it: Creates reserves and deposits, not loans; it works mainly through lower long-term interest rates.

How many dollars of assets sit on each dollar of the owners' money.

How it's worked out: Assets / equity.

How to read it: At 10x, losses equal to 10% of assets wipe the owners out; US banks' equity was about 10% of assets in 2026.

Many depositors asking for their money at once. A bank holds loans that cannot be sold quickly at full value, so even a sound bank can fail in a run.

How to read it: Selling good loans in a hurry at a discount turns a cash problem into a solvency problem.

The central bank lending to a sound bank that is short of cash, against good collateral, so it does not have to sell assets at fire-sale prices.

How it's worked out: Bagehot (1873): lend freely, at a penalty rate, against good collateral.

A guarantee that small deposits are repaid if a bank fails, so their owners have no reason to run.

How it's worked out: US: FDIC, up to $250,000 per depositor, per bank, per ownership category.

How to read it: Uninsured deposits run first: about 94% of Silicon Valley Bank's were uninsured in 2023.

How much less than an asset's market value a lender will lend against it.

How it's worked out: Loan = value x (1 - haircut).

How to read it: Rising haircuts force borrowers to sell, which pushes prices down and haircuts up again: a 2008 mechanism.

The bid is the best price a buyer will pay right now; the ask (or offer) is the best price a seller will take.

How it's worked out: Spread = ask - bid.

How to read it: A wide spread means trading is expensive or the market is nervous.

Every waiting order to buy or sell, price by price.

How it's worked out: Depth = how much waits at each price.

How to read it: Thin books move a lot when someone must trade.

Buy or sell now at the best prices waiting. Big orders eat through several prices.

How to read it: Fast, but you pay the spread and the slippage.

Buy or sell only at a price you choose or better; it waits in line until someone trades with it.

How to read it: Cheaper, but it may never fill.

The price you actually get is a bit worse than the quote.

How it's worked out: 5 basis points here.

Borrowing shares, selling them, and hoping to buy them back cheaper.

How it's worked out: Loss = (buy-back price - sale price) x shares, plus the borrow fee.

How to read it: A short's loss has no ceiling.

A lender's demand for more cash or collateral when a borrowed position loses value.

How it's worked out: Called when equity < maintenance margin x position value.

How to read it: Ignore it and the lender sells for you.

Someone who must sell whatever the price: a margin call, a rule, a redemption.

How to read it: The best prices often come from people who have no choice.

A loan secured by bonds: sell bonds today and agree to buy them back tomorrow.

How it's worked out: Cash lent = bond value x (1 - haircut).

How to read it: The plumbing behind most bond leverage.

The return a bond gives at today's price if held to the end.

How it's worked out: The rate that makes its coupons and repayment worth today's price.

How to read it: When yields rise, prices fall.

How sensitive a bond's price is to interest rates.

How it's worked out: About the % price change for a 1-point change in yield.

How to read it: A 30-year bond can lose about 14% when its yield rises one point from 5%.

The right to sell at a set price (the strike): insurance against a fall.

How it's worked out: Premium paid up front; pays max(strike - price, 0).

How to read it: Cheap in calm markets, dear in a panic.

The right to buy at a set price: a bet on a rise with a known cost.

How it's worked out: Pays max(price - strike, 0).

How to read it: Dealers who sell calls often buy shares as the price rises to stay hedged.

A 1980s rule: sell futures as the market falls to cap losses.

How to read it: A forced seller in disguise; it accelerated the 1987 crash (Federal Reserve History).

Buying what has risen most over the past year (skipping the last month).

How it's worked out: 12-1 month return.

How to read it: One of the most documented market patterns; prone to sharp crashes.

Hold an asset when its price is above its average, step aside when below.

How it's worked out: Price vs N-month moving average.

How to read it: Cuts big losses; lags in choppy markets.

Spreading risk (not dollars) equally across assets like stocks, bonds, gold.

How it's worked out: Weight ~ 1 / volatility.

How to read it: Dalio's All Weather idea.

Soros: prices affect fundamentals and vice versa, creating self-reinforcing booms and busts.

Buffett/Munger: excellent businesses at sensible prices, held long.

Many small, statistically-tested edges combined (Renaissance's approach).

All four lens strategies held together, sized by their recent risk.

How to read it: Diversification across philosophies.

Trading with simulated money and live prices.

How to read it: Test before risking real money.

One hundredth of a percent.

How it's worked out: 0.01%.

How to read it: 50 bp = 0.5%.

Data exactly as it was known on a past date, with no later revisions.

How to read it: Needed for honest backtests.

App's confidence in each imported number.

How it's worked out: Accounting checks and cross-checks.

How to read it: A clean, B fixed or derived, C needs your check.

Which kind of investor the verdict is written for: long-term, dividend / income, trader, or index + a few picks.

How it's worked out: Same valuation; the profile changes the verdict's wording and what leads.

How to read it: Space v cycles through them.

The companies you follow. Home shows each one's price, value range and verdict, plus alerts.

How it's worked out: Saved in ~/.config/prizume/prefs.json.

How to read it: Space a adds the open company; Ctrl-e reorders.

Your own record of each decision and why, reviewed automatically after 6, 12 and 24 months.

How it's worked out: Markdown files in data/journal/.

How to read it: The only honest way to learn whether your judgement adds value.