Every word behind the number

How much the stock moves relative to the S&P 500. A beta of 1.0 moves with the market, higher is more volatile, and higher volatility raises the discount rate.

The company's sector, which sets the growth ceiling and terminal method, combined with its lifecycle stage read from its own cash-flow history. Re-derived every run, and you can override it.

The implied interest rate the company actually pays. Applied after the tax shield in the WACC, because interest is tax-deductible and equity dividends are not.

The return shareholders demand for owning this particular company. Known as CAPM, the capital asset pricing model.

The figure the projection actually starts from: a trend-fitted or median value across several years, so one distorted year cannot become a runaway base. It can sit far from the latest year.

How much a dollar arriving in year t is worth today, adjusted to mid-year because cash arrives through the year rather than all on 31 December.

Valuing a company by projecting the cash it will generate and discounting each year back to today. Everything Angie does is one long DCF.

Earnings for every dollar invested, which makes a stock directly comparable to a bond yield. A company losing money has none.

The value of the whole operating business, before asking who has a claim on it. Debt holders are paid before shareholders, so this is not yet what the shares are worth.

The extra return investors demand for holding stocks instead of Treasuries. Beta scales it up or down for one particular company.

What is left for shareholders once the lenders have been paid.

The multiple applied to terminal-year free cash flow when the exit-multiple method is used instead of Gordon growth.

How many cents of spendable cash each dollar of sales leaves behind.

The same ratio against next year's expected earnings, so it depends on a forecast rather than a fact.

The cash left for investors after keeping the business running. This is the number a DCF actually values, taken from the most recent fiscal year.

Year-one growth, faded toward the perpetual rate across the horizon. Left blank it comes from the company's own history, capped at the sector ceiling, and a genuine decline is kept rather than floored up.

How many years are projected explicitly before the terminal value takes over. Longer horizons matter only when growth genuinely exceeds the perpetual rate.

Angie's estimate of what one share is worth, built from the company's own cash rather than from what anyone will pay. Compare it to today's price.

The discount below fair value a stock must reach before Angie will call it undervalued. It exists because the estimate itself can be wrong.

What the market currently values the equity at.

The expected long-run return of the broad market. Subtract the risk-free rate from it and what remains is the equity risk premium.

Positive means a net borrower, negative means net cash. Subtracted from enterprise value to reach what shareholders own.

Sets net debt by hand. Mostly used to supply industrial-only debt for a manufacturer whose in-house lending arm inflates the reported figure.

Angie's refusal. A discounted cash flow cannot honestly price a bank, an insurer, a REIT or a company burning cash with no history of earning any, so it declines rather than inventing a number.

A P/E adjusted for how fast the company is growing. Around 1.0 is conventionally treated as fair, though the convention has little theory behind it.

The rate the company is assumed to grow at forever, once the explicit horizon ends. It feeds the Gordon-growth terminal value and cannot sensibly exceed the economy's long-run growth.

What a future year's free cash flow is worth in today's money.

The 10-year Treasury yield, standing in for the return available with essentially no risk. Every other rate in the model is built on top of it.

The share count that converts a whole-company equity value into a per-share figure. Getting it wrong scales every valuation by exactly the same error.

Dampens an unstable post-IPO beta by 15%. A company with only months of trading history measures its own novelty as much as its risk.

Every discounted free cash flow across the explicit projection horizon, added up. Everything after the horizon is handled by the terminal value instead.

For a company not yet generating cash, the steady-state margin it is assumed to reach. The most fragile assumption in the model, because nothing in the history supports it.

How value beyond the horizon is set: Gordon growth, which assumes cash flow grows forever at a fixed rate, or a multiple applied to the final year.

Everything the business is worth beyond the projection horizon, set by Gordon growth or an exit multiple and then discounted back. It is usually the majority of the answer.

What the market pays today for each dollar the company earned last year.

How far intrinsic value sits above or below today's price. Positive means the model thinks the stock is cheap, negative means expensive.

How far the model will stand behind its own answer. Low means free cash flow has been volatile, the price sits far above what the business generates, or a margin had to be guessed. It is not the strategy signal score.

Undervalued, fair or overvalued, decided by where the price sits against intrinsic value once the margin of safety is applied.

Cost of equity and cost of debt blended by how much of each the company uses. This is the rate every future dollar is discounted at, and it is the single most leveraged input in the model.

Spending on property and equipment. Reported as a negative number and shown here as the amount subtracted from operating cash flow.

Cash, equivalents and short-term investments. Netted against debt to bridge the value of the business to the value of the shares.

Profit for the last twelve months divided by the share count. Accounting profit, not cash, which is why Angie values the cash instead.

Cash the core business actually generated, before anything is spent on maintaining or expanding it. Harder to massage than reported earnings.

Total sales over the trailing twelve months, before any cost is subtracted.

Short-term plus long-term borrowings, before any cash is netted off.

Measures how strong a trend is without saying which way it points. Low readings mean a market going sideways, where trend-following strategies bleed.

The typical distance a price travels in a day. Used to size stops in the stock's own terms, so a volatile name gets more room than a quiet one.

Running a set of rules over history to see what they would have done. It shows what did happen, never what will, and it flatters any rule tuned on the same data.

A moving average with bands drawn two standard deviations above and below it. Price at a band is unusual relative to its own recent range, not relative to any other stock.

Buying once and doing nothing. The benchmark every strategy must beat, measured over the identical window, because beating nothing is not a result.

Total return restated as a steady yearly rate, which makes runs of different lengths comparable.

Return measured against the worst loss along the way, rather than against general choppiness. It punishes the single deepest hole.

Tuning a strategy until it memorises one stretch of history. It produces beautiful backtests and no future returns, which is why results are held out and re-tested.

An average of recent prices that weights the latest days most heavily, so it turns faster than a simple average. A fast one crossing above a slow one is the classic trend signal.

Seeding a population of settings, keeping the strongest, then mutating and recombining them over generations to home in on what works.

Testing every combination of the settings you chose. Exhaustive and certain, and the count explodes as soon as you add a third or fourth knob.

The gap between a fast and a slow moving average, plotted against its own average. Widening upward means momentum is building, narrowing means it is fading.

The deepest fall from a peak before a new peak was reached. The number that decides whether a strategy is one you could actually hold through.

The tendency of a stretched price to drift back toward its own average. Angie measures how often that actually happened for a name rather than assuming it will.

Testing on data the search never saw. Angie hides the last quarter of history, then re-runs the winners on it, and flags any whose result collapses.

How many dollars the winners made for every dollar the losers cost. Below 1.0 the strategy loses money regardless of how often it is right.

Sampling combinations at random instead of testing all of them. Cheap coverage of a space too large to walk, at the cost of possibly missing the best corner.

A 0 to 100 gauge of how one-sided recent moves have been. Conventionally below 30 is oversold and above 70 overbought, though a strong trend can hold an extreme for weeks.

Return per unit of volatility. It asks how much turbulence you endured for the gain, so a smoother path scores better than a wild one that ends up level with it.

The weighted entry score a bar must clear before the strategy opens a trade, built from trend and momentum agreement. It says nothing about whether a stock is cheap.

What the account gained or lost across the whole test, start to finish.

The share of trades that made money. High on its own means little, since many small wins and one catastrophic loss still lose.

How many standard deviations today's price sits from its recent average. Two sigma below means a move this far down happened on roughly one day in forty.