The market, in plain words.

Return above what the risk taken would predict — the part the manager can claim credit for.

Almost all apparent alpha turns out, on inspection, to be an unrecognized risk factor: leverage, illiquidity, small size, or a value tilt. Real alpha is rare, expensive to find, and tends to erode once it is published.

The average distance a price travels in a bar — how much this instrument normally moves.

The correct unit for stops and position size. A 2% stop is tight on one stock and loose on another; two ATRs is the same amount of room on both. Sizing by ATR is why two positions can carry equal risk with very different dollar amounts.

The total value a manager or fund is responsible for.

Also the manager's revenue base, which creates the industry's central conflict: fees are charged on assets, so the incentive is to gather more, and past a point more assets make good returns harder. The rare managers who return capital are telling you they know that.

One hundredth of a percentage point. 25 bps is 0.25%.

Used because 'the rate rose 0.25 percent' is ambiguous — from 4% to 4.25%, or from 4% to 4.01%? Basis points remove the ambiguity, which matters when trillions are indexed to the answer.

How much an instrument moves relative to the market. Beta 1.5 means it moves half again as much.

A measure of shared movement, not of danger. A stock with beta 0.4 that is one bad quarter from insolvency is riskier than one with beta 1.8 and no debt. Beta also drifts: it is calculated on the past and the past regime may be over.

The highest price a buyer will pay and the lowest a seller will take. The gap is the spread.

The spread is the cost of impatience, paid every time you cross it. On a liquid mega cap it is a cent; on a thin small cap it can be a percent, which quietly destroys a strategy that trades often.

The constant annual rate that would take you from the start value to the end value.

The only honest way to compare returns over different periods. It hides the path completely — the same CAGR can come from a smooth climb or from a 70% collapse followed by a recovery, and those are not the same experience.

The right to buy at the strike price. Gains value as the underlying rises.

Buying a call is a bullish bet with a known maximum loss — the premium. Selling one you do not own is a bullish bet against you with no cap on the loss, which is why brokers require collateral for it.

The strike with the largest concentration of call (or put) gamma.

Where dealer hedging is heaviest, so price often stalls approaching it and accelerates once through. It moves as positioning changes, which is why yesterday's wall can be irrelevant today.

Earning returns on the returns you already earned.

The entire mechanism behind every long-term fortune in markets, and it is unspectacular until very late: most of Buffett's wealth arrived after he turned sixty. The corollary is that anything interrupting it — a large loss, a tax event, a fee — costs far more than it appears to.

How closely two instruments move together, from −1 to +1.

The foundation of diversification and the thing that betrays it: correlations rise toward 1 in a crash, exactly when the diversification was supposed to help. A portfolio that looks diversified in calm markets can be one position in a panic.

Selling a call against shares you already own.

You collect the premium and cap your upside at the strike. It converts an uncertain large gain into a certain small one, which is a good trade in a flat market and an expensive one in a rising market — where the shares get called away exactly when they were working.

Consumer Price Index — the headline measure of inflation.

Published monthly and moves markets more than almost any other release, because it decides what the central bank does next. Core CPI strips food and energy — not because they do not matter but because they are noisy — and it is core that policy actually reacts to.

Current assets divided by current liabilities — whether the company can pay what it owes this year.

Below 1 means short-term obligations exceed short-term resources, which is normal for a supermarket that turns inventory in days and alarming for a manufacturer that turns it in months. Context decides.

The nine-character code identifying a US security.

What regulatory filings speak in, and nothing else does. Turning a CUSIP into a ticker requires a mapping that no free official source publishes — which is why so much filing data is presented as nine-character codes nobody can read.

Shares short divided by average daily volume — how long the whole short side would take to buy back.

The number that turns short interest into a warning. Twenty million shares short says little on its own; twenty million that would take nine days to cover is a crowded exit. The regulator floors the published value at 1.

Total debt divided by shareholders' equity — how much of the business is financed with borrowed money.

Leverage magnifies both directions. The level that is safe depends entirely on how predictable the cash flow is: a utility can carry debt that would destroy a semiconductor company. What matters more than the ratio is when the debt comes due and at what rate it refinances.

How much the option price moves for a one-dollar move in the underlying.

Runs 0 to 1 for calls and 0 to −1 for puts. Also read as a rough probability of finishing in the money: a 0.30 delta call has around a 30% chance. A dealer short calls is short delta and must buy the underlying to stay neutral — which is where option positioning starts moving the stock.

Price making a new extreme that the indicator does not confirm.

A higher high in price with a lower high in RSI says the move is running on less force than the last one. It is a warning about momentum, not a signal to reverse — divergences can persist for a long time while price keeps going.

Annual dividend divided by share price — the cash return at today's price.

A yield that suddenly looks generous usually means the price fell, not that the dividend rose — and the market is often right that the dividend is about to be cut. Read it with the payout ratio and free cash flow before treating it as income.

The gap between reported earnings and what analysts expected.

The market trades the surprise, not the result — which is why a company can report record profits and fall 8%. Companies know this and guide expectations down beforehand, so a small beat against a lowered bar is a weaker signal than it looks.

Earnings before interest, taxes, depreciation and amortization — a rough proxy for operating cash generation.

Useful for comparing businesses with different debt loads and tax situations. Charlie Munger famously dismissed it, because depreciation is a real cost — the machines do wear out and have to be replaced. Trust it least where capital spending is highest.

A moving average that weights recent bars more heavily, so it turns sooner than the simple one.

The 9 and 21 pair is the short-term standard: price holding above both is the definition of an intact short trend, and the cross between them is one of the oldest signals there is. Faster response is bought with more false turns — that is the whole trade-off.

Market cap plus net debt — what it would actually cost to buy the whole company.

Buy every share and you also inherit the debt and pocket the cash. EV is that arithmetic. It is why a company with a small market cap and enormous debt can be far more expensive than it looks.

Net profit divided by shares outstanding — how much the company earned per share.

The denominator matters as much as the numerator: buying back shares raises EPS without the business earning a cent more, and issuing shares lowers it without the business earning less. Read EPS growth alongside share count, always.

A fund that trades on an exchange like a share, usually tracking an index.

Its price stays close to the value of what it holds because authorized participants can create and redeem shares in kind, arbitraging any gap. That mechanism is invisible until it strains, and when it strains — thin bond ETFs in a panic — the discount can get wide.

Enterprise value divided by EBITDA — the price of the whole business against its cash earnings before financing choices.

The multiple acquirers actually use, because it includes debt and ignores how the company chose to finance itself or depreciate its assets. That neutrality is also its flaw: EBITDA pretends capital expenditure is free, which for anything capital-intensive is a lie.

The annual percentage a fund charges, deducted from returns automatically.

Small numbers that compound into large ones: 1% a year on a 30-year horizon takes roughly a quarter of the final amount. It is the only variable in investing that is known with certainty in advance, which is why it deserves more attention than most forecasts.

An estimate of what a business is worth based on what it will earn, independent of its price.

Every fair value is a model with assumptions inside it — growth, margins, a discount rate — and small changes in those assumptions move the answer enormously. Treat it as a range and a starting point for argument, never as a number.

A price range that got skipped so fast no two-sided trading happened there.

Three bars where the first and the third do not overlap leave a gap in the middle where buyers and sellers never met. The idea is that price tends to return and fill it. It is a description of what happened, not a prediction, and plenty of gaps are never filled.

The shares actually available to trade, excluding insider and restricted holdings.

Always smaller than shares outstanding, and the difference matters: a small float means the same order moves the price much further, and it is why short interest as a percentage of float is a very different number from short interest as a percentage of shares outstanding.

The Federal Reserve committee that sets the US policy interest rate.

Meets eight times a year. What moves markets is rarely the decision, which is usually priced, but the projections and the press conference — the difference between what was expected and what was said.

The filing an officer, director or 10% owner must make within two business days of trading their own stock.

Far faster than a 13F and far more pointed. A single sale says little — it can be taxes, a scheduled plan, or diversification. What speaks is the cluster: several different insiders buying the same issuer within days.

P/E using next year's estimated earnings instead of the last reported ones.

More useful than trailing P/E for a business that is changing fast, and less trustworthy, because the denominator is an analyst's opinion rather than a filed number. When forward is much lower than trailing, the market is being told earnings will jump; whether it happens is the whole question.

Operating cash flow minus capital spending — the money genuinely available to owners.

Harder to manipulate than earnings because accruals wash out. A company whose profits rise while free cash flow falls is either investing heavily for the future or booking revenue it has not collected — and telling those apart is most of fundamental analysis.

How fast delta changes as the underlying moves — the acceleration of the position.

Highest at the money and near expiry, which is why 0DTE contracts have so much of it. Gamma is what forces a hedged dealer to keep re-trading: as price moves, the hedge that was right a minute ago is wrong, and the correction is a real order in the real market.

The price at which total dealer gamma crosses from positive to negative.

Above it, hedging damps moves and the market tends to grind; below it, hedging feeds moves and ranges expand. It is not a support level and it does not hold price — it is a change in the character of the tape.

The total gamma dealers are carrying across all strikes, expressed in dollars per 1% move.

When dealers are net long gamma they hedge against the move — selling into strength, buying weakness — and volatility gets suppressed. When they are net short they hedge with the move, and every push gets amplified. Positive GEX is a pinned market; negative GEX is where the fast days come from.

The 50-day average crossing above the 200-day. The reverse is the death cross.

The most reported signal in financial media and one of the slowest: by the time it triggers, the move that caused it is months old. Its value is as a regime label — which side of the trend you are on — not as an entry.

The sensitivities of an option's price: delta, gamma, theta, vega and rho.

Each answers one question. Delta: how much does the option move when the underlying moves a dollar. Gamma: how fast does delta itself change. Theta: how much value bleeds away per day. Vega: how much does implied volatility matter. Rho: interest rates, which almost never matters until it does. Every option position is a bundle of these five exposures whether the holder thinks in them or not.

What is left of each sales dollar after the direct cost of producing it.

The first and most durable signal of pricing power. Margins that hold through a recession say customers cannot easily leave; margins that compress every year say competition is winning, whatever the revenue line does.

Management's own forecast for coming quarters.

Usually moves the stock more than the quarter just reported, because the quarter is history and the guide is the new base for every model. Withdrawing guidance is a signal in itself, and rarely a good one.

A private fund for qualified investors, free to go short, use leverage and hold almost anything.

The classic fee is 2% of assets and 20% of profits, usually above a high water mark so the manager is not paid twice for the same gain. The freedom to short is what the word 'hedge' originally meant; most of the industry stopped hedging long ago.

The volatility the option's price implies the market expects, going forward.

Not a forecast anyone made — it is solved backwards from the price. It rises before events and collapses after them, which is why a correct directional bet through earnings can still lose money: the move happened, and the premium you paid for it evaporated.

The share of a company held by funds, banks and other professional managers.

Built by adding up every 13F. High ownership means the stock is well understood and heavily analyzed; it also means everyone owns it already, which limits who is left to buy. The change matters more than the level, and it must be measured on managers present in both quarters — filers reorganize, and comparing totals reads a corporate restructuring as a sell-off.

Operating profit divided by interest expense — how many times over the company can pay its interest.

The number that tells you whether debt is a tool or a threat. Under 2× a bad quarter becomes a solvency conversation with the banks; above 8× the debt is barely noticeable. It moves faster than the debt ratio because it responds to earnings, not just to the balance sheet.

How much you can trade without moving the price.

Abundant when you do not need it and absent when you do. Every liquidity crisis is the same story: an asset that traded freely for years stops having a bid on the one morning everybody wants to sell.

Borrowing from the broker to hold a position larger than your cash.

Multiplies gains and losses by the same factor, and adds a third outcome that cash never has: being forced out at the worst moment. Leverage does not change the odds of being right, only the cost of being early.

The broker demanding more collateral because the position moved against you.

You add cash or the broker liquidates — at market, at whatever price exists, usually into the same panic that caused the call. It is the mechanism that turns a temporary drawdown into a permanent loss, and it is why leverage kills people who were eventually right.

The gap between what you pay and what you think it is worth — room to be wrong.

Graham's central idea, and the one that survives everything else. You do not buy because your estimate says the stock is worth $100 and it trades at $98; you buy at $60, so that being wrong by a third still leaves you whole. The margin is not for precision — it is for the errors you have not thought of.

Share price times shares outstanding — what the equity of the company is worth at today's price.

The number people mean by 'how big is this company'. It is not what it would cost to buy it: that is enterprise value, which adds the debt you would inherit and subtracts the cash you would get.

A firm quoting both sides continuously and earning the spread for providing liquidity.

Not directional by intent: the market maker wants to be flat and collect the spread, which is exactly why its hedging is mechanical and predictable. That predictability is what makes dealer gamma positioning readable at all.

The largest peak-to-trough fall — the worst it got before recovering.

The number that decides whether a strategy is actually holdable. A 60% drawdown needs a 150% gain to get back to even, and almost nobody stays invested through one. Sharpe describes the ride; drawdown describes whether you would have stayed on it.

The strike where the largest total value of options would expire worthless.

Popular because the price often ends near it on expiry, and contested because the causation is unclear: hedging flows plausibly pull price there, but so does the fact that most open interest sits near where the stock has been trading anyway.

Whether the strike is favorable relative to the current price: in, at, or out of the money.

In the money contracts have real intrinsic value and behave more like the stock; out of the money contracts are pure time and volatility, and go to zero fast. At the money is where gamma and theta are both at their most violent.

Long positions minus short positions — how much market risk a fund is actually carrying.

A fund 130% long and 70% short is 60% net exposed but 200% gross — twice its capital is at work. Net says how much a market move hurts; gross says how much can go wrong inside the book even if the market goes nowhere.

The monthly count of US jobs added, excluding farms.

Released the first Friday of the month and revised heavily afterwards — the revisions often matter more than the original print and nobody reports them. Read alongside wage growth, which is the part that feeds inflation.

The number of contracts currently open and not yet closed or expired.

Different from volume: volume counts today's trades, open interest counts positions still standing. Critically, it is published once a night by the OCC — so every 'real-time' open interest number anywhere is last night's, repriced against today's spot.

Options expiration — the third Friday of each month, when monthly contracts settle.

Large blocks of open interest disappear at once, and the hedges attached to them get unwound. Quarterly expirations, where index futures and options expire together, are the ones that move volume most. When the third Friday is a holiday, it shifts to Thursday — and a lot of calendars publish it wrong.

A contract giving the right, not the obligation, to buy or sell at a set price before a date.

One contract normally covers 100 shares. The buyer pays a premium and can lose only that; the seller collects it and takes on an obligation whose loss can be far larger. Everything else in options is arithmetic on top of that asymmetry.

Price divided by book value per share — what the market pays for a dollar of accounting net worth.

Graham's favorite. It works for banks and insurers, whose assets are financial and marked near their real value, and it fails for anything whose value is people, brands or code, because accounting does not put those on the balance sheet. A software company at 40× book is not necessarily expensive.

Price divided by earnings per share — how many years of current profit you are paying for one share.

A P/E of 25 means you pay $25 for each $1 of annual profit. High multiples say the market expects growth; low ones say it expects trouble, or has stopped paying attention. It is meaningless across industries — a utility at 18 is expensive, a software company at 18 is cheap — and it breaks entirely when earnings are near zero or negative.

Price divided by revenue per share. Works when there are no profits yet.

The blunt instrument for companies that do not earn money: revenue is harder to manipulate than profit and always positive. Its blindness is that it ignores margins entirely — a dollar of software revenue and a dollar of grocery revenue are worth wildly different amounts and P/S treats them the same.

The share of earnings paid out as dividends.

Above 100% the company is paying shareholders with borrowed money or with cash it needed elsewhere, and that ends one of two ways. Below 40% there is room to keep raising the dividend for years, which is what compounds.

P/E divided by the earnings growth rate — a P/E adjusted for how fast profits grow.

Peter Lynch popularized the rule that a PEG near 1 is fair. A company at 40× earnings growing 40% a year has the same PEG as one at 10× growing 10%. The weakness is that growth rates are estimates and they mean-revert hard, so a low PEG built on a 60% growth assumption is not a bargain.

How much to put in a single trade — usually the most consequential decision.

The Kelly criterion gives the mathematically optimal fraction given an edge, and almost nobody trades full Kelly because its drawdowns are unbearable even when they are correct. Half Kelly gives three quarters of the growth with far less pain, which is the trade most professionals actually make.

The right to sell at the strike price. Gains value as the underlying falls.

Used to bet on a fall or to insure a position you own. Buying puts as insurance costs money every month it is not needed, which is exactly what insurance is — and why most people stop paying for it right before they need it.

A central bank buying bonds to push down long-term rates when short rates are already at zero.

Invented for Japan, industrialized by the Fed after 2008, and unwound as quantitative tightening from 2022. Its effect on the real economy is argued about; its effect on asset prices is not.

How much you can make against how much you can lose on a trade.

Meaningless without the probability attached. A 5:1 ratio that works one time in ten is a losing strategy; a 1:1 that works seven times in ten is a good one. The two numbers only mean something multiplied together.

Profit as a percentage of shareholders' equity — how well the company turns owner capital into earnings.

Above 20% sustained is the mark of a business with a real advantage. But leverage inflates it — borrow enough and a mediocre business shows a spectacular ROE — so it must be read next to debt. That is why ROIC is often the better question.

Gain divided by what you put in — the plain percentage return.

Its weakness is that it ignores time entirely: 50% in one year and 50% in ten years are the same ROI and are not remotely the same investment. For anything spanning more than a year, use CAGR.

Operating profit against all capital employed, debt included — the cleanest measure of business quality.

The number that says whether a company creates value: if ROIC exceeds the cost of capital, growth makes shareholders richer; if it does not, growth destroys value and the company should shrink. It is unfoolable by leverage, which is why it is harder to game than ROE.

A 0–100 oscillator measuring the speed of recent gains against recent losses.

Above 70 is conventionally overbought and below 30 oversold, which is exactly where the indicator misleads most: in a strong trend RSI stays pinned above 70 for weeks while the price keeps rising. Its real use is divergence — price making a new high that RSI does not confirm.

Return above the risk-free rate divided by volatility — return per unit of wobble.

Above 1 is good, above 2 is excellent, above 3 sustained deserves a hard look at whether the strategy is quietly selling insurance — steady small gains with a rare catastrophic loss produce a beautiful Sharpe right up until the day they do not.

The number of shares sold short and not yet bought back.

Reported to the regulator twice a month, so it always lags. High short interest is read as bearish conviction and as fuel: those shares must eventually be bought back, and if the price rises first they get bought back in a hurry.

A rising price forcing shorts to buy back, which pushes the price higher still.

Self-reinforcing while it lasts, and it needs three things together: heavy short interest, a small float, and a catalyst. It ends when the forced buyers are done, usually far above anything justified by the business.

The average closing price over a set number of bars, redrawn each bar.

The 50 and the 200 are watched because so many people watch them, which makes them self-fulfilling at the margin. A moving average is a lagging summary of where price has been, never a forecast of where it goes.

The 2012 law requiring members of the US Congress to disclose their trades within 45 days.

The filings are public and machine-readable, which is why retail traders now follow them. The trades are legal and disclosed; the debate is whether they should be allowed at all.

A standing order to exit if price reaches a level, deciding the loss in advance.

Its real function is removing the decision from the moment you are least able to make it. Placed by a round number rather than by volatility, it gets hit by noise — which is why ATR, not percent, is the right unit.

The price at which the option can be exercised.

Its distance from the current price decides almost everything about the contract: how much it costs, how fast it moves, and how likely it is to expire worthless. Strikes cluster at round numbers because that is where the exchange lists them, and that clustering shows up in dealer positioning.

A trailing line built from ATR that flips sides when price closes through it.

Below price in an uptrend, above it in a downtrend, and it only flips on a close — which is what keeps it from whipsawing on every intraday spike. It works in trends and gets chopped up in ranges, like everything of its family.

How much value the option loses per day just from time passing.

The buyer's rent and the seller's income. It accelerates as expiry approaches, and it does not care whether you were right about direction: an option can expire worthless on a stock that moved exactly the way you predicted, just not fast enough.

How much the option price changes when implied volatility moves one point.

Largest for longer-dated at-the-money contracts. It is why buying options right before an earnings release is expensive and selling them right after is not free money: everyone can see the same collapse coming.

The market's expected 30-day volatility on the S&P 500, from option prices.

Below 15 is complacency, above 30 is fear, above 50 is panic. It is mean-reverting and it spikes far more violently than it decays, which is why selling volatility looks like free money for years and then is not.

How much returns scatter around their average, usually annualized.

Finance uses it as a synonym for risk because it is measurable, not because it is the same thing. Volatility is symmetric — it counts a 10% rise as much as a 10% fall — and nobody has ever complained about upside risk.

Volume traded at each price level rather than at each moment in time.

Turns the chart sideways and asks where business actually got done. The Point of Control is the price with the most volume — the level the market agreed on — and the Value Area is the band holding 68% of it. Prices with little volume beneath them tend to be crossed quickly.

Treasury yields plotted from short maturities to long ones.

Normally upward sloping. When short rates exceed long ones the curve is inverted, which has preceded every US recession of the past half century — with lags of anywhere from six months to two years, which makes it useless as a timing tool and impossible to ignore.

An option expiring today. Zero days to expiration.

SPX and SPY now list expirations every weekday, and same-day contracts have become a very large share of index option volume. They have almost no time value left, so gamma is enormous and the price responds violently to small moves in the underlying. That concentration of gamma is what makes dealer hedging move the index in the last hours of the session.

The audited annual report a US public company files with the regulator.

The primary document, and the one almost nobody reads. The risk factors section and the footnotes are where the interesting things are — Enron's related-party transactions were disclosed in the footnotes for anyone who looked.

The quarterly filing every manager with over $100m in US equities must make, listing their positions.

Long US equity positions only. No shorts, no bonds, no currencies, nothing outside the US — and it arrives up to 45 days after the quarter closed. A fund that looks loaded up on something may be fully hedged with options the form does not require. It reads conviction and rotation, not what anyone holds today.