Trading Terms — the Index — Henny — Heinrich Songmin Berger

Every scene has its language, and trading’s is especially impenetrable: half exchange English, half forum slang, in between metrics that sound like laws of nature and are conventions. Newcomers understand nothing at first — and many veterans use half the words wrong without noticing.

This is the index against that: the terms of trading and the crypto space, explained briefly — with what they actually mean, which convention stands behind them, and where the usual half-knowledge goes astray. Slang entries state whether the origin is documented or just a common story — that’s researched, not guessed.

Three things up front. First: definitions describe what terms mean — not a single one is a recommendation to act. That this page explains what a stop-loss is doesn’t mean you should trade. Second: much of this is convention, not natural law — where a definition is contested or fuzzy (altseason, capitulation, confluence), the fuzziness is part of the entry. Third: this index is not finished and never will be — it grows, like the AI tools index does, and every extension lands in the version history at the end of the page.

Every term has its own anchor — you can link to each one directly (the # next to the name).

98 terms

Basics & market

# Long / Short: Long is the bet on rising prices, short the bet on falling ones. Going short means selling something borrowed to buy it back cheaper later — in crypto practice usually via derivatives rather than actual borrowing. Almost the entire derivatives vocabulary (squeeze, funding, liquidation) presupposes these two words.
# ATH / ATL: All-time high and all-time low — the highest and lowest price ever traded. Sounds trivial but is index-dependent: exchanges differ on historical extremes, which is why serious sources name the exchange or index alongside.
# Market Cap / FDV: Market cap = price × circulating supply. FDV (fully diluted valuation) = price × the total of all tokens that will ever exist. A large gap between the two means many tokens are still waiting to enter circulation — dilution is scheduled.
# Circulating Supply: The number of publicly tradable tokens — excluding locked, vesting, burned or provably lost holdings. Important: for projects with team and treasury holdings this is a data provider's estimate, not an exact on-chain number.
# Bitcoin Dominance: Bitcoin's share of the total market capitalization of all tracked cryptocurrencies. The value depends on the universe (does the provider count stablecoins?) — different data sources therefore show different dominance figures.
# Altseason: Ambiguous — and that ambiguity is part of the definition. Loosely: a phase in which altcoins broadly beat Bitcoin. The most common counting convention (Blockchaincenter index): altseason when at least 75% of the top 50 (excluding stablecoins) have outperformed Bitcoin over 90 days; below 25% counts as Bitcoin season. Other indices count differently and can contradict each other.
# DCA: Dollar-cost averaging: investing fixed amounts on a fixed schedule instead of guessing the "right" moment. It describes a method of execution, not a judgment on whether an asset is worth buying — the schedule replaces the timing decision, not the selection decision.
# Fear & Greed Index: A sentiment gauge (0–100) that blends several data sources — volatility, volume, social media, dominance, surveys — into one mood score (best-known version: alternative.me). It measures mood, not the future; the weighting is a provider convention. More in the post →
# Capitulation: A deliberately fuzzy term with three common readings: ① the final flush in which even the patient lose their nerve (price capitulation), ② miner capitulation when unprofitable miners switch off, ③ mass realized losses on-chain. Which one is meant often only context reveals — asking is not a crime. More in the post →
# Drawdown: The decline from a peak to the following trough, in percent; the maximum drawdown is the largest such decline before a new high. The uncomfortable arithmetic behind it: after −50% it takes +100% to get back to even — losses and required recovery are asymmetric. More in the post →

Orders & execution

# Market Order: Buys or sells immediately against the best available resting orders. It guarantees execution, not price — in a thin order book it eats through several price levels (slippage). Side note: professional venues like CME only execute market orders with a protection band; crypto exchanges mostly don't.
# Limit Order: Defines the worst acceptable price: a maximum when buying, a minimum when selling; fills only at the limit or better. The mirror image of the market order: price guaranteed, execution not — if the limit is never reached, nothing happens, and partial fills are normal.
# Stop Order: Rests invisibly outside the order book and only activates once the market trades at the trigger price — then it becomes a market order. The usually missed detail: the stop price is a trigger, not a guaranteed fill price. Across gaps or thin liquidity, the fill can be considerably worse.
# Stop-Limit Order: At the trigger, a limit order is placed instead of a market order — stop price and limit price are two separate parameters. That protects against slippage but trades the risk: if the market falls straight through the limit, the order stays unfilled and the position unprotected.
# Take-Profit / Trailing Stop: Take-profit closes a position automatically at a defined profit level. A trailing stop drags the stop mark behind the price at a fixed distance: if the market runs your way, the stop follows; if it reverses by the set distance, the position closes. Both inherit the execution mechanics of their base type (market or limit).
# OCO: Two linked orders — typically profit target plus protection — of which only one can live: once either is filled even partially, the system cancels the other automatically. Common misconception: that only a full fill counts — a partial fill already kills the sibling order.
# Time-in-Force (GTC · IOC · FOK): How long an order lives: GTC (good-till-cancelled) stays in the book until revoked, IOC (immediate-or-cancel) fills what it can instantly and discards the rest, FOK (fill-or-kill) demands an immediate full fill or nothing. Inconspicuous parameters with a big effect on what actually happens.
# Post-Only: An order flag guaranteeing the order enters the book as a maker: if it would execute immediately (i.e., as a taker), it is rejected instead. Exists precisely because of maker/taker fee logic.
# Maker / Taker: Makers add liquidity to the order book (orders that aren't immediately executable), takers remove it (orders that match instantly). Exchanges subsidize deep books, so makers pay less — sometimes even earning rebates. The most important anti-half-knowledge point: maker/taker is not a property of the order type — a limit order is a taker too if placed at an immediately executable price.
# Order Book & Depth: The order book is the price-sorted list of all resting limit orders (bids and asks); market orders match against it. Depth shows how much volume waits at each price level — and thus how much you can trade without moving the price yourself. Depth and slippage are two sides of the same thing; visible depth can also be distorted by spoofing.
# Spread: The gap between the highest bid and the lowest ask — both a liquidity measure and an implicit trading cost: buy and immediately sell, and you lose exactly the spread. Ambiguity alert: in futures contexts, "spread" also means calendar and product spreads.
# Slippage: The difference between expected and actual execution price — not an exchange error but a direct consequence of order book depth: large market orders work through several price levels, making the average price worse than the best quote first seen. On-chain adds a second source (see on-chain slippage).

Technical analysis

# Candlestick / OHLC: One candle summarizes a time window's trading in four values: open, high, low, close. Body = open to close, wicks = the extremes beyond. The entire TA vocabulary presupposes this format — and every candle looks different on every timeframe.
# Timeframe: The time window per candle — from one minute to one week. The same market can look like a downtrend on the 15-minute chart and an uptrend on the weekly; without a timeframe, almost any chart statement is incomplete.
# Support / Resistance: Price areas where demand has historically stopped declines (support) or supply has stopped advances (resistance). The standard references explicitly work with zones rather than exact lines — prices briefly overshoot without the level being "broken." After a genuine break, the levels conventionally swap roles.
# Trend / Range: The classic convention since Dow theory: uptrend = a sequence of higher highs and higher lows, downtrend = lower highs and lower lows, range = oscillating between horizontal support and resistance without new extremes. Trendlines conventionally connect at least two lows or two highs.
# Breakout: The move above a resistance or below a support area, read as the possible start of a continuation. Convention demands confirmation — closes beyond the level, ideally with volume — because mere intraday violations frequently fail.
# Fakeout: A failed breakout: price leaves the structure and returns shortly after. Fakeouts are why confirmation conventions exist. Honest footnote: there is no objective threshold at which a breakout officially becomes a fakeout — the fuzziness is part of the term.
# Retest: The return to a broken level that now conventionally plays the opposite role: old resistance gets approached as new support (and vice versa). If the level holds, the breakout counts as confirmed. Not a law of nature — not every breakout gets retested.
# Volume Confirmation: The convention since Dow theory: volume should confirm the trend — moves on rising volume count as strong, breakouts on thin volume as fragile. Important for honesty: this is a documented analysis convention, not a causal law — its predictive value is empirically contested.
# Divergence: Price and indicator part ways: a new price high without a new indicator high (bearish) or a new price low without a new indicator low (bullish) — read as fading momentum. The documented caveat belongs in the definition: in strong trends, divergences are notoriously misleading and can stack up long before anything turns.
# Confluence: Several independent signals meeting at the same price area — say, horizontal support plus a moving average plus a Fibonacci level. Deliberately informal umbrella term: there is no standard for how many signals make "confluence."
# RSI: Momentum oscillator by Wilder (1978), standard 14 periods, scale 0–100; above 70 conventionally counts as "overbought," below 30 as "oversold." The key anti-half-knowledge point sits in the standard reference itself: overbought is no automatic sell signal — in strong trends the RSI stays elevated for long stretches, and an overbought RSI can even signal strength.
# MACD: Trend-following momentum indicator (Appel): MACD line = difference of two EMAs (standard 12/26), plus a signal line (9-period EMA of the MACD line) and histogram. Read via crossovers, zero-line crossings, divergences. Built from moving averages, it lags — it confirms rather than predicts.
# SMA / EMA: Both smooth price: the SMA weights all periods equally, the EMA weights recent ones more and reacts faster — at the cost of more noise. Both lag by construction; the usual periods (20/50/200) are convention, not law.
# Golden / Death Cross: When the shorter moving average crosses above the longer (classically 50 over 200 days), that's a golden cross, read as bullish; the death cross is its counterpart. Built from heavily smoothed past data, both are late signals: they confirm a shift already underway rather than announce it.
# Bollinger Bands: A middle band (usually a 20-period SMA) plus two bands at a default distance of two standard deviations; they widen with volatility. The inventor's own rules: the bands provide a relative definition of high and low — touching the upper band is no sell signal by itself, and in trends price can "walk the band."
# VWAP: The volume-weighted average price of the day: the sum of price × volume of all trades divided by total volume; resets daily. Institutionally above all a benchmark for execution quality — and part of real exchange mechanics: CME uses VWAP formulas for settlement prices.
# Fibonacci Retracement: Horizontal levels at 23.6/38.2/50/61.8% of a prior move, used as zones where a correction might end — per the standard reference "alert zones," not hard reversal points. Two honest footnotes: the 50% level isn't a Fibonacci ratio at all, and there is no proven mechanism behind the levels — the documented common explanation is coordination: they work because many people watch them. More in the post →
# Wick: The thin lines above and below a candle body — extremes that traded but didn't hold. The scene calls extreme outliers in illiquid moments "scam wicks" — they sweep stops and liquidations; that's exactly why derivatives exchanges price liquidations on the mark price rather than the last trade.

Risk & statistics

# Risk/Reward (CRV): Relates the potential loss (entry to invalidation point) to the potential gain (entry to target). The central misconception: a "good" R/R alone says nothing about profitability — it's only readable together with the win rate. A high R/R with a low win rate can lose, and vice versa.
# Position Sizing: Deriving position size from account risk and stop distance instead of gut feeling: first fix the maximum amount at risk (a common convention: a fixed percentage of the account), then the invalidation point — size follows as the risked amount divided by the distance to it, fees and slippage included. A description of a method, not a recommendation.
# Win Rate & Expectancy: The win rate alone says nothing: what matters is expectancy = win rate × average win − loss rate × average loss. A system with 30% winners can be profitable, one with 90% can lose money. It only becomes meaningful over larger samples — single trades prove nothing.
# Backtest & Overfitting: A backtest applies a fixed rule set to the past — it proves nothing about the future. The documented main trap is overfitting: parameters get tuned until they model the noise of history; the curve looks great, live it breaks. Add survivorship bias (only survivors tested) and look-ahead bias (using data unavailable at the time). More in the post →
# Paper Trading: Simulated trading without real money — in the documented testing sequence, the forward test after the backtest. Its limit belongs in the definition: execution reality (slippage, partial fills) and above all the psychological pressure of real money are only partially reproduced.

Leverage & derivatives

# Leverage: Trading with more capital than you put up: at 10x, a position moves ten times your stake — gains and losses alike. The difference between nominal leverage (the setting) and effective leverage (position value divided by total capital) is chronically overlooked. At length and in peace: in the leverage post. More in the post →
# Initial / Maintenance Margin: Initial margin is the collateral that opens a position; maintenance margin the minimum that keeps it alive. If collateral falls below it, liquidation begins — which is why you get liquidated before zero, not at zero. The rates rise in tiers with position size.
# Cross / Isolated Margin: Isolated: the position has a fixed margin budget — liquidation costs only that budget. Cross: all positions share the entire account as collateral — lowering single-position risk while putting everything at stake in the extreme. Not "better or worse" but two different failure modes.
# Liquidation: The forced closure of a leveraged position by the exchange once margin falls below maintenance margin — triggered on the mark price, not the last traded price, and with an extra liquidation fee. At 10x the point sits roughly at −9 to −10% of price movement, at 100x below −1%. More in the post →
# Insurance Fund: The derivatives exchange's safety buffer: if a position is closed worse than its bankruptcy price, the fund covers the shortfall; it is fed, among other things, by liquidation fees and residual margins. If it runs short, the last stage kicks in: ADL.
# ADL: The last step of the liquidation chain: if the insurance fund can't absorb a bankrupt position, the exchange forcibly closes profitable opposing positions of other traders — the most profitable and highest-levered first. You can get closed out despite being right.
# Perpetual Futures: Futures without an expiry date — instead of settlement, the funding mechanism ties the price to the underlying. The first crypto perp was BitMEX's XBTUSD swap (May 2016); today perps are the space's dominant leverage instrument.
# Funding Rate: A periodic payment directly between longs and shorts (not an exchange fee) that anchors the perp price to the index — typically every 8 hours, per contract also 4 or 1. A subtlety that often gets lost: besides the premium, a fixed interest component enters the rate — so it can be slightly positive even when the perp isn't trading above the index.
# Open Interest: The total of all open, un-closed contracts of a derivative. Rising OI means new capital in the market; falling OI, positions being closed. What OI never tells you: direction — every contract has exactly one long and one short, so it cannot measure an "imbalance."
# Long / Short Squeeze: A short squeeze occurs when rising prices force shorts to buy back — voluntarily or via liquidation — fueling the rise further; the long squeeze is the mirror image downward. In perp markets, squeezes often run as liquidation cascades: forced closures fire market orders that trigger the next forced closures.
# Mark / Index / Last Price: Three prices, three jobs: last price = the latest trade on this exchange (what you trade at). Index price = a weighted spot average across exchanges. Mark price = a smoothed "fair" price from index plus a basis component — liquidations and open-position valuation run on it, so a single order book outlier (a "scam wick") doesn't liquidate anyone.
# Basis / Contango / Backwardation: The basis is the difference between futures and spot price: futures above spot is contango, below is backwardation. In crypto, the annualized basis of dated futures is read as a gauge of leverage demand — and is the foundation of the cash-and-carry trade.
# Cash-and-Carry: The classic basis strategy as a mechanism: buy spot, simultaneously sell the dated future, and collect the price difference (basis) until expiry — market-neutral to direction, but not risk-free (counterparty, margin and execution risk remain). Described here as a term, not a recommendation.

On-chain

# UTXO: Bitcoin's accounting model: no accounts, but unspent transaction outputs — an address's "balance" is the sum of its UTXOs. Because every UTXO carries a timestamp and cost basis, many on-chain metrics (coin age, realized cap, SOPR) build precisely on this. More in the post →
# Mempool: Every node's waiting room for valid, still-unconfirmed transactions. Miners pick the highest fees first — a full mempool means costlier, slower transactions. There is no single "the" mempool: every node holds its own view. More in the post →
# Gas: Ethereum's unit of computational effort: fee = gas used × gas price. Since EIP-1559 (2021), the price consists of a protocol-set base fee, which is burned, plus a voluntary priority fee to the validator.
# Hashrate: The estimated total computing power of a proof-of-work network in hashes per second. It isn't measured but inferred from difficulty and observed block times — daily values are estimates with visible noise.
# Halving: Every 210,000 blocks (roughly four years), Bitcoin's block subsidy halves — fixed in the protocol. Most recently in April 2024 to 3.125 BTC; the next is expected around 2028. The series' result: the hard cap of just under 21 million. More in the post →
# Realized Cap / Price: Values every coin not at the current price but at the price of its last on-chain move — an aggregate cost basis of the market that down-weights lost old holdings (concept: Nic Carter & Antoine Le Calvez, 2018). Realized price = realized cap divided by supply.
# MVRV: Market value to realized value — market cap divided by realized cap (Mahmudov & Puell, 2018). Above 1, the average holding sits in unrealized profit. Historical extreme zones are often read as overheating or capitulation — that's interpretation in hindsight, not a property of the metric.
# SOPR: Spent Output Profit Ratio (Shirakashi, 2019): per moved UTXO, the ratio of sale value to cost basis. Above 1, moved coins sold at an average profit; below 1, at a loss. Variants compute adjusted (aSOPR) or by holder cohort (STH/LTH SOPR).
# NUPL: Net unrealized profit/loss: the share of market cap consisting of unrealized profit or loss — computed as (market cap − realized cap) divided by market cap. The popular zone names ("belief," "capitulation") are after-the-fact labels, not measurements.
# LTH / STH: Glassnode's split at roughly 155 days of holding: statistically, coins become increasingly unlikely to move after that quiet period. Important: a data provider's heuristic, not an on-chain property — other providers draw other lines. More in the post →
# Exchange Flows: On-chain volume onto addresses the data provider has labeled as exchange addresses (inflow) or away from them (outflow); netflow is the difference. Two caveats belong in the definition: labels are mutable — values get revised retroactively — and the reading "inflow = sell pressure" is interpretation, not measurement. More in the post →

DeFi & tokens

# AMM: Replaces the order book with a pricing formula against a liquidity pool — in Uniswap v2, x·y=k keeps the product of the two reserves constant; every swap shifts the ratio and thus the price. Your counterparty is the smart contract, not another human.
# Liquidity Pool: The smart contract reserve of usually two tokens an AMM trades against. Liquidity providers deposit both sides, receive shares and a cut of trading fees — and carry impermanent loss risk in return.
# Impermanent Loss: The difference between the value of an LP position and simply holding the same tokens — it arises as soon as the pool tokens' price ratio shifts. "Impermanent" because it vanishes if the ratio returns; it becomes realized (permanent) if liquidity is withdrawn before that. Whether fees compensate for it is an empirical question, not a definitional one.
# Slippage (on-chain): On-chain, slippage has two sources: your own order's price impact against the pool formula (the shallower the pool, the larger) and price movement between submission and block inclusion. The adjustable slippage tolerance is a protection — set too generously, it becomes an invitation for sandwich attacks.
# MEV / Sandwich: Maximal extractable value: profit block producers or bots extract by viewing, reordering, or bracketing transactions with their own. The sandwich attack is the best-known form: buy just before the victim's trade, sell right after — paid out of their slippage tolerance.
# TVL: Total value locked: the current dollar value of assets in a protocol's smart contracts. A convention-dependent number — whether staked governance tokens, lent assets or double-counted wrapped positions are included is defined by each aggregator.
# Airdrop: Free token distribution to wallets — as marketing, for distribution breadth, or as retroactive reward for early users. Around expected airdrops, "farming" regularly emerges: activity performed solely to qualify.
# Vesting / Unlocks / Cliff: Vesting is the staged release of team and investor tokens; the cliff is the initial lock period, after which a large block often frees up at once; unlocks are the release dates themselves. Every unlock increases tradable supply — which is why tokenomics trackers keep calendars of them. More in the post →
# Depeg: When a stablecoin deviates persistently from its peg. Small, short-lived deviations are everyday market noise; exactly where a "deviation" becomes a "depeg" is convention. The canonical example remains TerraUSD's collapse in May 2022. More in the post →

Manipulation & fraud

# Rug Pull: An exit scam: the team drains the liquidity or vanishes with the funds — linguistically, the rug yanked from under you. Established as a fraud category since DeFi summer 2020 and tracked as its own category by analytics firms; the linguistic first use is undocumented.
# Pump and Dump: Coordinated inflation of an illiquid asset (pump) in order to sell into the self-created demand (dump) — the buyers at the top are the exit liquidity. Classic market manipulation in regulated markets; in crypto, the pattern thrives in group chats around small coins.
# Wash Trading: Trading with oneself (or in collusion) to fake volume and activity — on exchanges for prettier statistics, in NFTs for apparent demand. The reason raw volume figures of smaller venues deserve suspicion.
# Spoofing: Placing large orders never meant to execute — only to fake depth and interest, pulling them before they fill. The reason visible order book depth is no reliable picture of real demand.
# Shill origin documented: Promoting an asset self-interestedly or for pay, often without disclosure. Pre-crypto slang with a documented history: the "shill" was the decoy in carnival and gambling cons (dictionary first attestation: early 20th century).

Scene slang

# HODL origin documented: Holding, come what may. Origin documented: the bitcointalk thread "I AM HODLING" by GameKyuubi, December 18, 2013 — a typo for "holding" in a self-declaredly whisky-soaked rant mid-crash. The backronym "Hold On for Dear Life" came later.
# FOMO origin documented: Fear of missing out — the fuel of procyclical buying. Older than crypto and documented: coined by Patrick McGinnis in the Harvard Business School student paper, 2004.
# FUD origin documented: Fear, uncertainty and doubt — crypto's blanket label for negative news or targeted unsettling. Documented as a tech term: popularized in 1975 by Gene Amdahl, for the tactic IBM sales used to keep customers off competitors' products.
# WAGMI / NGMI common story: "We're All Gonna Make It" as the community's rallying cry, "Not Gonna Make It" as its mocking counterpart. Origin: common story without a documented source post — attributed to the bodybuilding scene around Zyzz, adopted by crypto and NFT Twitter from 2021.
# Diamond / Paper Hands common story: Diamond hands hold through every drawdown; paper hands sell at the first headwind. Partly documented: traceable on WallStreetBets from around 2018, mainstreamed by the GameStop squeeze in January 2021 — migrated from stock meme culture into crypto, not the other way around.
# Ape / Degen common story: To "ape in" is to enter big without analysis; "degen" (from degenerate gambler) is the self-ironic job title for it, established in DeFi summer 2020. Both origins: common story, source posts missing.
# Rekt origin documented: From "wrecked": liquidated, total loss, done for. Documented as gaming slang (traceable from 2011/2012), adopted from there into crypto jargon — the crossover itself can't be dated.
# DYOR common story: "Do Your Own Research" — officially the request to verify claims yourself; in practice often the liability boilerplate under a shill post. Origin undocumented; precursors existed in classic finance forums.
# GM common story: "Good morning" as a greeting and belonging ritual on crypto Twitter and in Discords, mass-adopted since 2021. Function: community signal, not information. Origin undocumented.
# Bagholder common story: Someone sitting on heavily fallen positions — the "heavy bags." Derived from the old English idiom "left holding the bag"; a market-specific first attestation is missing.
# Exit Liquidity common story: Derogatory: the buyers whose demand funds insiders' and large holders' exit at high prices — "don't be someone's exit liquidity." Established jargon without a documented origin.
# Whale common story: An actor whose holdings are large enough to move prices. The casino-jargon origin is a common story without first attestation — the entire sea-creature hierarchy behind it gets taken apart in a post of its own. More in the post →
# Moon / Wen Lambo common story: "To the moon" celebrates or hopes for the vertical rise; "wen lambo?" asks ironically when the gains will cover the Lamborghini. Both memes: mid-2010s, source posts undocumented.
# Flippening common story: The hypothetical event of Ethereum overtaking Bitcoin by market capitalization — generalized, any such swap of places. Widespread since mid-2017; who coined the word is undocumented.
# Sats / Stacking Sats origin documented: A satoshi is Bitcoin's smallest unit: 0.00000001 BTC. The unit is documented: bitcointalk user ribuck proposed it in 2010/2011, and the hundred-millionth version stuck. "Stacking sats" — regularly accumulating small amounts — became popular around 2019; no single first attestation exists.

The terms are the toolkit — know them and the rest starts to make sense.

How this index grows

Individual terms become posts of their own when they deserve the space — leverage already has one, so does the whale hierarchy. Conversely, new terms migrate in here as they keep crossing my path. If one is missing, or one is wrong: write to me — corrections are the fastest way this page gets better.

And if you’d rather have the most important of these terms on your wall than in a browser tab: the Crypto Collection includes the Trading Cheat Sheet — a tool, not investment advice, exactly like this index.

Sources and limits

The definitions are checked against standard references: exchange documentation (CME, Binance) for orders and derivatives mechanics, StockCharts ChartSchool and John Bollinger’s original rules for technical analysis, Glassnode documentation for on-chain metrics, Uniswap docs for DeFi, original forum posts for slang origins (the 2013 HODL post is directly verified). Where references diverge — execution mechanics are exchange-dependent, index conventions provider-dependent — the entry says so. A glossary remains a simplification all the same: every entry is the beginning of a topic, not its end.

Questions, or spotted a mistake? Write to me.