Reference

Trading glossary

Every term SIGNAL9 uses on a signal card, in the ATLAS Bot Ledger or in a guide, defined in plain English — with the practical caveat that matters, not just the textbook line. If a term you saw on the site is missing, tell us and we will add it.

SIGNAL9-specific terms

Words that mean something particular inside this terminal. If a number on the site confuses you, it is almost certainly defined here.

Confluence score: A 0–10 summary of how many independent pieces of evidence point the same way on a ticker — trend structure, momentum, volume, news tone and fundamentals. A 9 does not mean a 90% chance of profit; it means the inputs disagree very little at that moment.
Conviction gate: The floor a setup must clear before ATLAS is allowed to open it: model confidence of at least 72% and a reward-to-risk ratio of at least 1.8:1 measured against the buy zone. Setups below the gate are shown as watch-only rather than traded.
ATLAS Bot Ledger: The autonomous paper-trading account. Every entry and exit it makes is written to a public record at the time it happens, including the target and stop it was opened with. Its statistics are kept completely separate from human users' trades.
Buy zone: The price range where a thesis is valid. Chasing above the zone changes the reward-to-risk of the trade even when the analysis itself is unchanged, which is why ATLAS measures risk against the zone rather than the current spot price.
Watch-only: A setup the scanner finds interesting but which fails the conviction gate. It appears for context and is never counted in the track record.
Runner-up: A ticker that was scanned and rejected on the same pass that produced the published pick, shown so you can see what the model passed on and why.
Paper trading: Simulated trading with no real money. Fills are assumed at the quoted price, which excludes slippage, spread and commissions — real results are always somewhat worse.
XP and levels: A progression score for human users only, earned from manual paper trades and daily streaks. It measures activity and discipline, not skill, and it has no effect on any signal.

Price and chart structure

How a chart is described before any indicator is applied.

Trend: The direction of successive swing highs and lows. Higher highs with higher lows is an uptrend; the reverse is a downtrend; overlapping swings are a range, where most breakout strategies bleed.
Support / resistance: Price areas where buying or selling has repeatedly appeared. They are zones, not lines — treating them as exact prices is the fastest way to get stopped out by noise.
Breakout: Price closing beyond a defined range boundary. Without an expansion in volume, a breakout is more likely to be a liquidity grab than the start of a move.
Gap: A price jump between one session's close and the next session's open, usually on news or earnings. Gaps are why a stop-loss is an instruction, not a guarantee of exit price.
Candlestick: A bar showing the open, high, low and close of a period. The body shows where the period settled; the wicks show what was rejected.
Timeframe alignment: When the daily, hourly and 15-minute charts agree on direction. Setups where timeframes conflict get lower confidence in our pipeline.

Indicators

Every indicator is a transformation of price or volume. None of them add information the chart did not already contain; they make certain patterns easier to see consistently.

Moving average (MA / EMA): The average closing price over a lookback window, plotted as a line. An exponential moving average weights recent prices more heavily, so it turns faster and whipsaws more often.
RSI: Relative Strength Index, 0–100, comparing the size of recent gains to recent losses. Above 70 is conventionally 'overbought' and below 30 'oversold', but in a strong trend RSI can stay pinned for weeks — it is a momentum gauge, not a reversal signal.
MACD: The difference between two moving averages plus a signal line. Crossovers describe a change in momentum that has already begun; it is a confirming tool, not a leading one.
VWAP: Volume-Weighted Average Price — the session's average price weighted by traded volume. Institutional desks benchmark fills against it, which is why intraday price so often reacts around it.
RVOL: Relative volume: today's volume against the same time of day on an average day. Above roughly 2× means something genuinely changed; below 1× means the move lacks participation.
ATR: Average True Range, the typical distance price travels in a period. Used to place stops outside normal noise and to size positions to volatility rather than to a fixed percentage.

Risk and position sizing

The part of trading that decides survival. Sizing and exits matter more than entry selection over any meaningful sample.

Stop-loss: The price at which the thesis is wrong and the position is closed. It should be placed where the setup breaks structurally, then sized down to fit your risk — not tightened to justify a bigger position.
Take-profit / target: The price where the expected move is considered complete. Publishing it in advance is what makes a track record auditable.
Reward-to-risk (R:R): Distance to target divided by distance to stop. At 2:1 you can be right only about 40% of the time and still break even; at 0.5:1 you need to be right roughly 70% of the time.
R-multiple: A result expressed in units of the amount risked. A +2R trade earned twice what it risked, regardless of account size — this makes trades comparable across positions.
Fixed fractional risk: Risking the same small percentage of equity per trade, commonly 0.5–2%. Position size is then derived from the stop distance rather than chosen first.
Drawdown: The decline from an equity peak. Recovery is asymmetric: a 50% drawdown requires a 100% gain to get back to even.
Correlation exposure: Holding several positions that are really one bet — five AI semiconductor names move together. Correlated risk is additive even when each position is sized correctly on its own.
Slippage: The difference between the expected fill and the actual fill. It grows with position size and shrinks with liquidity, and it is the main reason paper results beat live results.

Fundamentals and market structure

Terms used when a signal is justified by the business rather than the chart.

Market capitalisation: Share price multiplied by shares outstanding. It sorts the universe: mega cap, large, mid, small and micro — each with different liquidity and gap risk.
P/E ratio: Price divided by earnings per share. Useful for comparing similar businesses; meaningless for companies with no earnings.
Free cash flow: Cash from operations minus capital expenditure — the money a business actually generates. Harder to flatter with accounting choices than net income.
Earnings date: The scheduled release of quarterly results. Holding a technical setup through earnings converts a chart trade into a coin flip on a report.
Float: Shares actually available to trade. A small float exaggerates both moves and slippage, which is why small caps need smaller positions.
Liquidity: How much you can trade without moving the price. Measured practically as average dollar volume; thin names invalidate normal stop placement.
Short selling: Selling borrowed shares in the expectation of buying them back lower. Losses are theoretically unlimited, so stops matter more, not less.

Crypto-specific

Digital assets trade continuously and carry mechanics equities do not have.

BTC dominance: Bitcoin's share of total crypto market value. Rising dominance in a falling market usually means capital is retreating from altcoins first.
Funding rate: A periodic payment between long and short holders of perpetual futures. Strongly positive funding shows crowded longs — a condition that precedes long squeezes.
Liquidation: Forced closing of a leveraged position when margin is exhausted. Clusters of liquidations create the fast candles that take out stops before reversing.
Stablecoin: A token designed to track a fiat currency, used as the quote asset in most crypto pairs.
24/7 session: Crypto never closes, so there is no opening auction to structure the day around. Volume still concentrates around the London and New York equity hours.

Signals and models

Language used when we describe what the AI layer did.

Model confidence: The reasoning layer's own stated certainty in a thesis, reduced when independent models disagree on the same evidence pack. It is a self-assessment, not a measured probability.
Consensus / disagreement: Two frontier models review the same data separately. Agreement raises confidence; material disagreement lowers it and often blocks the trade at the conviction gate.
Backtest: Applying rules to historical data. Useful for eliminating bad ideas, weak as proof — it excludes slippage and rewards rules that were tuned on the same history.
Overfitting: Rules that describe the past precisely and predict the future poorly. The usual symptom is a strategy with many parameters and a flawless historical curve.
Settled trade: A position closed by target, stop or timeframe expiry. Only settled trades count toward win rate; open positions are marked to market and shown separately.
Win rate: Settled winners divided by settled trades. On its own it says nothing useful — a 40% win rate at 3:1 beats an 80% win rate at 0.3:1.

Where these terms show up

The confluence score, conviction gate and settlement rules are described in full in our editorial policy and methodology. You can see them applied to live positions in the bot track record, and the sizing maths behind R-multiples is worked through in risk management and position sizing. Definitions here are educational; nothing on this page is investment advice — see the risk disclaimer.