Investment strategy glossary

What every term on this site means, from performance bases and risk measures to the rules strategies state and the types we sort them into. Tap any dotted-underlined term in the app for its short definition.

219 terms

Where text comes from

Manager text: The strategy’s own published description, written by whoever runs or publishes it (its manager, creator or platform).; For example IA methodology pages, Quantbase descriptions, a Composer creator’s own note or a Collective2 trader’s “about”. A rule quoted from manager text is the manager’s own claim, the strongest kind of stated evidence.
Platform AI summary: Text that Composer’s AI wrote about a creator’s strategy. It’s a paraphrase, so details can be wrong.; Composer’s “how it works”, “summary” and “value proposition”. Treat rules quoted from it with more care than manager text.
Form ADV brochure: The firm’s regulatory filing with the SEC (Form ADV Part 2A). Legally reviewed, but usually general.; Used for programs like Schwab Intelligent Portfolios, Robinhood and Webull: allocations, fees, minimums, tax-loss harvesting.
Analyst: A note written by our own staff. None exist yet.

Stated rules

Stated rule: Something the strategy’s own text says it does (a screen, exclusion, weighting, rebalance…), with the exact quote it came from.; Stated rules describe what a strategy says. Whether its holdings back that up is a separate check: see Evidence.
Proposed: A rule extracted from the text that hasn’t been approved yet. Proposed rules aren’t shown on this site.
Approved: A rule confirmed to be stated by its quote. Every rule shown on this site is approved.; Rules are extracted from each strategy’s own text, every quote is verified to appear word for word in it, and each rule is checked a second time before it is approved.
Rejected: A rule whose quote turned out not to state it. Rejected rules are never shown.
Provider not named: The text says it uses outside data (e.g. “third-party ESG data”) but doesn’t say whose. We never guess the provider.
Rule type: The kind of rule, such as an include screen, an exclusion, a weighting scheme or a rebalance schedule. The full list is below.

Performance basis

Basis: What a performance number is based on: real accounts, a composite, results since the last edit, or a simulation. Only compare numbers with the same basis.
Live: Results of a real account (the manager’s or a model account), as published by the source.
Composite: Combined results of the firm’s client accounts, as published by the source.
OOS (out-of-sample): Results only since the strategy’s logic was last edited, so they weren’t tuned with hindsight. Used for Composer.
Backtest: A simulation of how the rules would have done in the past. Hypothetical; never mix with live results.
Hypo (hypothetical): Results of a hypothetical model account, including fees and commissions, as published by the source (Collective2).
Model: The platform’s tracking of its model portfolio, not a specific client account.
Recon (reconstructed): Rebuilt by us from published holdings and prices. An estimate, labelled separately.

Evidence

Evidence: A check of whether what a strategy holds backs up what it says, using tests fixed before any results were seen.; Two tests so far: leveraged or inverse funds, and fossil-fuel companies (compared with IVV, the iShares S&P 500 ETF).
Supports: Its stated rules make a claim, and its holdings back it up.
Neutral: A stated rule exists, but the holdings neither clearly back it up nor contradict it (for example, out of that asset right now).
Contradicts: Its stated rules promise one thing and its holdings show the opposite.
Held, not stated: It holds (or may hold) these, but none of its stated rules mention it. Worth checking the source text.
Rules pending: This strategy’s rules haven’t been extracted yet, so there’s no claim to compare with. The measurement is still shown.
No claim: Nothing stated to test. The measurement is shown for reference only.
Low coverage: Fewer than 70% of the holdings could be identified, so there’s no verdict.
Stale: The holdings on file are more than 120 days old, so there’s no verdict.
No holdings: No holdings are on file for this strategy, so nothing can be tested.
Identified: The share of the portfolio we could classify for this test. Below 70%, no verdict is given.
Look-through: Seeing inside the funds a strategy holds, using each fund’s own SEC holdings filing (Form N-PORT).
Proxy: A stand-in fund that tracks the same index, used when the real one files no holdings (SPY is read through IVV, QQQ through QQQM).
May-hold list: Every fund a Composer strategy’s rules can switch into, even if it isn’t holding them right now.
Leveraged / inverse fund: A fund that aims for a multiple (2x, 3x) or the opposite of an index’s daily move, as its official name says.
Fossil-fuel company: A company whose SEC industry code is coal, oil or gas production, drilling, refining, pipelines or gas distribution. Electric utilities aren’t counted.
Oil & gas commodities: Exposure to oil and gas themselves (through commodity funds such as DBC, USO or BOIL), shown separately from fossil-fuel companies.; Weight in each commodity fund times its energy exposure: 2x for a 2x natural-gas fund, about 0.37 for DBC (its energy share in its 10-K). Negative for inverse funds. Never added to the fossil-company share.
Holds no companies: A fund that holds futures, physical commodities or currencies (gold, oil, the dollar, VIX), not company shares. It counts as identified, with 0% in fossil-fuel companies.
Estimated (est.): A verdict based on a mix estimated from the strategy’s returns, because it publishes no holdings. Weaker evidence than holdings.; Used only where holdings can’t give a verdict, only when the estimate fits well, and only while the method passes its acceptance check (see Estimated mix).

Holdings

Simulated holdings: Holdings from the platform’s backtest of the rules, not from a real client account.
Holdings as of: The date of the holdings on file. Older holdings may no longer reflect the portfolio.
Holdings type: Where the holdings on file come from: reported by the source, simulated by its backtest, or a target mix stated in its own documents.
Reported holdings: Positions the source itself publishes for its account or model portfolio (e.g. Quantbase, SoFi, Surmount).
Stated target mix: Fixed target weights quoted from the strategy’s own documents (e.g. “60% VTI, 20% VEU, 20% BND” in a brochure). A promise, not observed holdings, so it’s never tested as evidence.
Estimated mix: The blend of sector funds (plus small caps and cash) whose returns best match the strategy’s daily returns. An estimate for strategies that publish no holdings; it can’t name stocks.; Returns-based style analysis: weights of 0–100% that add up to 100%, fitted over the latest year of returns, with a 90% range from resampling. Shown only as a guide when the fit is poor.
Fit (R²): How much of the strategy’s day-to-day movement the estimated mix explains: 1 = all of it. Below 0.90 an estimate gives no verdict.

Metrics

Return: How much the strategy gained or lost over the period, as a percent.
Annualized: A multi-year return expressed as the equivalent yearly rate, so periods of different length can be compared.
Volatility: How much returns swing up and down, as a yearly percent. Higher means a bumpier ride.
Sharpe ratio: Return above the risk-free rate (T-bills) for each unit of volatility. Higher is better; above 1 is good.
Sortino ratio: Like the Sharpe ratio, but only counts downside swings as risk. Higher is better.
Max drawdown: The largest fall from a peak to a low during the period. Closer to 0% is better.
Beta: How much the strategy tends to move with its benchmark: 1 moves in step, below 1 moves less, above 1 moves more.
Return vs benchmark: The strategy’s return minus its benchmark’s over the same dates, in percentage points.
LH (Lighthorse standard): Computed by us with one fixed method for every source: trading days, annualized with √252, 3-month T-bill as risk-free.; Sources each use their own method (IA counts calendar days with √365), so a source’s number and ours can differ slightly. Compare strategies on the LH values.
Risk score: The source’s own risk rating. Scales differ by source (e.g. 1–5 for IA, 0–5 for Quantbase), so compare only within one source.

Data quality

Data completeness: The share of core fields we have for this strategy: description, inception, fee, minimum, asset classes, rebalance, holdings, performance, history, stated rules and evidence.; Fields that don’t apply (e.g. a fee for a Composer strategy) are left out of the score. Every missing field says why it’s missing.
Not published: The source doesn’t publish this for the strategy.
Login required: Only visible to the source’s signed-in users or subscribers. We never sign in to collect data.
Blocked: The source stopped our requests (e.g. “too many requests”). We stop and never work around it.
Not collected yet: The source publishes it, but we haven’t collected it yet.
Pending: Waiting for a pipeline step, such as rule extraction (on hold for Composer).
Doesn’t apply: Not meaningful for this strategy (e.g. a per-strategy fee on a subscription platform). Left out of the score.
Partly identified: There are holdings, but too little of them could be classified (or the estimate fits too poorly) to give a verdict.
Too old: There are holdings, but they’re over 120 days old, so no verdict.
Unexplained: Missing without a known reason: a gap in our pipeline that should be fixed.

Screener labels

Source type label: The category the source itself gives the strategy, word for word (e.g. “Socially Responsible Investing”).
Strategy types: Our own consistent categories across all sources (e.g. Copy trading, Leveraged ETFs). A strategy usually has several.
Themes / tags: Keywords the source attaches to the strategy, as published.
Coverage count: Under a column name: how many strategies in the current results have a value for it (e.g. 101/2,881).
ADV brochure program: The program in the firm’s SEC Form ADV brochure that describes this strategy. Its fee, minimum, text and rules are merged in here where the firm’s site gives none, so the program isn’t listed separately.; One program can cover several strategies (e.g. Interactive Advisors’ 13 socially responsible portfolios). Brochure programs with no match on the firm’s site (private funds, planning services) are still listed as their own rows.
Original name: The name the source uses. We show a clearer descriptive name; search finds both.
LOCAL MODE: This copy reads a database on one staff Mac, with sign-in off. The shared screener reads Supabase.

Rule types

The kinds of rule extracted from strategy text. Each stated rule has one.

Allocation target: Target mix across asset classes (e.g. 60% stocks / 40% bonds).
Copy / tracker source: Mirrors trades or holdings disclosed by others (Congress, 13F filers, insiders, a named investor).
Data provider / source: Who supplies the data or model (named, or 'unnamed' when the text only says e.g. 'third-party ESG data').
Event handling / exception: How corporate actions, index changes or exceptions are handled.
Exclude screen: Removes securities or industries (e.g. fossil fuels, tobacco, high polluters).
Hedge / defensive overlay: What the strategy moves to or adds for protection (volatility, Treasuries, cash, gold, inverse).
Include screen: Selects securities that meet a criterion (e.g. high LGBTQ+ inclusion scores, carbon leaders).
Leverage use: Uses leveraged or inverse instruments, or borrowing.
Number of holdings: How many positions the strategy holds (range or target).
Position cap: Maximum (or minimum) weight per position.
Rebalance schedule: Scheduled rebalancing (daily, weekly, monthly, quarterly, semi-annual, annual).
Rebalance threshold: Rebalancing triggered by drift beyond a band.
Risk control: Stop-losses, drawdown limits, volatility targets.
Score / metric threshold: A numeric cut-off on a score or metric (e.g. market cap > $50M, score >= 90).
Sector / asset cap: Maximum weight per sector, industry or asset class.
Tax management: Tax-loss harvesting, tax-aware construction, municipal bonds for taxable accounts.
Trading signal: A rule that switches holdings on an indicator (moving average, RSI, momentum rank, volatility, spread).
Universe / index: The pool securities are drawn from (an index, exchange, market-cap band, or list).
Weight by score: Position sizes scale with a score (higher score -> larger weight).
Weighting scheme: How positions are weighted: equal, market-cap, inverse-volatility, fixed target weights, risk parity.

Themes

What a rule is about, e.g. the screen in “Include screen: LGBTQ+ inclusion & equality”.

13F filings (hedge funds / investors): A theme rules can refer to.
Alternatives / managed futures: A theme rules can refer to.
Artificial intelligence: A theme rules can refer to.
Bonds / Treasuries: A theme rules can refer to.
Buybacks / net payout: A theme rules can refer to.
Carbon & emissions: Part of ESG (general).
Cash / T-bills: A theme rules can refer to.
Clean energy / EV: A theme rules can refer to.
Commodities: A theme rules can refer to.
Communication services: A theme rules can refer to.
Congress members' trades: A theme rules can refer to.
Consumer: A theme rules can refer to.
Controversial industries (tobacco, weapons, gambling…): Part of ESG (general).
Corporate governance & ethics: Part of ESG (general).
Corporate insider trades: A theme rules can refer to.
Corporate transparency: Part of ESG (general).
Credit spreads: A theme rules can refer to.
Crypto / blockchain: A theme rules can refer to.
Currencies / US dollar: A theme rules can refer to.
Customer welfare: Part of ESG (general).
Data security & privacy: Part of ESG (general).
Defense & geopolitics: A theme rules can refer to.
Dividends / yield: A theme rules can refer to.
Drawdown / loss from peak: A theme rules can refer to.
Energy: A theme rules can refer to.
Equities: A theme rules can refer to.
ESG (general): A theme rules can refer to.
Fair labor & communities: Part of ESG (general).
Financial strength / cash flow: A theme rules can refer to.
Financials: A theme rules can refer to.
Fossil fuels: Part of ESG (general).
Gender equality: Part of ESG (general).
Gold / precious metals: A theme rules can refer to.
Government contracts: A theme rules can refer to.
Growth: A theme rules can refer to.
Healthcare / biotech: A theme rules can refer to.
Industrials / defense: A theme rules can refer to.
Interest rates / yield curve: A theme rules can refer to.
International / emerging markets: A theme rules can refer to.
Inverse / short exposure: A theme rules can refer to.
Land & ecosystem health: Part of ESG (general).
LGBTQ+ inclusion & equality: Part of ESG (general).
Liquidity / trading volume: A theme rules can refer to.
Lobbying spending: A theme rules can refer to.
Low volatility: A theme rules can refer to.
Macro / economic regime: A theme rules can refer to.
Market share / competitive position: A theme rules can refer to.
Materials / commodities producers: A theme rules can refer to.
Momentum: A theme rules can refer to.
Ocean & marine life: Part of ESG (general).
Pollution control: Part of ESG (general).
Product safety & quality: Part of ESG (general).
Quality / profitability: A theme rules can refer to.
Quantum computing: A theme rules can refer to.
Racial & ethnic equality: Part of ESG (general).
Real estate: A theme rules can refer to.
Relative strength vs another asset: A theme rules can refer to.
RSI overbought / oversold: A theme rules can refer to.
Size / market cap: A theme rules can refer to.
Social / news sentiment: A theme rules can refer to.
Space / aerospace: A theme rules can refer to.
Sustainable products & lifecycle: Part of ESG (general).
Technology / semiconductors: A theme rules can refer to.
Trend (moving averages): A theme rules can refer to.
Utilities: A theme rules can refer to.
Valuation ratios: A theme rules can refer to.
Value: A theme rules can refer to.
Volatility products (VIX): A theme rules can refer to.
Volatility regime (VIX): A theme rules can refer to.
Water use & clean water: Part of ESG (general).
Workplace diversity: Part of ESG (general).

Strategy types

Our own categories, consistent across sources. A strategy usually has several.

13F / hedge fund trackerSignal source: Follows positions disclosed in 13F filings
Aggressive riskObjective: High-risk profile (high volatility, deep drawdowns, leverage)
AI / ML-drivenSignal source: Machine-learning or AI models drive selection
AI-personalized adviceStructure: Advice or portfolios generated per client by AI
Alternative dataSignal source: Signals from non-price data (social media, web traffic, lobbying, contracts)
AlternativesAsset focus: Hedge-fund-like or alternative strategies (managed futures, market neutral)
Buy and holdMethod: Long-term holdings with infrequent changes
Capital preservationObjective: Prioritizes limiting losses / cash-like returns
Commodities & goldAsset focus: Gold, commodities or real assets
ConcentratedMethod: Few holdings or a single dominant position
Conservative riskObjective: Low-risk profile
Copy tradingStructure: Mirrors another person's or entity's trades
CryptoAsset focus: Bitcoin, Ether or crypto assets
Direct indexingStructure: Owns an index's stocks directly, often with customization and tax-loss harvesting
ESG / SRIObjective: Screens or weights on environmental, social or governance criteria
Event-drivenMethod: Trades around events (earnings, corporate actions, filings)
Factor / smart betaMethod: Weights or screens on factors (value, quality, size, low volatility, dividends)
Fixed incomeAsset focus: Mainly bonds, Treasuries or credit
ForexAsset focus: Trades currencies
Fundamental researchSignal source: Company analysis (cash flow, valuation, competitive position)
FuturesAsset focus: Trades futures contracts
GrowthObjective: Seeks capital growth
Income / dividendsObjective: Seeks income or dividends
Index trackerStructure: Follows a published index's rules
Insider trackerSignal source: Follows corporate insider purchases
International equitiesAsset focus: Developed or emerging markets outside the US
Leveraged ETFsAsset focus: Uses 2x/3x or inverse ETFs
Long-shortMethod: Holds both long and short (or inverse) positions
Macro signalsSignal source: Economic, rate, credit-spread or yield-curve signals
Manager discretionaryMethod: A manager decides trades with judgment
Mean reversionMethod: Buys dips / sells rallies expecting a snap-back (e.g. RSI oversold)
Model portfolioStructure: A target allocation the adviser maintains for many clients
MomentumMethod: Buys recent winners / ranks by recent performance
Multi-assetAsset focus: Mixes stocks, bonds and other asset classes
OptionsAsset focus: Trades options contracts
Options incomeMethod: Sells options to collect premium (covered calls, puts)
Pairs / relative valueMethod: Trades the spread between related assets
Politician trackerSignal source: Follows disclosed trades of members of Congress or officials
Private fundStructure: Pooled private vehicle (e.g. master-feeder fund)
Real estateAsset focus: REITs or real-estate securities
Risk parity / all-weatherMethod: Balances risk across economic regimes or asset classes
Rules-based algorithmicMethod: Fixed, mechanical rules decide trades
Sector rotationMethod: Rotates among sectors or industries
Separately managed accountStructure: Client-owned account run by a manager (SMA)
Short-term tradingMethod: Day trading or holding periods of days
Small capAsset focus: Mainly small companies
Strategic asset allocationMethod: Fixed long-term mix across asset classes, periodically rebalanced
Tactical allocationMethod: Shifts between risk-on and defensive assets on signals
Tax-managedObjective: Designed for after-tax results (TLH, munis, tax-aware)
TechnologyAsset focus: Mainly tech / semiconductor / Nasdaq exposure
ThematicObjective: Invests in a theme (AI, clean energy, space, defense...)
Treasury ladderStructure: Ladder of Treasury securities by maturity
Trend followingMethod: Holds assets in uptrends (moving averages, trend filters)
US equitiesAsset focus: Mainly US stocks or US stock ETFs
User-builtStructure: Created by a platform user or community member
Volatility hedgeMethod: Uses VIX products or hedges to protect in stress
Volatility tradingMethod: Trades volatility itself as the main return source