1. What is a 13F
Form 13F is a quarterly position report filed with the U.S. Securities and Exchange Commission (SEC). It discloses the long U.S. equity holdings of institutional investment managers — hedge funds, mutual funds, pensions, family offices, and large registered investment advisers.
Each 13F lists the name, CUSIP, share count, and market value of every reportable U.S. equity position the manager held at quarter-end. The document is the single most transparent public window into how professional investors actually allocate capital.
2. Who must file a 13F
Any institutional investment manager that exercises investment discretion over $100 million or more in 13F-reportable U.S. equity securities must file Form 13F. Once an entity crosses that threshold, it is required to file every quarter for the remainder of that calendar year and the entire next year.
- Hedge funds and family offices that meet the threshold (e.g., Berkshire Hathaway, Pershing Square, Bridgewater, Tiger Global).
- Mutual fund families and ETF sponsors holding U.S. equities directly.
- Banks, insurance companies, and pension funds with discretionary U.S. equity portfolios.
- Registered investment advisers and trust companies above the threshold.
3. 13F disclosure timing
13F is filed within 45 calendar days of the end of each calendar quarter. In practice that produces four predictable filing windows per year:
- Q1 holdings (as-of March 31) — filed by mid-May.
- Q2 holdings (as-of June 30) — filed by mid-August.
- Q3 holdings (as-of September 30) — filed by mid-November.
- Q4 holdings (as-of December 31) — filed by mid-February of the next year.
- 13F.chat ingests filings from EDGAR within hours of publication and rebuilds aggregates each quarter.
4. Why backtests cannot use quarter-end prices
A naive 13F backtest treats quarter-end positions as if they were buyable at quarter-end prices. That is not what actually happened. The market did not see the holdings on March 31 — the public learned about them roughly 45 days later, when prices, sentiment, and even the portfolio composition itself may have already shifted.
Pricing a 13F at quarter-end therefore creates look-ahead bias: it grants the simulated investor information no real participant could have had. Any track record produced this way is, at best, optimistic — and often outright wrong.
5. Executable simulation: first trading day after disclosure
13F.chat takes the conservative, replicable approach. We assume the investor reads each 13F on the day it becomes public on EDGAR, then transacts at the open of the next regular U.S. trading day.
This rule keeps every simulated buy or sell strictly inside the information set a real-world reader would have had. It also models the realistic delay — sometimes a price gap up, sometimes a gap down — between disclosure and the moment a portfolio can actually be rebalanced.
- Entry / exit price = next regular session opening print after the EDGAR publication timestamp.
- Half-day or holiday sessions roll forward to the next full trading day.
- Cash, derivatives, short positions, and non-U.S. holdings are excluded — see Limitations below.
6. U.S. default benchmark: SPY
Every U.S.-listed fund on 13F.chat is benchmarked against SPY, the SPDR S&P 500 ETF Trust. We use SPY rather than the raw S&P 500 index because SPY is what an actual investor can buy: its returns include the management fee drag, dividend reinvestment, and execution friction a real index passive holder would experience.
Alpha, beta, and excess return on every fund detail page are computed against the same daily SPY total-return series.
7. Future Hong Kong benchmark: Hang Seng Index
13F coverage today is U.S.-only because the SEC mandate is U.S.-only. As 13F.chat expands to Hong Kong-listed funds and Asia-Pacific managers, the default benchmark will shift to the Hang Seng Index (HSI) total-return series for HK-domiciled portfolios. The same rules — disclosure-day open execution, daily mark-to-market, no look-ahead pricing — will apply.
8. Metric definitions
Every fund and every strategy page exposes the same eight core metrics. They are computed on the simulated post-disclosure return stream, never on raw 13F snapshots.
- Return
- Total cumulative return over the simulated holding period, including reinvested dividends and net of modeled execution at the next-open rule.
- Alpha
- Annualized return of the simulated portfolio minus the annualized return of SPY over the same window. Positive alpha means the disclosed positions, executed on the realistic timeline, outperformed a passive S&P 500 holder.
- Beta
- Slope of the portfolio's daily returns regressed on SPY's daily returns. Beta near 1.0 means the portfolio moves in lockstep with the market; beta below 1.0 means it dampens market moves; beta above 1.0 means it amplifies them.
- Sharpe Ratio
- Annualized excess return (over the risk-free rate) divided by annualized return volatility. Higher is better — Sharpe rewards smooth excess returns over jagged ones.
- Sortino Ratio
- Like Sharpe, but the denominator only counts downside volatility. Useful when a strategy has fat upside and you don't want to penalize big positive moves.
- Max Drawdown
- The largest peak-to-trough decline in the portfolio's cumulative equity curve over the period. Always reported as a negative number; lower (more negative) is worse.
- Turnover
- Average proportion of the portfolio that changes per quarter. A turnover of 0.20 means roughly 20% of holdings are rotated each filing window. High turnover signals an active manager; low turnover signals a buy-and-hold style.
- Concentration
- Share of the portfolio held in the top 5, top 10, or top 20 positions. A top-10 concentration above 60% indicates a high-conviction fund whose returns are driven by a small number of names.
9. Data limitations — what 13F cannot tell you
13F is a powerful but partial signal. Anyone treating it as a complete view of a manager's book will draw the wrong conclusions. The mandate has explicit blind spots:
- Long U.S. equities only. Short positions, options, futures, swaps, and other derivatives are not reportable on 13F.
- No international holdings. ADRs and U.S.-listed foreign stocks appear; positions on non-U.S. exchanges do not.
- No cash, bonds, or private credit. A 13F never tells you a manager's overall cash allocation or fixed-income book.
- Snapshot, not real-time. Quarter-end positions are weeks old by the time the public sees them.
- No execution costs of the manager. Our backtest assumes the next-open price; the real fund may have entered at very different prices days or weeks earlier.
- 13F omits portfolio leverage, margin, and prime-broker structure entirely.
10. Investment disclaimer
13F.chat is a research and education product. Nothing on this site — including fund pages, strategy backtests, AI-generated analyses, optimization outputs, and methodology notes — constitutes investment advice, a recommendation to buy or sell any security, or a solicitation of any kind.
Past performance, simulated or actual, does not guarantee future results. All strategy returns shown are hypothetical, frictionless except where explicitly modeled, and would have varied materially if executed in live markets. Consult a licensed financial professional before making investment decisions.