13F Backtest

Background: how 13F backtests work

Every quarter, U.S. institutional investment managers with at least $100 million in equity AUM file Form 13F-HR with the SEC. The filing lists the firm's long positions in U.S.-listed stocks as of quarter-end, including ADRs, listed options, and registered closed-end funds. Filings are due 45 days after the period closes, so a portfolio held on December 31 doesn't surface until mid-February.

This page reconstructs a filer's holdings at the period-end, scales the share counts to your target portfolio value, and prices the basket forward to today. The result is a per-symbol P&L showing what the same dollar allocation would have done from the filing date to now — useful for evaluating a manager's conviction, generating ideas, and stress-testing strategies against real holdings rather than synthetic factor models.

How to use the tool

The five-step flow above moves left to right. Search for a fund by name (the local index covers the past two quarters and falls back to live SEC EDGAR for older filers), pick a quarterly filing, set a budget, preview the allocation, then run the backtest.

A few non-obvious knobs:

• Target portfolio value. Default $100,000 scales most large-cap baskets cleanly. Concentrated funds (Pershing Square, Scion) work fine at this size. Highly diversified books (Renaissance, D.E. Shaw) often have low-weight tail positions that cannot be expressed at small budgets — the tool flags these as Underfunded and suggests a higher target that would capture them.
• Mirror options as cash. Off by default. The tool renormalizes stock weights to deploy your full budget when this is unchecked. Turn it on if you want a strict reproduction of the filer's stock-vs-option ratio (the cash that would have funded the missing options remains uninvested).
• Treat long calls as long shares. A 13F lists call options as long positions. Treating them as the underlying captures the directional exposure without modeling option Greeks. Long puts are intentionally skipped here because converting them to long shares would invert the trade direction.
• Backtest from date. Defaults to the filing's period-end. Use a later date (the filing date, or even later) if you want to test the portfolio under "what a follower could actually have bought" timing.

Notable filers worth exploring

A short list of widely-watched managers, biased toward portfolios that are interesting to backtest:

• Berkshire Hathaway — Buffett's book; concentrated in financials and consumer staples; turnover is unusually low for an institution this large.
• Pershing Square Capital — Bill Ackman; under 15 names most quarters; high conviction and unusually durable holding periods.
• Scion Asset Management — Michael Burry; small portfolio with aggressive rotations between concentrated themes (semiconductors, China, regional banks).
• Tiger Global Management — Public-equity sleeve of Chase Coleman's tech-heavy growth book; the 2022 drawdown is visible in any pre-2023 backtest.
• Greenlight Capital — David Einhorn; long-short in spirit but the 13F shows only the long side, so backtests look more directional than the live fund.
• Appaloosa Management — David Tepper; high turnover, often heavy in tech and select industrials.
• Renaissance Technologies, D.E. Shaw, Citadel Advisors — quant platforms; thousands of small statistical positions. The "average" backtest tracks broad equity beta. Fun to look at, less useful for replication.

The local advisor index syncs weekly from EDGAR's quarterly company index, so newly filed managers appear within a few days of their first 13F submission. Older or one-off filers will fall through to a live SEC EDGAR search.

Reading the backtest results

Three caveats worth keeping in mind:

The 45-day lag means a 13F published February 14 reflects positions held December 31. Six weeks of price action and trading happen between those dates. If you're evaluating "what a follower could have done," start the backtest from the filing date, not the period-end — anything earlier is information the public did not have.

This tool buys at the chosen start price and holds. The filer was almost certainly trading inside the quarter, taking partial profits and adding to positions. Their realized return diverges from this static replication, sometimes meaningfully.

High-turnover books (quant funds, event-driven shops, short-dated-strategy managers) churn most of their portfolio quarter to quarter. A 13F snapshot represents a single moment that no longer exists a week later. These managers' filings are still useful for due diligence and style attribution, but the backtest P&L should be read as a study aid, not a prediction.

What 13Fs do not include

A 13F is a long-equity disclosure, not a complete portfolio. Missing from every filing:

• Short positions, in any form. A market-neutral hedge fund will look like a long-only equity book.
• Cash and fixed income. A 30 percent cash buffer is invisible.
• Foreign equities not registered with the SEC. ADRs are included; primary listings on the LSE, TSE, HKEX, and similar exchanges are not.
• Currencies, commodities, futures, swaps, and most derivative structures.
• Positions held in private vehicles or sub-advised separately managed accounts.

This is why a backtest of a long-short fund's 13F can look wildly off versus the firm's own performance reports — half the book is hidden.

Suggested use cases

• Manager due diligence. Run a fund across its last four quarters. Persistence in top names suggests conviction; complete rotation suggests trading-style turnover.
• Idea generation. Identify positions held in size by multiple high-quality managers — overlap is a common starting point for further research.
• Style attribution. Replicate the same firm's filing across regimes (2022 drawdown vs 2023 rally, pre- and post-Fed pivot) to see how their long book held up.
• Educational replication. Mirror a famous bet at its inflection point — Burry's Q1 2020, Ackman's pre-COVID hedge, the Big Short setup — to study a known thesis with real prices.
• Benchmarking your own ideas. Compare your watchlist against an institutional book to see whether your conviction overlaps with smart money.

SEC 13F filing schedule

For reference, the deadlines and typical filing-cluster windows:

Funds frequently file in the final week before the deadline. Same-quarter peer comparison is most reliable a few days after the deadline date, once the laggards have submitted.

Frequently asked questions

How accurate is a 13F backtest?

Accurate as a snapshot, lossy as a strategy. The holdings, share counts, and reported values come straight from the SEC filing — those numbers are exact. The backtest then prices each position from a chosen start date forward, which assumes you bought at that day's close and held without rebalancing. Any intra-quarter trading the manager did is invisible, and short positions, options writes, fixed income, and cash are not part of the disclosure at all. Read the P&L as "what this static long-equity sleeve would have returned," not "what the fund returned."

Why do some holdings show as Underfunded?

The filer's weights are converted to share counts against your target budget. If a position represents (say) 0.05 percent of a $50 billion fund, expressing that weight in a $100,000 portfolio requires fewer than one share. The tool flags those rows and offers a bump suggestion — the smallest target value at which the holding becomes representable.

Can I backtest dates before 2010?

Quarterly indexes go back to 1993, but our local price history covers more recent windows. Backtests rely on those local prices, so very old start dates may return partial or empty results.

Is this investment advice?

No. The page is informational. Past returns — including those of sophisticated institutional investors — do not predict future performance. 13F data is delayed by 45 days; positions you see in the filing may have been sold before you read them. Trading involves substantial risk of loss. Read the Terms of Service §17 for the full disclaimer.