Every quarter, institutional investment managers with at least $100 million in 13(f) securities disclose their U.S.-listed equity holdings to the SEC. That makes the 13F the closest thing retail investors have to a public window into what large funds actually own. It is also, in practice, the most misread dataset in retail finance: people treat a delayed, partial snapshot as a live portfolio and trade on headlines about it. Reading 13Fs well is mostly about knowing what the form does not say.

What a 13F actually shows, and what it hides

A 13F lists the long positions in 13(f) securities that a manager exercised investment discretion over at the end of a quarter. That scope excludes a lot. Short positions do not appear, so a stock listed in the filing could be one leg of a hedge you cannot see. Most non-U.S. holdings are omitted. Cash, bonds, and most derivatives are outside the picture too. A 13F tells you what a manager held, in one sliver of their book, on one day. Treat it as evidence, not as the portfolio.

Read quarter-over-quarter deltas, not snapshots

A single filing is a photograph. The information is in the film strip: what changed since last quarter. A holding that appears for the first time, a position that doubled, a longtime core holding that shrank by half, a full exit. Comparing two consecutive filings for the same manager turns a static list into a statement about behavior. Most 13F headlines skip this work and report the biggest positions instead, which are usually the oldest and least newsworthy lines in the filing.

The 45-day lag problem in practice

Managers can file up to 45 days after quarter end, and many file close to the deadline. That means a position disclosed in mid-February reflects holdings as of December 31, and the manager has had a month and a half to change their mind. The lag does not make 13Fs useless. It makes them useful for different questions: not “what is this fund buying today” but “how has this fund’s thesis evolved over quarters,” which the lag barely affects. If you catch yourself reacting to a 13F as breaking news, stop and check the report date.

Changes worth flagging

Not every delta deserves attention. The ones that tend to carry information:

  • New positions, especially ones large relative to the fund, because they represent a fresh, deliberate decision.
  • Full exits of positions the manager held for multiple quarters.
  • Concentration changes: a position growing from a rounding error to a top-ten holding says more than a large position drifting with the market.
  • The same name appearing across several unrelated managers in the same quarter.

Position sizes that merely rose or fell with the stock price are usually noise; back out price moves before crediting a manager with conviction.

Why copying a 13F is not a strategy

The obvious temptation is to clone a famous fund’s filing. The mechanics work against you: you buy 45 or more days late, you cannot see the shorts or options that may hedge the position, you do not know the manager’s cost basis or exit plan, and you will learn about their exit a quarter and a half after it happened. None of that means the data is worthless. It means the value is context, not copy-trading. A 13F delta tells you a serious investor changed their mind about a company, which is a reason to look, not a reason to buy. None of this is a recommendation to buy or sell anything.

How QuantConomy turns 13F changes into signals

We read 13F filings as they post to EDGAR, link each holding to the asset, and compare against the manager’s prior quarter, so the deltas are computed rather than eyeballed. A notable position change can raise an INSTITUTIONAL signal with a direction, a strength score, and the filing attached as the source, and the same holdings are queryable through the /sec/institutional-holdings endpoint on our SEC filings API. It is the same approach we take with insider buying: do the tedious comparison work automatically, keep the primary source one click away, and let you judge it.

A quarter-over-quarter delta with the filing attached beats a headline about a famous fund every time. The fund’s name is the least informative field in the document.