The most common 13F analysis mistake: comparing raw dollar values between quarters and calling it a "buy" or "sell" signal.
Here's why that doesn't work and what to do instead.
The False Signal Problem
A position goes from $100M to $120M quarter-over-quarter. Did the manager add shares? Maybe. Or maybe:
- The stock price went up 20% and they did nothing
- They actually sold 5% of their position but price appreciation masked it
- A stock split or corporate action changed the share count
Dollar changes ≠ manager decisions.
The Right Comparison Framework
1. Compare Share Counts, Not Dollars
Share count changes isolate actual buying and selling from market moves. If shares held went from 1M to 1.2M, that's a real add.
2. Compare Portfolio Weights
Did the position go from 3% to 4% of the portfolio? Or from 3% to 2.8%? Weight changes tell you about relative conviction, not absolute market moves.
3. Context Matters
A manager adding 10,000 shares of AAPL to a $50B portfolio is noise. The same add in a $500M portfolio is a signal.
4. Watch the Exits and Entries
Complete exits and brand-new positions are the clearest signals. No price appreciation or rebalancing ambiguity — they made a binary decision.
Tools to Do This Right
13F Insight shows quarter-over-quarter changes with both dollar and share-count deltas, so you can separate real decisions from market noise.
What's your approach to comparing quarterly filings? Share counts, weights, or something else entirely?
Top comments (0)