Tracking Insider Grants

Using Cursor To Create an Options Tracker

Throughout my investing career, I have been unusually obsessed with evaluating management teams. Obviously, the financials are important. That said, businesses are helmed by people, and the quality of those people can determine whether a stock outperforms or not.

At the same time, much of the analysis about management teams is qualitative, not quantitative. Much of it is based on feel, whether it's interviewing management teams at a conference or determining their skills as capital allocators.

One of the best ways to determine management quality (at least as it relates to shareholder return) is to monitor open purchases and sales. It can be messy at times, but it is perhaps one of the cleanest looks at a manager's conviction (or lack of conviction) in the company.

With that said, there are other signals that we can look at. Specifically, I like to look at the timing of option grants. If you look through history, you can identify patterns of companies granting options days before an earnings announcement with good news. On the flip side, you also see companies granting options days after a bad earnings report.

This practice, called spring loading and bullet dodging, is heavily scrutinized by the SEC. However, companies still do this. Because they still do this (and all of this information is in public SEC filings), we can try to gather some signal into companies' near-term prospects.

It's why I created a very simple options tracker. It's really only for my internal use, but I wanted to share it in case you were interested in checking it out. You can find it here. It's password-protected, but I'm happy to provide it if you're interested.

Essentially, the tracker takes a watchlist and analyzes the timing of historical option grants. From there, it builds a profile of the historic timing behind those grants. Then, when new Form 4s are filed, the app evaluates the grant, compares it to the historic timing, and flags grants that appear to be potential spring loading or bullet dodging.

It's not the most sophisticated or complicated app. At the same time, it speeds up my research process and flags ideas that I would potentially miss. The stack is also simple. It's a simple NextJS app with a Supabase backend. I deployed it with Vercel. Because it's just for me, there is virtually no auth and the UI is bare bones.

It's yet another example of leveraging the power of AI to execute latent ideas that many of us have. It's not going to be a billion dollar startup, but it makes my life better. Sometimes, that's good enough!

Until next week,

Adam