Your money has three choices. Only one shows its work.
Maine Street runs institutional-grade screens over SEC filings, doing the same normalized earnings, owner-cash-flow and balance-sheet work a fund analyst does by hand. Every ranking shows the formula, the inputs, and the filing it came from.
The same windows, three ways.
Plays entered 2024-09-11 to 2026-09-10, each measured over its own holding window, 8 months on average. Cost figures are industry averages.| Index fundS&P 500, buy and hold | Financial advisorManaged account, 1% of assets | Maine StreetYou, with the screens | |
|---|---|---|---|
| Result over the same windows | +13.6% | +12.5% | +30.1% average Conviction List play over 19 plays, ahead of the S&P by +16.5% over the same windows. |
| Annual cost | About 0.03% in fund expenses. | About 1.37% in advisory and fund fees. | Free plan at $0. Pro at $99 a month. |
| What you can inspect | The index. | A quarterly statement. | Every formula, every input, and the filing each number came from. |
| All three columns are measured over the same windows: each Conviction List play's own holding period, from its entry to its exit or to the latest close. The index column is SPY's stored closing prices over those windows, averaged, price only. The advisor column is the same return net of a 1.37% average annual all-in fee charged for the average holding length. The Maine Street figure is the average return of every Conviction List play entered in the last two years. For reference, buying SPY on 2024-09-11 and holding through 2026-09-10 returned +35.5%. This is a research record, not a managed portfolio, and past results never predict future ones. | |||
Four ways of finding the same thing: a good business the market has mispriced.
Each screen is a different, independent argument. A company that surfaces on one is interesting. A company that surfaces on several is the reason the Conviction List exists.
107 open now.
54 open now.
15 open now.
73 open now.
8 open now.
Every screen keeps its own ledger, in the open.
The night a company appears on a screen, a play opens at that day's close and follows written exit rules. Entries, targets and exits are all published, including the plays that are losing.
Plays are stamped before the move
Every play is written to the database the night its screen refreshes, with the entry price and the score that triggered it. Nothing is added after the fact.
Point-in-time data only
Screens run on what was public on the filing date. A restated figure never travels backwards into an old score.
The losers stay on the page
A losing play stays open on the ledger until its rules close it. Nothing is quietly deleted.
Normalized P/E (10)
<10 ideal, >25 earns nothing; negative normalized earnings score 0.
Scored for YELP INC in the 2026-09-11 run, on TTM figures as filed through 2026-06-30.
No black box. Click any number and the math opens up.
Most screeners hand you a rating and ask for trust. Every score here decomposes into five categories, each category into sub-factors, each sub-factor into its formula, its inputs, and the filing those inputs came from.
Reported facts and interpretation are labelled separately
Anything pulled straight from a filing is marked reported. Anything computed or written by a model is marked as such, every time.
Originals are preserved, not summarized
Full filings are archived on ingest. A summary never replaces a source document, so a score can always be walked back to the page it came from.
Two plans. The difference is the screens that are working hardest.
Free
Forever- Value Board and Relative Value screens, ranked in full
- Company research pages with the score breakdown, formulas and inputs
- Five-year financials, insider Form 4 activity, and source filings
- The full published play ledger for every screen, locked or not
Pro
All screens- Everything in Free, plus:
- Josh's Formula and the Conviction List, the two strongest ledgers
- Data Grabbing exports a company's KPI history from its own filings to CSV or Excel
- AI Screener filters the whole database from a plain-English sentence