How Face Value is worked out

Congress names bills for how they should be received. H.R. 1 of the 119th Congress is called the One Big Beautiful Bill Act; as enrolled it runs to 338 sections and amends thirteen separate titles of the United States Code, from Government Organization to Veterans' Benefits. No four-word name accounts for that, and there is no way to discover it without opening two thousand pages.

Face Value is that gap, counted: how much of a bill you get by taking its name at face value. It is not an opinion about a bill, and there is no model reading the text and forming a view — every input below is a count taken from the bill's own published XML, and you can reproduce the whole thing with the file open.

The formula

Every bill starts at 100. Two things subtract.

Areas of law — up to −50

8 points for every title of the United States Code the bill amends beyond the first, capped at 50. A bill that changes one area of law loses nothing.

Counted by matching the statutory phrase “title N, United States Code” in the bill text. How often each title appears is deliberately ignored: a bill that amends the tax code in ninety places is doing one thing thoroughly, while a bill touching nine titles once each is doing nine things. The second is what breadth means.

Length — up to −35

One point per eight sections beyond the first 20, capped at 35. Counted as <section> elements in the published text, so it is a count of elements rather than a reading of a document.

A name that admits it — both halved

When the short title contains omnibus, consolidated, continuing resolution, reconciliation or similar, both penalties are halved. An omnibus called an omnibus has not misled anyone.

Halved rather than cleared, because warning a reader that a bill is broad is not the same as the name describing what is in it. Without this rule the score would measure length and nothing else — the 2026 continuing resolution is almost exactly the size of H.R. 1.

What the numbers are called

The labels describe documents, not people. “Sprawling” is a property of a bill; there is no band that describes a sponsor.

Which version was counted

A bill's text is not fixed. It is printed on introduction, again when reported from committee, again when it passes a chamber, and again when enrolled — and it can triple in size along the way. That growth is exactly what this score exists to catch, so scoring the wrong draft would defeat the point.

The latest published version is always used, and every bill page names which one it read: as enrolled, as passed the House, as introduced. A bill scored on its introduced text may score differently once it has been through committee, and the date it was read is recorded.

What it does not claim

It is not a measure of honesty. It counts how far a bill reaches, not whether anyone was trying to mislead you. A low score means the contents go further than any short title could describe — which is sometimes deliberate, sometimes the ordinary result of legislating, and this score cannot tell the difference.

It cannot tell a descriptive name from a marketing one. The National Defense Authorization Act scores the same as bills with far vaguer titles. It is enormous and touches a dozen titles of the Code, and it is also named accurately for its main purpose. Whether a name is apt is a judgement, and judgements are not countable, so this score does not attempt one. This is the score's largest limitation and it is stated here rather than buried.

It says nothing about the sponsor. Introducing an omnibus is frequently a leadership duty rather than a choice, which is also why there is no leaderboard ranking members by the Face Value of bills they sponsored. Bills are ranked; people are not.

It is not a quality judgement. A narrow bill can be bad and a sprawling one can be necessary. The score measures reach.

Bill text comes from GovInfo's bulk data, the same source as everything else here, and every bill page links to the full text so the counts can be checked against it. The bills this ranks highest · Where the rest of the data comes from

If a count looks wrong to you, it may well be — tell us and we'll publish the correction, the same as any other number on this site.