Observe
Record what happened, when it happened, and which safeguard it affected. Include the source and evidence pointing the other way.
How the clock works
See how the score is built.
The review asks whether the system can limit power, protect rights, settle disputes, and correct mistakes peacefully. It does not ask whether your preferred policy won.
The short version
How the clock is set.
Evaluate
Use the same scoring guide for all eight measures. Ask whether each safeguard works in real life, not only whether it exists in law.
Calculate
Apply each measure's weight, add the results, and convert the total to the clock. Show the reasonable range beside the result.
New evidence can move a score in either direction. Harmful actions and successful corrections both count. The clock is not a list of bad headlines.
“A party, candidate, belief, or policy does not move the clock by itself. The clock moves only when the evidence shows a change in how the constitutional system works.”
The framework
Eight measures, combined into one score
The weights add up to 100%. Version 0.7 uses the same weights for every review. Any future method change will be explained, and earlier results will remain available.
Limits on Government Power
Whether real limits can stop government power.
Checks and Balances
Whether Congress, the president, and the courts can check one another.
Elections and Transfers of Power
Whether elections are competitive and power changes hands peacefully.
Rule of Law
Whether the law is stable and applied fairly.
Individual Rights
Whether speech, belief, association, and due process are protected.
Federal and State Balance
Whether federal and state governments can use their lawful powers.
Trust in Institutions
Whether people accept lawful institutions as a way to settle disputes.
Peaceful Politics
Whether political competition remains overwhelmingly peaceful.
What does not move the clock
A policy result by itself
Tax increases aloneNo movement
Tax cuts aloneNo movement
Strict immigration policy aloneNo movement
Loose immigration policy aloneNo movement
What the review asks instead
Questions about how the system works
- Did the Constitution or a law authorize the action?
- Was Congress involved when the law required it?
- Were protected rights respected?
- Could a court review the action?
- Could voters reverse it?
- Could people oppose it lawfully?
- Were state and federal powers respected?
Quarterly reviews
Review on schedule. Move only when the evidence does.
Reviews take place on January 1, April 1, July 1, and October 1. Each one covers information available through the previous day and publishes the evidence considered, the decision for each measure, and the next review date—even when the clock does not move.
When a score changes
A score changes only when more than one reliable source shows a lasting or system-wide shift large enough to move it by five points. One headline, unpopular policy, court filing, or poll does not move the clock by itself.
When an early review may happen
An early review may happen if a national election or transfer of power is interrupted, court-ordered solutions stop working broadly, organized force replaces lawful resolution, or another event directly changes the country's ability to correct itself peacefully.
Current review: July 1–September 30, 2026. Next scheduled review: January 1, 2027.
Confidence in historical reviews
Strong facts do not always produce an exact score.
Every historical review shows its time period, main findings, evidence pointing the other way, sources, and two confidence ratings. Evidence confidence asks how strong and complete the historical record is. Score confidence asks how reliably that record can be turned into eight numbers.
Several direct or trusted sources agree on the central facts.
The record is strong, but the time period or comparison still affects the reading.
The evidence supports the direction of the score, but reasonable judgment remains.
The period is broad or the evidence is indirect, so the number deserves extra caution.
Important: a well-documented event does not make its 0–100 score a measured fact. The score still includes human judgment.
From score to clock time
A simple way to picture the result
A score of 100 appears as 6:00 PM. A score of 0 appears as midnight. Each point equals 3.6 minutes, and the result is rounded to the nearest five minutes. Midnight has one additional meaning: peaceful national self-correction has failed and organized force has replaced constitutional resolution. That rule is applied to 1861. The clock is a visual guide, not a prediction or countdown.
Research note · October 1, 2026 review · v0.7
How the latest score was reached
65/100 · About 8:05 PM
Effective October 1, 2026; evidence through September 30. AI tools helped organize this editorial review. Independent scoring review is still pending.
Effective October 1, 2026; evidence through September 30. AI tools helped organize this editorial review. Independent scoring review is still pending.
What the number means
Public evidence is compared with the same five guideposts for each measure. Scores between those guideposts use five-point steps and require judgment. A 55 instead of a 60 is not a directly measured statistical difference. The weights are part of the original design and have not yet been independently tested.
0The safeguard does not work in practice.
25It is badly weakened, and solutions rarely work.
50It works unevenly, with serious repeated failures.
75It generally works, but important weaknesses remain.
100It works reliably and includes everyone it should protect.
| Measure | Score | Weight | Points in total |
|---|---|---|---|
| Limits on Government Power | 55 | 15% | 8.25 |
| Checks and Balances | 60 | 12.5% | 7.50 |
| Elections and Transfers of Power | 80 | 15% | 12.00 |
| Rule of Law | 60 | 15% | 9.00 |
| Individual Rights | 70 | 15% | 10.50 |
| Federal and State Balance | 70 | 10% | 7.00 |
| Trust in Institutions | 50 | 7.5% | 3.75 |
| Peaceful Politics | 65 | 10% | 6.50 |
Calculation: each score is multiplied by its weight, then the results are added together. The total is 64.50, displayed as 65/100. No outside index is simply copied into this number.
How to read the labels
Version 0.7 uses these ranges: 82–100 resilient; 68–81 guarded stress; 52–67 elevated stress; 35–51 high stress; and 0–34 self-correction at risk. These are descriptive categories, not medical-style diagnoses. The reasonable score range may cross more than one category.
Why show a range?
To show how judgment affects the result, every measure is also calculated 10 points lower and 10 points higher. That produces a range of 55–75/100, or about 7:30 PM–8:40 PM. This is an illustration, not a statistical confidence interval. Different weights or readings of the evidence could produce a result outside it.
What supports the score?
Each section below separates what a source reports from how the review interprets it. The evidence is stronger than the apparent precision of the numbers. The October review incorporates an August Supreme Court order and July–September survey findings, but some measures still have less coverage than others. A court remedy counts as a real check; the review does not assume that every ruling will be followed. Low public trust also does not, by itself, make an election illegitimate.
Limits on Government Power · 55/100
55: significant constitutional disputes remain, but judicial constraints remain available. No corroborated five-point change was established.
What could change this score: Raise the score if government repeatedly follows legal limits; lower it if officials keep avoiding or defying binding limits.
Checks and Balances · 60/100
60: both meaningful judicial checks and substantial oversight concerns persist; neither a durable restoration nor systemic breakdown met the movement threshold.
What could change this score: Raise the score if checks are used and followed across all three branches; lower it if a branch cannot enforce its lawful authority.
Elections and Transfers of Power · 80/100
80: competitive electoral processes remained in place by the cutoff, without evidence of a nationwide blocked certification or transfer. Future election outcomes are not assumed.
What could change this score: Raise the score if access and election administration remain reliably fair; lower it if competition, certification, or a transfer of power is blocked.
Rule of Law · 60/100
60: substantial concerns continue, but no sufficiently corroborated new five-point deterioration or recovery was demonstrated in July–September.
What could change this score: Raise the score if legal remedies are applied promptly and fairly; lower it if officials repeatedly ignore rulings, enforce laws selectively, or punish people for seeking review.
Individual Rights · 70/100
70: legal protections remain substantial, with uneven application and continuing government pressure; this review does not equate criticism of a policy with rights violations.
What could change this score: Raise the score if dissent and due process are protected consistently; lower it if retaliation becomes systematic or meaningful review is lost.
Federal and State Balance · 70/100
70: states retained meaningful remedies and authority, despite significant disputes. The evidence does not establish durable nationwide displacement of state power.
What could change this score: Raise the score if disputes over authority are resolved more consistently; lower it if lawful state powers are repeatedly displaced without an effective remedy.
Trust in Institutions · 50/100
50: acceptance of the democratic ideal persists alongside growing dissatisfaction with its performance. Dissatisfaction is not itself rejection of lawful results.
What could change this score: Use broader evidence about acceptance of results, compliance with court rulings, and trust across parties before narrowing this range.
Peaceful Politics · 65/100
65: encouraging evidence does not yet establish a durable five-point improvement in actual peaceful political competition. Concern remains warranted.
What could change this score: Raise the score if political intimidation declines over time; lower it if organized, lasting violence prevents lawful competition.
Limits and next review
This quarterly review cannot measure every government action or institution. Federal-state balance, peaceful politics, and trust in institutions need more direct records and broader public data. Some measures overlap, so one problem can affect more than one score. Strength in one area can also offset a serious failure elsewhere in the total. For those reasons, the number alone cannot prove that peaceful self-correction remains available.
Before the index can be called independently validated, at least two independent reviewers should score the same evidence, publish where they disagree, and test other weights and overlaps. Earlier versions should remain available when an input changes. The next scheduled review is January 1, 2027; news does not update the clock automatically.
About the sources
The organizations below have not endorsed this method or score. Freedom House’s 2026 report covers 2025, so it is background context rather than a new quarterly observation. The review also considers the August 2026 Supreme Court order, July Bright Line Watch survey, Kettering/Gallup research published in September, and September UC Davis preprint. Research findings about attitudes are not direct measures of institutional behavior. The separately labeled November election update falls outside the September 30 evidence cutoff.
Freedom in the World 2026: United States (covers 2025)Freedom House · Annual assessmentLearning Resources, Inc. v. Trump · February 20, 2026 opinionSupreme Court of the United States · Primary recordThe Persistence of Diminished Democracy in a Second Trump PresidencyBright Line Watch · December 2025–March 2026 surveysLegal experts say rule of law has hit its lowest point in a decadeUCLA Law / Bright Line Watch · May 20, 2026Survey shows little shift in Americans’ views on political violenceUC Davis Centers for Violence Prevention · June 1, 2026Public Trust in Government: 1958–2025Pew Research Center · December 4, 2025A Partial Democratic Rebound? · July 2026 surveysBright Line Watch · August 2026Trump v. California · August 24, 2026 orderSupreme Court · Primary recordFrustration Rises, Commitment EnduresKettering Foundation / Gallup · September 23, 2026Slim Majority Doubt U.S. Leaders Are Held Accountable to LawsGallup · September 22, 2026Election results, canvass, and certificationU.S. Election Assistance Commission · July 2026 guidanceNovember 2026 midterms: vote counting and congressional controlAssociated Press · November 4–5, 2026Election Day unfolded with relatively few problemsAssociated Press · November 20262026 Midterms election integrity litigation trackerJust Security · November 2026 · Court proceedings indexDemocracy before the 2026 midterms · October surveysBright Line Watch · October 2026Decreased Support for Political Violence in Mid-2026UC Davis · September 8, 2026 · Preprint