NLI

Scoring Methodology

Every score on NLI is computed from evidence — not opinion. This page documents exactly how scores are calculated so you can evaluate them yourself. If you disagree with a score, you can trace it back to the specific data points that produced it.

Core Principles

Evidence, Not Opinion

Scores are computed from countable, verifiable data points — votes cast, endorsements received, court findings, financial filings. No editorial judgment enters the calculation.

Same Framework for Everyone

The identical formula is applied to every official regardless of party, incumbency, or our relationship with them. The math doesn't know whose side anyone is on.

Transparency Over Precision

We publish the exact methodology. A score you can verify is more valuable than a score you have to trust.

Proven vs. Alleged

A court judgment and an unproven allegation are both tracked, but they don't carry equal weight. Provability multipliers ensure proven findings count more than allegations.

Score Scale (0-100)

80-100
Strong — Substantial evidence of strength in this category
60-79
Above Average — More positive evidence than negative
40-59
Mixed — Significant evidence on both sides
20-39
Concerning — Notable negative evidence outweighs positive
0-19
Poor — Overwhelming negative evidence or critical proven findings

Provability Multipliers

Not all evidence is equal. A court judgment carries more weight than an unproven allegation. Every vulnerability vector has a provability rating that adjusts its impact on scores.

Proven1.0x

Court judgment, official government finding, documented legislative record, confirmed factual data. Full weight.

Documented0.8x

In public records or filings but not formally adjudicated. Credible evidence exists but no official ruling.

Alleged0.5x

Claimed in legal proceedings or credible reports but unproven. No charges filed, no court finding, no official determination.

Unsubstantiated0.25x

Single source, no corroboration. Tracked for completeness but minimal scoring impact.

Scoring Categories

Leadership

Ability to lead, manage, and attract support.

  • •Offices held (12 pts each)
  • •Endorsements received (weighted: President 25 pts, national PAC 15 pts, state org 5-10 pts)
  • •Political career depth (3 pts per role)
  • •Currently serving bonus (+15 pts)
Ethics & Integrity

Personal conduct, legal record, financial conflicts. Only legal, financial, and personal misconduct vectors count — policy disagreements do not affect this score.

  • •Starts at 85 (presumption of integrity)
  • •Critical proven findings: -18 pts × 1.0x
  • •High documented issues: -10 pts × 0.8x
  • •Alleged issues: severity × 0.5x multiplier
  • •Active ethics cases: -15 pts each
  • •Financial disclosures filed: +5 pts each (max +15)
Electability

Fundraising, endorsement strength, ad spending, minus political vulnerabilities.

  • •Fundraising ($1M+ = 25 pts, $500K+ = 18 pts, $100K+ = 10 pts)
  • •Endorsements (weighted same as Leadership)
  • •Ad spending (5 pts per campaign)
  • •Minus: policy_record/association/flip_flop attack vectors (provability-weighted)
Experience

Breadth and depth of political and professional career.

  • •Offices held (12 pts each)
  • •Career history entries (4 pts each)
  • •Distinct branches served (8 pts per branch)
  • •Distinct jurisdictions (10 pts per level: federal, state, local)
Policy & Substance

Depth of policy positions and legislative record.

  • •Issue positions documented (6 pts each)
  • •Legislative votes recorded (1 pt each)
  • •Bills sponsored (12 pts each)
  • •Public statements tracked (3 pts each)
Constituent Service

Responsiveness, promise-keeping, measurable outcomes.

  • •Promises kept (+15 pts each)
  • •Promises in progress (+8 pts each)
  • •Improving constituent metrics (+10 pts each)
  • •Minus: Promises broken (-12 pts each)
  • •Minus: Promises stalled (-5 pts each)
Fiscal Responsibility

Budget and fiscal policy positions.

  • •Starts at 50 (neutral baseline)
  • •Tax/fiscal/budget positions (+10 pts each)
  • •Legislative votes on fiscal matters (+1 pt each)
Transparency

FOIA compliance, financial disclosures, public accountability.

  • •Financial disclosures filed (12 pts each)
  • •Transparency grade (A = 30 pts, B = 20 pts, C = 10 pts)
  • •High-reliability sourced claims (3 pts each)

Endorsement Weight Tiers

25 ptsSitting or former President of the United States
15 ptsSitting or former U.S. Senator, Governor, or major national PAC with significant financial commitment
10 ptsLarge state legislative delegations, major statewide organizations (ALFA, BCA, etc.)
5 ptsState-level PACs, trade associations, local organizations (default)

Confidence Levels

Each score has a confidence level based on the total amount of evidence available. More evidence = higher confidence that the score is accurate.

Verified
15+ evidence items — Substantial evidence base. Score is reliable.
Likely
5-14 evidence items — Moderate evidence. Score is directionally accurate but may shift.
Unverified
<5 evidence items — Limited evidence. Score should be interpreted cautiously.

This methodology is version 3.0, last updated March 2026. The scoring engine is a Supabase SQL function that runs identically for every person in the database. Source code is available for audit. If you believe a score is incorrect, check the evidence on the person's profile — every data point that feeds into the score is visible and linked to its source.