Beta. An independent, unofficial civic tech project for education. Ratings are this tool's own judgments and are still being verified. Confirm your ballot with your county elections office. About and disclaimer
Ballot Weighing Tool
A guide for working out your own yes or no on all 14 statewide propositions. You say how you feel about each measure's goal. The tool pairs that with evidence ratings from its Standard Evaluation Framework and shows where you lean. It never tells you how to vote.
For each measure you see two separate things. Your preference is whether you support its goal. The evidence score shows how well the measure is built on the factors you care about. The tool shows how the two line up, but it never decides your vote.
Our Standard Evaluation Framework
To make it easier for voters to compare California's ballot propositions, this tool uses one consistent evaluation framework across all 14 propositions.
The framework considers factors such as public benefit, strength of evidence, safety, accountability, feasibility, and fiscal impact. Each factor is assessed using available evidence and supporting sources, with ratings and explanations provided for voters to examine.
The framework also uses default weights so voters can explore the propositions without having to configure a scoring model themselves. The standard weights are already set, so you can go straight to Step 2.
Factors and standard weights
Each measure's evidence score is the weighted average of its ratings on these factors, from 1 (weak) to 5 (strong). A factor with weight 5 counts two and a half times as much as one with weight 2. With the standard weights, Net benefit and Evidence strength together make up 10 of the 24 weight points.
How scoring works
- Choose what matters to you. Set how important each of the seven factors is. Your weights become percentages of your total weighting. With the standard weights, Net benefit is .
- Review the evidence ratings. Each proposition has a rating from 1 (weak) to 5 (strong) on each factor, with a written reason and sources.
- Calculate the weighted result. Each factor adds points in proportion to its percentage and its rating. A factor rated 5 of 5 adds its full percentage; one rated 3 of 5 adds three fifths of it.
- Understand what it means. The result summarizes this tool's recorded ratings under your priorities. It is not a chance of success, a share of voters, a guarantee of benefit, or a recommendation to vote Yes or No.
Show calculation
Weight share of a factor = its weight ÷ the total of all seven weights × 100.
Points from a factor = its weight ÷ the total weight of the factors that are weighted and rated × its rating ÷ 5 × 100.
Weighted evidence rating = the sum of those points. The tool rounds the average rating to one decimal (for example 3.19 becomes 3.2) and uses that rounded value for the score, the quality word and your lean, so the percentage is shown from it (3.2 of 5 = 64%). Displayed numbers may not add up exactly because of rounding.
A factor set to Ignore (weight 0) is left out. A factor that is Not rated is left out too, and the other weights are scaled up to fill its place. Missing evidence is never counted as 0 or as 5. If every weight is 0, no evidence score is calculated.
| Proposition | Weighted evidence rating | Average of 5 |
|---|
How were the weights determined?
The criteria and weights reflect the framework's design choices about what matters when evaluating a ballot proposition. They provide a consistent starting point; they are not a scientifically proven or universally agreed-upon set of ideal weights. The choice of weights can influence the results, and reasonable people may disagree about the relative importance of different factors.
The purpose is to make the evaluation process more structured and transparent, not to present a numerical score as an objective verdict or a recommendation on how to vote. Voters are encouraged to review the evidence, examine the assumptions, and make their own decisions.
What the framework cannot establish. It cannot predict whether a measure will succeed, and it cannot weigh your values for you. Ratings are analytical judgments based on the cited sources, and where information is thin a rating is marked as limited or Not rated rather than scored low.
Customize the weights (optional)
How do you feel about each measure's goal?
Each card gives the official title, what the measure does, and what a YES or NO vote legally means. Open More facts for the money, accountability, the official arguments on each side, ratings and numbered sources. Not sure records that you are undecided. Leaving a card blank skips it. Neither counts for or against.
Your ballot at a glance
This tool helps you compare your priorities with available evidence. Its results depend on your answers, selected weights, and the quality of the available information. It does not tell you how to vote.
Tap any row to see that measure's details and sources.
How your result is worked out
- Your preference is how you feel about the measure's goal. Against or Strongly against always shows Leans toward NO. That reflects your view, not a finding that the measure is bad. Not sure and Not answered never count as support or opposition.
- The evidence score is a weighted average of this tool's ratings, counting each factor by its weight (the standard weights unless you customize them). Factors set to Ignore, and any rating marked Not rated, are left out. It runs from 1 (weak) to 5 (strong), shown to one decimal place. If every factor is set to Ignore, it shows Evidence not weighted. If you weigh factors but none of them has a rating for a measure, it shows No evidence score available. Either way the result follows your preference alone.
- When you support the goal, the evidence score decides whether the evidence backs you up, using the table below. Either way, the vote is yours.
| Your preference | Evidence score | Result |
|---|---|---|
| Strongly support | 2.0 or higher | Leans toward YES |
| Strongly support | Below 2.0 | Mixed / uncertain |
| Support | 3.0 or higher | Leans toward YES |
| Support | Below 3.0 | Mixed / uncertain |
| Not sure | Any | Not sure |
| Oppose or Strongly oppose | Any | Leans toward NO |
| Skipped | Any | Not answered |
About the cutoffs. The 2.0 and 3.0 lines are illustrative rules of thumb. They are not probabilities that a measure will work, and they have not been statistically validated. Mixed / uncertain means you like the goal but the measure rates below the line on what you said matters. Some people vote yes anyway because the goal matters most. Others vote no and wait for a better-written version.
Sources and methodology
Facts. Each card's facts come from the Secretary of State's official voter guide (titles, what YES and NO mean, official arguments and legal text) and the Legislative Analyst's Office (fiscal estimates). Independent reporting and research are used where official sources are silent, and are labeled as Independent. Campaign claims are attributed to the side that makes them. Every card lists its sources, numbered so each rating can point to them.
Ratings. The ratings are structured analytical judgments made for this tool, drafted with AI research assistance and published by the tool's creator. They are not official government ratings or scientific measurements. One rubric is applied to all 14 measures, whatever the topic. Dashed dots mean a rating rests on limited information, and Not rated means there was too little verified information to score it. Missing information is not treated as a good or bad sign.
See the rating rubric
| Factor | 1 | 3 | 5 |
|---|---|---|---|
| Net benefit | Likely costs or harms outweigh benefits | Real benefits and real costs, net unclear | Large, likely benefits well above costs |
| Evidence strength | No credible evidence this approach works | Mixed or indirect evidence, such as similar policies elsewhere | Strong, direct evidence for this approach |
| Safety | Serious documented risk of harmful side effects or legal failure | Real but limited or familiar risks | Few plausible downsides |
| Accountability | Nothing beyond ordinary lawsuits | Required public reporting or a built-in enforcement remedy, but no independent audit | Independent audits, public reporting and penalties all required |
| Feasibility | New, untested systems with many undefined terms | Existing agencies, but tight timelines or open questions | Existing systems with little new work |
| Fiscal clarity | No official estimate and unclear direction | Official dollar range or bound, key costs unknown | Clear official yearly and total figures |
| Outcome clarity | No measurable result defined | Spending or rules are measurable, but real-world results are not tracked | Specific numeric result written into law and reported |
A 2 or 4 falls between the descriptions. For Accountability, 2 means a self-executing rule or limited reporting with no review, and 4 means independent audits plus public reporting. For Fiscal clarity, 2 means the official analysis gives a direction but no dollars. For Outcome clarity, 4 means a clear observable rule or required reporting on results.
Last checked. Facts and source links were first reviewed October 15 and 16, 2026. On October 22, 2026, every fact on this page was audited again by automated research against the Secretary of State's voter guide, the LAO analyses, the official arguments and, where it could be read, the legal text. 93 statements were corrected or labeled as reported by news sources. The ratings themselves were not changed. On October 23, 2026 an independent review led to three more wording fixes (Props 3, 39 and 44), again without changing any rating. On October 24, 2026 four flagged ratings were resolved against the legal text and the LAO analysis: Prop 44's accountability went from 5 to 4 because audits are allowed but not required, and the evidence strength of Props 37 and 38 and Prop 37's accountability stayed at 3 with clearer reasons. On October 24, 2026 a cross-check against the Official Voter Information Guide led to 19 more corrections of wording, attribution and source links, again without changing any rating. No person has yet reviewed the audit. Official materials can change, so confirm anything important in the official voter guide (opens in a new tab), which also has all 14 propositions (opens in a new tab) and an audio, large-print and other-language version (opens in a new tab).
Your privacy. Your answers and weights are saved only in this browser's local storage so they are there when you return. They are never sent anywhere, and the tool asks for no name or contact details. Clear my answers removes saved answers. Weights stay until you change them or pick Standard. The page loads fonts from Google Fonts and a visitor counter from GoatCounter, and like any web request these send your IP address and browser details to those services. GoatCounter records one visit each time this page loads, using only the page's fixed name and title, plus the referring site, browser, screen size, language and country. It also records two anonymous counts, when someone answers at least one question and when someone answers all 14. It is never sent which propositions you open or how you answer. Its privacy policy says it does not store IP addresses or set cookies. This page keeps a small note in local storage so each count is sent only once. If storage is blocked, each count is sent at most once per page load. Source links take you to other websites only when you click them.