This is a nonpartisan, sourced timeline of executive orders and major actions by Trump, Biden, and Obama. It is intended to make the underlying records and the strength of the evidence visible, not to deliver a verdict. Here is exactly how it is built, where the numbers come from, and what they can and cannot claim.
Every numbered executive order is pulled directly from the U.S. Federal Register API (federalregister.gov), filtered to presidential documents of type "Executive Order" for each president. That comes to 278 for Obama, 162 for Biden, and 486 for Trump (926 total), and each entry links back to its official Federal Register record so any title, number, or date can be checked at the source.
The larger or more consequential actions get a deeper write-up ("spotlight"). Each spotlight states what happened, how, and lays the positive and negative impact side by side with sources. Every individual impact claim then carries an evidence label. The label describes how well-established that specific claim is — not whether the order is good or bad — and the same rubric is applied across all three presidents and in both directions:
Substantiated — rests on something not seriously in dispute: a court ruling, an official record, or a measured statistic. Example: "the Supreme Court struck down the IEEPA tariffs 6-3"; "the uninsured rate fell from 15.5% to 7.9%" (Census/CBO).
Projected — rests on a model or forecast that has not fully played out. Example: CBO's estimate that the 2025 tax law adds ~$3.3 trillion to deficits over a decade.
Contested — the evidence does not settle it and credible analysts disagree. Example: how much of 2021–22 inflation the American Rescue Plan caused.
The impact slider is a directional tally, not a severity score. It counts the positive impact findings against the negative ones for whatever is currently in view (respecting filters), takes the net as a share of the total, and positions the marker between "more negative" and "more positive." The visible counts beneath it (positive / mixed / negative, and the number of entries and impact points) are the raw inputs, so it is auditable. A minor negative and a sweeping one each count as one finding, and it reflects only the curated spotlights, not all 926 orders. One honest skew worth naming: negative effects are often easier to document concretely than projected positives, which can tilt a simple tally — so it is best read as a summary of the documented findings, not a judgment of a presidency.
Polling is included only where a reliable measurement exists, and it is left blank — not guessed — everywhere it does not, which is the large majority of entries. Every figure comes from a named, methodologically transparent pollster, and each polled card shows the exact question measured, the approve/disapprove (or support/oppose) split, the pollster, the date, and a link to the source.
Figures are drawn only from organizations that publish their methodology, including:
The second slider ("aggregate public opinion") averages the polled spotlights for the president in view:
(avgApprove − avgDisapprove) / (avgApprove + avgDisapprove) — between "more disapproval" and "more approval."The "Projects" section covers physical or institutional undertakings that are not executive orders but still shape a presidency (e.g., the White House ballroom and East Wing demolition, the Lincoln Memorial Reflecting Pool, the Kennedy Center). Each carries a status, a cost estimate, the same positive/negative impact treatment with evidence labels, and sources. Cost figures are reported as stated by officials or as documented by reporting, with disputes (e.g., a claimed cost versus the figure fact-checkers found) noted on the card.
This is a living document, and the deeper write-ups are curated rather than exhaustive — the full executive-order catalog is comprehensive, but the spotlights cover the most consequential actions, not every order. Selection, source authority, and the line between "projected" and "contested" all involve judgment; the goal is not to eliminate that judgment but to make it transparent and checkable, which is why every claim links to its source. If you find a claim labeled "substantiated" that does not hold up, a weak source, or a major action that was missed or mislabeled, that is exactly the kind of correction I want — please get in touch and I will fix the label or show the source it was based on.
Data: U.S. Federal Register API. Polling: named pollsters as listed above. Built in vanilla HTML/CSS/JavaScript; data pulled and structured with Python. The full structured dataset is available at /timeline_data.json. Not affiliated with any government body.
← Back to the timeline