Editorial policy

Last updated July 1, 2026

This site does not sell dates. Dates are commodity information: the operator announces them, and within the hour they are on fifty sites. What is scarce is knowing which of those fifty pages checked, and which one is running last year's figures with the year swapped.

That is what the rules below are for. Most of them are not editorial preferences at all. They are checks that run in the build, and a page that breaks one does not publish.

The problem this site is built around

Search for the dates of any Disney hard-ticket party in February, months before anybody has announced anything, and the first page of results will state them anyway. Some of those pages are last year's dates with the year edited. Some are last year's dates with nothing edited. Some are a genuine extrapolation from the pattern, written by somebody who knows the event well, which is closer to honest and is still presented to the reader as fact.

None of them mark which. The omission is the part worth being uncomfortable about, because the omission is what makes it work: a page that said "these are last year's dates and this year is unannounced" would lose the click to the page that simply stated them, so nobody says it. The incentive runs one way and the whole category has followed it.

We mark it. Every dated claim here says which of three things it is, and the labeling is not a footnote — it renders next to the claim, in color, on the page.

Two kinds of page, one set of rules

Half of this site is built to be true for years and half of it expires. A height requirement, a ride system, an accessibility provision, and the shape of a park do not move much between one visit and the next. A party-night price, a festival window, a Lightning Lane figure, and a construction wall move every few months.

The two halves get different treatment and one shared stamp. Every page carries the month a person last checked what is on it, permanent or not, because a height nobody has rechecked since 2024 is a claim about 2024. Dated claims carry two things more: a confidence level, and a month by which they have to be rechecked.

A permanent page is not exempt from being wrong. It is exempt from expiring on a schedule. That is the whole difference, and it is why the freshness contract below governs the dated sections of the site while the verified stamp governs all of it.

The three confidence levels

Every dated claim on this site carries one of three labels. The label describes what has actually been announced, not how sure we feel.

Confirmed has to name its source

A confirmed label is a claim that somebody with the authority to set the date has published it. So the entry has to carry a source note saying what that announcement was, and the note renders on the page where a reader can weigh it rather than sitting in a file only we can see.

A confirmed entry with no source note is a validation error and the build stops. That is a low bar and it is set low on purpose: the label is the strongest thing we say anywhere on this site, and the cost of being able to apply it casually is that it stops meaning anything.

Expected is where the real work is

Expected means the pattern, not the date. An event that has opened in mid-August for six consecutive years will almost certainly open in mid-August again, and that is genuinely enough to book flights around. What it is not is a date, and the gap between those two things is where every seasonal site in this category quietly cheats.

So an expected entry may tell you the window it typically occupies, roughly how many nights it usually runs, which nights of the week it usually falls on, and what the last confirmed cycle charged. It may not tell you that the first night is the fifteenth. A fact checker greps our data files for date shapes in any entry that is not confirmed, and a single one fails the build.

This is a harder page to write than the alternative. Stating the pattern well takes more research than stating a date badly, and it converts worse. It is also the only version of the page that is still true in six weeks.

Historical is not a downgrade of the writing

Historical is the label a good part of the dated material wears in the quiet months. It means the last cycle we could stand behind, with its year attached — a party that ran thirty-eight nights and charged between one hundred nineteen and two hundred nineteen dollars in the 2025 season, stated as exactly that rather than as what this year will cost.

The analysis on a historical page is as good as it will ever be. What the event is like, who it suits, which nights are worth paying for and which are a waste — none of that decays. Only the numbers do, and the label is attached to the numbers.

Money is always a range, and the range carries a date

No single price appears anywhere on the dated side of this site. Every monetary claim is a range with the cycle it describes attached, and the component that renders a price will not render one without the as-of, which makes the pairing structural rather than a habit somebody could forget.

A single figure implies a precision that seasonal pricing does not have. Party nights are priced night by night, and the cheapest night and the priciest night of the same event can be a hundred dollars apart. Quoting one number for both is not more useful than a range. It is just wrong in two directions at once.

Snack and menu prices are the one exception, because a single verified item price is the honest shape for a churro. Those carry the month they were checked instead of a cycle, and an older month is a weaker claim — which is exactly what the stamp is there to tell you.

Every page carries the month it was checked

The verified month is on every data page on this site, permanent and dated alike, and it is not decoration. It is the honest statement of when a person last put eyes on what is written there.

Heights, ride systems, operating status, and accessibility provisions are checked against the operator's own published information. Prices are checked against published pricing and cross-checked against the independent trackers that do this full time; where two sources disagree, the most recently verifiable figure wins and it is dated. Where a figure was checked in person, the entry says so. Where it was not, it does not.

A correction changes the underlying data rather than the page, so every page that used the figure picks up the fix on the next build, and the verified month moves with it. A fix that left the old date in place would have the page claiming a diligence nobody performed.

The staleness banner is computed, not authored

Every dated entry declares the month by which it must be rechecked. At build time the site compares that month against the month the build is for, and if the review month has passed, three things happen with nobody deciding: a banner appears above the content saying details on the page are past their review date, the page drops out of the high-priority band of the sitemap, and its structured data stops publishing a price.

There is no field in our data files that suppresses the banner. There is no editor setting, no override, and no "reviewed, still looks fine" checkbox. The only way to remove it is to recheck the facts and move the review date, which is precisely the work the banner exists to ask for.

This is deliberately hostile to us. The person most likely to want the banner turned off is the person who forgot to do the check, which is the exact case it was built to catch, and a control that trusted an author to be honest about their own diligence would be a control that does nothing. An entry that arrives with no freshness block at all resolves to stale as well, because an unstamped page is more likely to be wrong than a stamped one, not less.

What these rules cost us

Search engines reward specificity, and the specific answer is the one we refuse to invent. An event page that will not assert a start date cannot carry the rich result that requires one, so for a stretch of every year about half our event pages publish as ordinary articles while competing pages carry a date-stamped result they earned by making something up.

We also lose the reader who wanted a date and got a window. Some of them go to the page that gave them a date, and most of them never find out it was wrong, because by the time the real dates land nobody goes back to check who called it.

The compensating asset is the only one available: when the announcement finally comes, our page is the one that was not wrong for six months, and the reader who opened the operator's own site in the next tab found us telling the truth. That is a slower asset than a rich result and a more durable one.

Closed attractions keep their pages

When something closes, we mark it closed, date the change, and say what replaced it if anything did. We do not delete the page.

Deleting is what most sites do and it is the wrong call twice over. The page about a ride that is gone is exactly what somebody searches for when they come back after four years and cannot find it on the map. And a site that quietly removes the evidence of what used to be there is a site whose archive cannot be checked against — which matters most for the pages where we were wrong.

Reopening and construction timelines are the same problem with the same answer. We state a timeline as far as the operator has stated it and no further. Anything past that — a date circulating from a permit filing, a construction-watch account, or the pattern of past refurbishments — is labeled as an estimate, because that is what it is. An estimate never renders under a confirmed ribbon, and a tracker entry that offers a firm reopening date without a source note does not publish.

Verdicts, grades, and the overrated call

A guide that likes everything is useless, so every page that could hedge is required not to.

Every month page carries a letter grade, and the grades are spread across the range because a guide that awards all twelve months a B is not a guide. September at Walt Disney World and the week after Christmas are not the same trip and should not receive the same letter. Where the answer depends on who you are — and it usually does, because a retired couple and a family tied to a school calendar are looking at different years — the page names the cases and answers each one, and the trip-timing tool re-ranks all twelve months against whichever of them applies to you.

Every ranked page commits to a verdict. Our best-of pages put things in an order and defend the order rather than listing ten things in no particular sequence. They name the overrated — the headliner whose queue is not worth its payoff, the snack that sells on its photograph, the restaurant charging table-service prices for counter-service food. They also name the underrated, which is the harder and more useful half: the ride nobody queues for that deserves better, the quiet month, the counter-service window with the best food in the park.

This applies to things we would earn money from. Where a ticket product is bad value for a particular trip shape, that is on the page carrying the link to it.

Sourcing, and what we do not claim

The operator's own published information is the primary source for heights, ride systems, dates, prices, hours, and closures. Where an independent tracker and the operator disagree, the operator wins and the disagreement gets a note. Where we are reasoning from a pattern rather than reading a source, the confidence level says so, which is the entire reason for having one.

We do not claim attendance we did not have. You will not find "when we visited last October" on a page where nobody visited last October. Writing from documented fact and stated reasoning is a weaker rhetorical position than a first-person anecdote and it has the advantage of being checkable.

We do not republish another site's text, photographs, tables, or datasets. Where we rely on somebody else's open data — as with the geographic data underlying our maps — it is credited on the page that uses it.

Money cannot move a verdict, a grade, or a confidence level

This site carries a small number of affiliate links, and display advertising if and when it ships. No seller, operator, hotel, restaurant, or brand can buy a place on a list, a better month grade, a softer verdict, or inclusion of any kind. We decline sponsored posts, paid reviews, and paid link insertions, and those requests are declined without a reply.

The confidence level deserves its own sentence, because it is the one a payment would be most valuable to move. Confidence is a factual claim about whether an announcement exists. It is not an opinion, it is not for sale, and there is no mechanism through which a commercial relationship could reach it — the level is set against a source note in a data file that a fact checker reads on every build.

Where we earn a commission, it is disclosed immediately above the link rather than only in the footer, and affiliate links stay on pages where a purchase is genuinely the next step. There are none on the height charts, where somebody is looking for a fact rather than a product. The commercial arrangements themselves are listed in full on the affiliate disclosure.

Corrections

A correction changes the data file rather than the page. Every page that used the figure picks up the fix on the next build, and the entry's verified month moves to the month somebody rechecked it, so the stamp on the page reflects the new check instead of the old one.

Where we cannot confirm a correction quickly, the claim drops to a lower confidence level or comes out until we can. An unresolved doubt does not get to sit underneath a green ribbon. The full handling standard, including how long we take, is on the contact page.

How AI is used here

AI tooling is used in producing this site, in the way most publishers now use it: drafting, structuring, and checking a large dataset for consistency faster than a person can read it. It does not set a grade, a verdict, or a confidence level, and it is never the source of a fact.

That last restriction is the one that matters most here, and it is not a general principle about machines. A model trained on the open web has read tens of thousands of pages confidently asserting unannounced Disney dates, and it will reproduce them fluently and without hesitation. On the single question this site exists to answer correctly, it is the worst available source.

So every height, price, closure, mechanic, and dated claim is checked by a person against a primary source before it publishes, and the verified month records when that person did it. We say all this because the alternative is pretending otherwise, and the pretense collapses the first time anyone looks closely. The standard we are asking to be judged on is whether the page is accurate, original, and useful.

What never appears here

No logos, wordmarks, character art, official artwork, official photography, or official park maps appear on this site. We do not use photography whose primary subject is a copyrighted character, and we do not imply affiliation or endorsement. Park, event, attraction, restaurant, and character names appear because there is no other way to identify the thing being written about. The full position is on the terms of use.