The AP Top 25 college football and men's basketball polls are released every Monday during their respective seasons, typically around noon ET for football and early-to-mid afternoon ET for basketball. The exact clock time varies week to week because the poll is not a scheduled feed but a snapshot of a voluntary vote that closes on a deadline the night before.
That distinction matters more than it sounds. The "release" is the public moment when the Associated Press publishes a completed tally. It is not the moment the ranking is decided. Understanding the gap between those two events explains most of the confusion around when the poll appears, why it sometimes slips, and why a Sunday-night social media leak is neither authoritative nor a sign the poll "moved."
This explainer walks through the mechanism by which the AP poll is produced and published, quantifies the timing windows you can actually plan around, and flags the edge cases that shift the release.
Anatomy of the Monday release
The AP Top 25 is a human survey, not an algorithm. That single fact drives the entire release schedule.
A panel of voters — sports writers and broadcasters who cover the sport — submits a ranked ballot each week. The AP aggregates those ballots into a points total, where a first-place vote is worth 25 points and a 25th-place vote is worth 1. The final ordering is a straight sort on total points. The release is simply the publication of that sort.
Because the input is human, the poll cannot finalize until every ballot is in. The AP sets a submission deadline, generally late Sunday or very early Monday, so staff can verify ballots, resolve ties in the point totals, and prepare the accompanying editorial package. That production window — ballot close to public post — is why the poll lands near midday rather than at a fixed minute.
The mechanism has a second consequence. There is no underlying statistical model to "run," so there is no computation bottleneck. The delay is logistical and editorial, not computational. When the poll is late, the cause is almost always a late or corrected ballot, not a server.
In college basketball the same architecture applies, but the poll runs from roughly mid-October through the end of the regular season, then gives way to the NCAA tournament selection process. The football poll runs from the preseason release in mid-August through the final post-bowl poll in January.
The timing window in numbers
You cannot pin the AP poll to a single clock time, but you can plan around a reliable window. The ranges below reflect how the release typically behaves across a season; treat them as planning brackets, not published schedules.
| Poll | Typical day | Typical release window | Ballot close | Season span |
|---|---|---|---|---|
| Football preseason | Monday, mid-August | Midday ET | Prior weekend | August |
| Football weekly | Monday | Around noon ET | Late Sunday to early Monday | August to December |
| Football final (post-bowl) | Tuesday after title game | Midday ET | Overnight after title game | January |
| Men's basketball weekly | Monday | Early-to-mid afternoon ET | Sunday night | October to March |
| Men's basketball final | Tuesday after title game | Midday ET | Overnight after title game | April |
A few numbers do more work than the table. The football regular season is 12 games for most teams, which means roughly 13 to 15 weekly releases including preseason and championship weekend. The basketball regular season spans about 31 games per team, producing roughly 18 to 19 weekly polls. The panel itself is in the dozens of voters, which is small enough that a handful of late ballots can visibly shift the deadline.
The single most useful planning number: if you need the poll for a Monday-morning task, do not wait for it. Build your workflow around a 1 p.m. ET buffer on Mondays and a Tuesday buffer for the final polls.
Edge cases that move the release
The presumptive Monday schedule bends in predictable ways, and knowing the triggers prevents wasted refresh cycles.
Holidays are the most common shift. A Monday that falls on a federal holiday can push the poll later in the day or, in some seasons, to Tuesday. The AP has historically adjusted around major holidays rather than forcing a same-day release.
The postseason is the second trigger. The final football poll follows the national championship game, which is played on a Monday night. Because ballots close overnight, the final poll publishes Tuesday, not Monday. The same pattern applies to the final basketball poll after the men's title game.
Breaking news is the third. A late Sunday game, a corrected result, or a disputed ballot can extend the verification window. The AP would rather publish a clean poll an hour late than a wrong one on time.
A counterintuitive detail worth internalizing: a Sunday-night "leak" of the new Top 25 is almost never the AP's own product. Because ballots close Sunday and the tally is mechanically sortable, aggregators and message boards sometimes reconstruct or claim the ranking before publication. The AP's published version remains the authoritative record, and reconstructed versions occasionally differ in the bottom five because ties and points totals are hard to reverse-engineer exactly.
Do not cite a Sunday-night ranking as the AP poll. Do not assume a holiday Monday follows the normal window. Do not treat a delayed release as evidence the poll "changed late." Do not expect the final poll on the same weekday as the weekly polls. Do not assume preseason and Week 1 share a release pattern.
Misconceptions about the drop
Several persistent beliefs about the AP poll's timing are wrong, and each has a mechanical explanation.
The first is that the poll "comes out at a set time." It does not. There is no cron job. The release tracks the human submission and verification process, which is why the window is described in hours rather than minutes.
The second is that the poll is decided Monday morning. In reality the ranking is substantially locked by Sunday night, once most ballots are in. Monday's work is verification and publication. This is why a Monday-morning injury or coaching change rarely reshapes that week's poll — the ballots already reflect Sunday's information, and voters do not resubmit for Monday news.
The third is that the AP poll and the coaches' poll release together. They are separate surveys with separate voter panels and separate production schedules. They often publish within hours of each other, but neither waits on the other. A lag between them is normal, not a signal.
The fourth misconception is subtler and worth a professional caveat. The poll is a measure of voter sentiment, not a predictive model. Its release time is engineered for editorial consumption — a Monday news peg — not for gambling markets, fantasy platforms, or automated feeds. If your system needs the ranking at a precise moment, the AP poll is the wrong dependency; build for the window and handle the miss.
Plainly: the poll is a Monday editorial product built from Sunday-night human input, and it behaves like a news item, not a data feed.
Using the schedule without getting burned
If you depend on the poll — for content, dashboards, or fan communication — the practical actions below remove most of the frustration.
Set your refresh target to 1 p.m. ET Monday for football and 3 p.m. ET Monday for basketball, then treat anything earlier as a bonus. Add a 24-hour fallback for any Monday adjacent to a federal holiday, and a Tuesday fallback for the final polls of each season.
Separate "ballot close" from "publication" in your own workflow. If you are summarizing the week, you can often draft against Sunday-night information and publish the moment the AP confirms. If you are quoting the poll, wait for the published version and link to it.
Watch the preseason and final polls as distinct events with their own timing, rather than assuming the weekly cadence carries over. The preseason football poll arrives in mid-August; the final arrives on a Tuesday in January.
For anything downstream of the ranking — brackets, projections, automated alerts — decouple the trigger from the poll's publication. Use a scheduled window and a manual confirmation step rather than a hard dependency on the release moment.
Limits of the schedule and what it can't tell you
Two limits deserve explicit statement, because they bound what any timing guide can promise.
First, the AP does not publish a minute-level schedule in advance, so every window in this article is a planning bracket derived from how the release behaves, not a guaranteed timestamp. The release is voluntary-vote-driven, and voluntary processes occasionally run long.
Second, timing tells you nothing about content. A poll that publishes at noon is not more or less accurate than one that publishes at 2 p.m. The release schedule is a logistics fact; the ranking is a sentiment measurement with known volatility, especially in the bottom third of the Top 25 where points gaps are small.
If your question is "when does the AP poll come out," the durable answer is Monday around midday ET for football and Monday afternoon ET for basketball, with Tuesday exceptions for the final polls and holiday weeks. Build for the window, verify against the published poll, and treat the schedule as a bracket rather than a clock.