The PlayerSells YouTube upload-time analyzer reads up to 100 of a channel's most recent public uploads through the official YouTube Data API, takes the publication timestamp of each one, and buckets them into a 168-cell grid of seven days by twenty-four hours. Every timestamp YouTube publishes is UTC, so the grid is computed in UTC and redrawn in the reader's own timezone in the browser using the same function, which is why the two boards cannot disagree about anything except the frame.
Slots are ranked on an age-adjusted performance index rather than on view counts. Each video's views are divided by the median views of the up-to-eight videos published either side of it, so a video is compared only against videos that had roughly the same time to accumulate views and roughly the same subscriber base behind them. An index of 1.00 is typical for that period, 1.30 is thirty per cent above it. Without that adjustment a channel that grew over two years would rank its oldest uploads highest and call the result a schedule.
The minimums are deliberate and they are the reason the tool sometimes declines to answer. A day or hour bucket needs 2 uploads before it can be ranked, a single day-and-hour cell needs 2, and a channel with fewer than 6 public videos is refused outright. With 168 cells and at most 100 videos, most channels legitimately have no cell that qualifies, and saying so is the honest answer rather than a fault.
The page also prints a 0 to 100 schedule score, which is a measure of upload discipline and not a claim about YouTube's ranking system. Forty of those points come from how steady the gaps between uploads are, and sixty from the share of uploads that land in the channel's own above-baseline day and hour buckets, with the day axis weighted 0.6 against the hour axis at 0.4. Only buckets that clear the 2-upload minimum count towards it, and when too few qualify the score reads "not measured" rather than inventing one.
For scale on the channels this is usually pointed at: across the 35,149-channel YouTube sample PlayerSells measured on 24 August 2026, the median channel has 62,000 subscribers, the 90th percentile has 1,530,000 and the 99th has 12,000,000. Those percentiles describe the channels in our index, not every channel on YouTube.
Upload timestamps, view counts and like counts come from the YouTube Data API v3 and are the same public numbers shown on the video page. Subscriber percentiles are measured over a 35,149-channel sample of the PlayerSells YouTube index on 24 August 2026. PlayerSells is not affiliated with YouTube or Google.
Every parameter the analysis runs on. There is nothing else in it.
| Parameter | Value |
|---|---|
| Uploads read per channel | 100 most recent |
| Minimum videos to analyse a channel | 6 |
| Grid size | 7 days x 24 hours = 168 cells |
| Minimum uploads to rank a day or hour | 2 |
| Minimum uploads to recommend a slot | 2 |
| Performance baseline | median of the 8 neighbouring uploads |
| Schedule score split | 40 consistency, 60 concentration |
| Concentration axis weights | day 0.6, hour 0.4 |
| Source timezone | UTC, as YouTube publishes it |
Start with the day table, not the heatmap. Day of week is the bucket with the most videos behind it, so it is the number least likely to be noise, and it is also the one you can act on: a channel can move its release day far more easily than it can move a specific hour.
Then read the hour table as a window rather than an instant. The tool buckets by the hour a video went live, so "9 AM" means the whole 09:00 to 09:59 window. An upload fifteen minutes either side of a good hour is the same upload.
Treat a single strong cell in the heatmap with suspicion unless it appears in the recommended slots list, because that list is the only part of the page that enforces a minimum sample. A dark cell holding one video is one video.
Finally, weigh consistency above the optimum. The subscriptions tab and the notification bell both reward a schedule an audience can learn, and a channel that holds a merely good slot every week beats one that chases the best slot and misses.
YouTube publishes every timestamp in UTC and nothing else, so the board is computed in UTC and then redrawn in your browser's own timezone. The frame is printed above the heatmap and you can switch between the two. A tool that shows you an hour without telling you whose hour it is has given you a wrong answer that looks precise.
A YouTube video keeps collecting views for years, and a growing channel published its earliest work to its smallest audience. Ranking upload slots on raw view counts therefore measures how old a video is and how big the channel was, not what time worked. Every comparison here uses a performance index instead: a video's views divided by the median views of its neighbours in publish order, so it is only ever scored against videos with roughly the same age and roughly the same subscriber base. An index of 1.00 is typical for that period.
The heatmap has 168 cells and the sample is at most 100 videos, so most channels never put 2 uploads into the same day-and-hour cell. When none reaches that minimum the tool says so instead of recommending a cell that holds one video. The day and hour tables need 2 uploads in a bucket before it can be ranked, and they usually have far more.
Up to 100 of the channel's most recent public uploads, read through the official YouTube Data API. Channels with fewer than 6 public videos are refused rather than analysed, because there is nothing honest to say about the timing of five uploads.
Less than it does on a feed-based network, and more than nothing. YouTube's home page is a recommendation surface rather than a chronological feed, so a video can find its audience days after it goes up. What the upload hour reliably controls is the first few hours: subscriber notifications, the subscriptions tab and the browse impressions that follow from early click-through. That first window is where a video is either picked up or not, which is why the strongest signal in this tool is usually a day of the week rather than an exact minute.
A 0 to 100 measure of upload discipline, and it is explicitly not a claim about YouTube's ranking system. Forty points come from how steady the gaps between uploads are, and sixty from how much of the channel's upload volume lands in its own above-baseline day and hour buckets. It answers one question: is this channel uploading where its own uploads do well, on a rhythm an audience can learn. When too few buckets clear the minimum sample it reports not measured rather than a number.
Yes. Everything read here is public: the upload timestamps, the view counts and the like counts YouTube shows on the video itself. Nothing needs a login and nothing is stored against the channel.
what a channel would sell for
what a channel earns from ads
the percentile a subscriber count sits in
which signals the ranking model rewards or ignores
the whole tracked market, sorted and filtered
what is charting in any country right now, and how fast it got there
score a title before you publish it and see what your own best titles do
PlayerSells is not affiliated with, endorsed by or connected to YouTube or Google. Upload times, view counts and like counts are read from the public YouTube Data API v3 at the moment you run the tool. The performance index, the minimum sample sizes and the grading of a slot are this model's own and are not published by YouTube.
Reads the channel's last 100 public uploads through the YouTube Data API. Nothing is stored against your account and no login is needed.
