Van Born Labs / Guides / YouTube & Creators
How to Find YouTube Outlier Videos for Free
A video with 400,000 views can be a miss on one channel and a breakout on another. Here is how to tell the difference without a vidIQ subscription.
By Arkadiusz Van Born. Updated 2026-11-18 (America/Phoenix).
A million-view video on MrBeast’s channel is normal. A million-view video on a channel that usually does 12,000 is a different object. The second one is an outlier: it performed far above that channel’s usual range. If you make YouTube videos, or you research topics and formats for people who do, outliers are one of the most useful things you can collect. They tell you which ideas already found an audience, even when the channel itself is still small.
You do not need a full vidIQ (or similar) subscription to do this work. You need a clear definition, a honest baseline, and a way to apply it to the videos in front of you. This guide covers the method, the traps, a free Chrome option for browsing, and a bulk option when you want a spreadsheet.
What a YouTube outlier score actually is
An outlier score is a ratio, not a quality rating and not a prediction.
In plain language:
How many times bigger are this video’s views than the views this channel usually gets?
Two reasonable ways to define “usually” are in circulation:
- Channel median. Take the channel’s latest 30 public videos, exclude the video you are scoring, take the median view count, and divide the video’s views by that median. A score of 8.4x means the video has 8.4 times the channel’s typical recent views. This is the definition I use in the Chrome extension described later, because the median is not pulled around by one old viral hit or by one dead upload.
- Views ÷ subscribers. Divide the video’s views by the channel’s subscriber count. A 500,000-view video on a 5,000-subscriber channel scores 100. This is a coarser, but still useful, “punching above its weight” number, and it is the definition used in the YouTube Outlier Score Apify actor when you want video details in bulk. It does not know whether the channel’s last 30 uploads were already huge.
Those two numbers will not match, and they should not be mixed in the same column of a spreadsheet. If you are comparing videos on one channel, the median of recent uploads is the fairer baseline. If you are scanning many channels of very different sizes and you only have subscriber counts, views ÷ subscribers is a decent first filter. Just label the column so you remember which one you used.
What an outlier score is not:
- It is not a forecast of future views.
- It is not “this thumbnail will work for you.” The idea worked on that channel, in that week, for that audience.
- It is not a substitute for watching the video. The first 15 seconds still matter more than the badge.
Van Born Labs
Get a live outlier score
YouTube blocks automated video-stat lookups from our server, so a live score cannot run in this box. The free Chrome extension computes the score in your browser, and Pro unlocks badges, sort and export.
Use the free extension for a live score
Pro adds badges on Home, Search and channel pages, sort, filters and CSV export.
Why raw views and subscriber counts mislead
Creators often sort a niche by view count and then copy the top result. That method systematically over-weights big channels. A video with 400,000 views on a channel whose median is 350,000 is not a breakout. It is the baseline. Meanwhile a video with 80,000 views on a channel whose median is 6,000 is the one you should actually study.
Subscriber count has a similar problem, just inverted. A new channel can have few subscribers and a couple of lucky spikes; views ÷ subscribers will look extraordinary. A large channel can publish a solid video that is still an outlier against its own recent median even if the subscriber ratio looks modest. When you can, prefer a same-channel median for “is this unusual for them?” and use subscribers only when you cannot build that median.
Time matters too. A three-year-old video with high total views may have had a long tail, a recast into a recommendation cluster, or a one-time news event. Pair every outlier score with views per day (views divided by the age of the video, with a floor of one day) so a slow, old pile of views does not look the same as a two-day-old spike.
A manual method you can do today
You can score a video with a notebook and the YouTube UI. It is slow, which is why people stop doing it, but the logic is the same as any tool.
- Open the video. Note the view count and the publish date. If the view count is rounded (“4.2K”), write it as approximate.
- Open the channel’s videos tab. Collect the view counts of the latest 30 uploads, skipping the one you are scoring, and skipping active live streams or premieres if you can see them.
- Sort those 29 or 30 numbers and take the middle value. That is the median.
- Divide the video’s views by the median. That is the outlier score.
- Also divide the video’s views by its age in days. That is views per day.
A practical reading of the ratio, if you want buckets:
- Below 1x: under-performed against the recent baseline. Not automatically a bad video.
- 1–2x: normal-to-strong.
- 2–5x: a real outlier, worth watching for topic, title, hook and packaging.
- Above 5x: a breakout on that channel. Watch it twice: once as a viewer, once as a maker.
Do this for ten videos in your niche and you will already have a better list than “sort by most popular.”
The traps that inflate a score
A few cases should make you downgrade or discard the number:
- Rounded public view counts. YouTube often shows 1.2K or 3.4M in the UI. Any score built on that is approximate. Mark it with a tilde (~) and do not pretend you have studio-level precision.
- Too small a sample. A channel with four public videos does not have a stable median. Either wait, or be honest that the baseline is fragile.
- One old mega-hit in a mean (average) baseline. This is why the median is better than the average. If you ever use an average, exclude obvious outliers from the baseline first.
- Shorts mixed into a long-form baseline, or the reverse. A Short that gets 200,000 views is not comparable to a 12-minute video that gets 8,000. Score Shorts against Shorts, long-form against long-form.
- Lives, premieres, and members-only uploads. Skip them in the baseline when you can.
- Stolen or re-uploaded videos. An outlier score cannot see copyright. If the video is a straight re-upload of someone else’s work, the number is meaningless for idea research.
If a score cannot be calculated honestly, write “unknown” and move on. A missing score is better than a made-up one.
A free way to do this while you browse
If you already spend time on YouTube search, Home, and other people’s channels, the manual loop gets tedious. I built a free Chrome extension, YouTube Outlier Score, that uses definition (1) above: views divided by the median of the channel’s latest 30 videos, excluding the video itself.
On the free tier, you get a badge on the watch page — the score, the channel median, the sample size, the video age, and views per day — with 20 unique video lookups a day. Repeat visits of the same video do not count against you. You can set your own colour thresholds. There is no account to create, no analytics, and no ads. Add it from the Chrome Web Store if you want the badge; the method in this article does not depend on it.
Pro is there if you later want badges on thumbnails, a sort-by-score view, Shorts scored against a Shorts baseline, and a CSV export. It is a narrow tool. It is not a vidIQ replacement, and it should not be sold as one.
When you need a spreadsheet, not a badge
Browsing is the right way to watch outliers. It is a poor way to inventory them. When you want every recent upload from a list of channels, or every result for a search term, with views, likes, tags, chapters and a score in one table, run it in bulk.
The YouTube Video Details & Outlier Score actor on the Apify Store does that without a YouTube Data API key. As noted earlier, its outlier score is views ÷ subscribers, not the 30-video median, so keep the column labelled and do not mix it with extension scores. You also get views per day, which is the better column if you are ranking a large export. Pricing is pay-per-result ($0.80 per 1,000 videos at the time of writing), billed to your own Apify account.
A simple bulk workflow:
- Collect 10–50 channels that already serve the audience you want, or a handful of search terms you are considering.
- Run the actor, export CSV or Sheets.
- Filter to videos from the last 30–90 days, then sort by views per day, then by outlier score.
- Watch the top 20 yourself. Keep a short note: hook, topic, packaging, and whether the idea is copyable or just luck (news, a celebrity, a one-time event).
The last step is the work. No actor will do it for you.
How this compares with vidIQ and similar suites
People search for a “vidIQ alternative” when they want outlier research without paying for a full creator suite. That is a fair instinct, as long as you are honest about the gap.
vidIQ and tools in that class usually bundle keyword tools, competitor tracking, tag suggestions, thumbnail tests, and a browser overlay. An outlier score is one of those features. A free, private extension that only does outlier scoring will not replace the rest of that bundle. If you need the bundle, you should compare those products on their own terms.
If you only need to answer “did this video beat its own channel?”, you do not need the bundle. You need a baseline, a ratio, views per day, and enough humility to watch the video. That is the whole method.
Independent note: YouTube Outlier Score and the Apify actor above are software I make at Van Born Labs. They are not affiliated with YouTube, Google or vidIQ. Public view counts can be rounded; a score is relative performance, not a forecast.
Questions
What is a YouTube outlier score?
It is a ratio: how many times bigger a video’s views are than the views that channel usually gets. It is not a quality rating and not a forecast of future views.
How do I find YouTube outlier videos for free?
You can compute the score by hand from the last 30 uploads, or use a free Chrome extension such as YouTube Outlier Score, which badges the watch page and includes a free daily lookup cap.
Is this a vidIQ alternative?
It can replace vidIQ-style outlier research, but not a full creator suite (keywords, thumbnail tests, competitor tracking). Use a suite if you need those features every week.
Do the Chrome extension and the Apify actor use the same formula?
No. The extension uses views divided by the median of the channel’s latest 30 videos. The Apify actor uses views divided by subscribers. Do not mix the two in the same spreadsheet column.
Tools mentioned
Related guides
Canonical URL: https://arkadiuszvanborn.com/labs/guides/how-to-find-youtube-outlier-videos. Support: arkadiuszvanborn@gmail.com.