What is a YouTube comment picker?

A YouTube comment picker is a tool that reads the comments on a public video, Short or livestream replay and selects one or more at random as giveaway winners. A good one loads the complete comment list rather than the ranked "Top comments" view, and lets you deduplicate by channel and filter by keyword before drawing.

Is there a free random comment picker for YouTube?

Yes. Drawing a winner is free with no signup and no Google login. Downloading the underlying entry list is free for the first 100 comments of a video; beyond that the complete file is a one-time payment rather than a subscription.

Do I need to log into my Google or YouTube account?

No. Everything a picker needs is on the public video page, so there is no OAuth screen, no channel connection, and no permission grant. This also means it works on videos from channels you do not own — useful for verifying someone else’s giveaway, and the reason no login is possible in the first place.

Will commenters be notified that I picked from their comments?

No. Reading comments produces no notification, no like, no reply and no visible trace on the video. The only person who knows a draw happened is you, until you announce it.

Are giveaways allowed on YouTube?

Yes, with conditions. YouTube’s contest policies require you to run the contest lawfully, publish official rules, make clear that YouTube is not a sponsor, and not use the contest to inflate engagement metrics artificially. Requiring a comment to enter is fine. Requiring viewers to like or subscribe as a condition of entry sits closer to the line and is worth reading the policy on before you write your rules.

Why does the exported entry list have fewer comments than the count on the video?

Because the counter and the readable list are not the same population. Comments held for review, comments caught by the channel’s blocked-words list, and comments from accounts that were later removed are counted in some views and unavailable in others. YouTube also lets a channel disable comments entirely mid-campaign, which strands every entry already left.

Can I pick a winner from a YouTube Short?

Yes. Shorts are ordinary videos with an ordinary comment list, and a /shorts/ URL resolves to the same video id as a watch URL. In practice Shorts giveaways generate far more single-emoji and one-word comments than long-form ones, so a keyword requirement in your rules is worth more here than anywhere else.

Can I draw from livestream chat?

Not from live chat — that is a separate, ephemeral stream that disappears when the replay is processed. You can draw from the comments left on the livestream’s video page, before, during and after the stream. If your giveaway depends on live chat, tell entrants to also leave a comment on the video, or you will have nothing to draw from tomorrow.

How do I make sure one person only gets one entry?

Deduplicate by channel rather than by comment. Each exported row carries the commenter’s channel name and a stable author identifier, so removing duplicate identifiers leaves one row per channel however many times they commented. Publish this rule when the giveaway opens — a channel that left twenty comments and finds nineteen removed afterwards will say so loudly.

Can someone enter a hundred times with new accounts?

Creating accounts is cheap, so yes, and no picker can fully prevent it. What you can do cheaply is sort the entry list by created_at and look for clusters: a dozen entries in ninety seconds with near-identical text is visible in a spreadsheet in seconds and invisible while scrolling the page. Excluding them is a rules decision; spotting them is a filter.

What columns does the exported file have?

CSV, Excel and JSON exports carry author, username, avatar_url, text, likes, replies, created_at, language, is_pinned, id and reply_to_id, plus the YouTube-specific fields the platform returns. That covers every filter a normal draw needs — time cutoffs, keyword matching, top-level versus reply, and per-channel deduplication.

What proof should I publish with the winner?

Three things: how many eligible entries there were, where the random number came from, and the fact that you kept the entry file. "Entry 812 of 2,447, drawn from the exported comment list" is verifiable by anyone who cares. A wheel animation is not — it is rendered after the winner has already been chosen.

YouTube Comment Picker — Free Random Winner Draw for Videos, Shorts & Livestreams (2026)

If you searched for a YouTube comment picker, you are probably staring at a video with a few thousand comments and a prize to hand out. The good news is that a YouTube draw is technically the easiest of the major platforms to run and the hardest to fake, because the comment list is genuinely readable and the video id in the URL is stable. The bad news is that YouTube quietly removes comments from view in ways Instagram and TikTok do not, and if you do not know which ones, your entry list will be short in a way you cannot see. Here is how to draw a free, random, defensible winner from any public video, Short or livestream replay.

Why a YouTube giveaway is drawn differently from an Instagram one

Three things about YouTube change the shape of the job.

The URL is honest. Every form of a YouTube link — a watch URL, a youtu.be short link, a /shorts/ link, a live replay, a link with a timestamp — resolves to the same eleven-character video id. There is no equivalent of the Facebook problem where a link copied from the app points at a feed story instead of the post. Paste anything and it works.

The default view is a ranking, not a list. YouTube sorts comments by "Top" out of the box, which is an engagement score. Scroll all day and you are still reading a ranked subset in an order the platform chose. Every serious mistake in a hand-run YouTube draw starts here.

The channel is the identity. On Instagram an entrant is a handle; on YouTube an entrant is a channel with a display name that anyone can change to match anyone else’s. Two commenters called "Sarah M" with the same avatar are trivially easy to arrange and genuinely confusing at announcement time. Draw on the stable channel identifier in the file, display the name.

Held for review, spam-filtered, and the comments you will never see

This is the part that catches hosts out, and it is worth being precise about because it changes how you should write your rules.

A YouTube comment can be absent from the readable list for several reasons that have nothing to do with the entrant. It can be held for review because the channel enabled that setting or because YouTube's automated systems flagged it — held comments are invisible to everyone but the channel owner until approved. It can be caught by the channel’s own blocked words list, which many creators populate years earlier and then forget; a giveaway keyword that happens to collide with a blocked term will silently eliminate every correct entry. It can be classified as likely spam, which on a giveaway video is a real risk precisely because giveaway comments look like spam: short, repetitive, emoji-heavy, posted in bursts.

Google's own commentThreads API documentation is explicit that it returns the comments a video exposes publicly — which means anything held, filtered or removed is not there for any tool, ours included. No picker can draw an entry that YouTube is not showing.

The practical consequences for your rules: check your blocked-words list before you choose a giveaway keyword, do not require an exotic string, and if a viewer messages you saying their comment vanished, believe them — it probably did, and it was not the picker.

Picking a winner from a video, a Short, or a livestream

The steps are the same for all three; the pitfalls differ.

Copy the URL. Address bar or Share button, any format.
Load the comments. Paste into the YouTube comment exporter to get the whole list as a file, or the free comment picker to draw a name directly. Neither needs a Google account.
Apply your rules and count. Keyword, one-per-channel, before-the-deadline. The surviving count is N.
Draw and publish. A random integer from 1 to N via random.org, announced with N and the number.

On long-form videos, the main risk is a stale entry list: comments keep arriving for months, so fix the closing time in your rules and enforce it with the created_at column rather than by exporting "around" the deadline.

On Shorts, entries skew tiny — single emoji, one word, a bare channel tag. If your rules require a specific word, a large share of comments will fail them, and you need to have said so in advance. Shorts also accumulate comments faster after the giveaway closes than long-form videos do, which makes the timestamp filter more load-bearing.

On livestreams, the trap is chat. Live chat is a separate ephemeral stream that does not survive as comments on the replay. Entries typed in chat during the stream are simply gone afterwards. If you announce a giveaway mid-stream, say out loud and in the description that entries must be left as a comment on the video, or you will finish the stream with an empty entry pool.

Subscriber-only and "like the video" rules: what is actually verifiable

Giveaway rules on YouTube tend to ask for three things: comment, like, subscribe. Only one of them is checkable.

Comments come back as a list with an author, a channel identifier, a timestamp and text. Anything expressed in comment text is verifiable after the fact.

Likes are a count, not a roster. There is no public per-user list of who liked a video, so "like to enter" cannot be checked at draw time by you or by anyone auditing you.

Subscriptions are worse: a viewer’s subscription list is private unless they have chosen to make it public, and most have not. "Subscribers only" is therefore unenforceable in practice for anyone except the winner, whom you can ask directly.

YouTube’s contest policies are worth reading in full before you write rules, and they also address the underlying reason this matters: contests must not be structured purely to inflate engagement signals. The defensible pattern is the same one that survives scrutiny on every platform — comment to enter, everything else encouraged but not required. It keeps the draw checkable and keeps you clearly inside the policy.

Building an entry list you can defend

The draw is one line of arithmetic. The file is the artefact that answers questions for the next fortnight.

Each exported row carries author, username, avatar_url, text, likes, replies, created_at, language, is_pinned, id and reply_to_id, plus YouTube’s own fields. Four of those carry the weight in a giveaway:

created_at — sort descending, cut at your closing time. A deadline you can point at beats a deadline you remember.
reply_to_id — empty for top-level comments, populated for replies. This is how "top-level entries only" becomes a filter instead of an argument.
text — keyword matching, and the fastest way to spot a copy-paste entry ring: identical strings sorted together.
is_pinned — your own pinned rules comment is in the list too. Exclude it, or one day you will draw yourself.

Number what survives, draw once, and archive the file with the announcement. If you want the spreadsheet mechanics spelled out, the YouTube comments to Excel guide covers the formulas, and analyzing YouTube comments covers what else the file is good for once the giveaway is over.

What YouTube exports look like from where we sit

We build and run the exporter and the picker, so what we can speak to is the aggregate shape of what passes through them, not any individual channel’s numbers.

YouTube exports behave differently from every other platform we support. They are steadier — spread across the week rather than spiking in the hour a contest closes, the way Facebook and Instagram giveaway exports reliably do. The likeliest explanation is that YouTube comment sections stay alive for months, so hosts are not racing a closing bell in the same way, and a meaningful share of YouTube exports are not giveaways at all but creators mining feedback from an old video.

The gap between the displayed comment count and the number of rows you get back is real and asymmetric on YouTube: the file is often smaller than the badge, which is the opposite of what we see on Instagram and Facebook. Held, filtered and removed comments are the explanation, and it is the single most common thing hosts write in about. It is not a truncated export; it is the platform showing a count that includes comments it will not serve.

One more thing worth saying plainly: giveaway comment sections are the least readable text on the internet. Emoji-only entries, tag chains, and the same eight words repeated three thousand times. It is exactly the kind of data a human cannot audit by scrolling and a spreadsheet handles in a minute — which is the entire argument for drawing from a file.

The free cap of 100 comments per post, the no-login picker, and the column names listed above are the live product values at the time of writing, not marketing approximations.

Running the same giveaway across platforms

Very few creators run a YouTube-only giveaway; the same prize usually goes up on Instagram and TikTok too, and the winner comes from the combined pool. The method scales, with one rule.

Export each platform separately, add a platform column, then stack the files in one sheet. Deduplicate within each platform, not across — the same human entering on YouTube and on TikTok is two legitimate entries unless your rules explicitly say one per person overall, and trying to match identities across platforms by display name will produce false matches you cannot defend. Number the combined list, draw once, publish the combined N.

The platform-specific quirks are covered in the TikTok picker walkthrough and the Instagram picker comparison. And on disclosure, the US FTC endorsement guidance applies to giveaways just as it does to sponsored posts once entrants are posting on your behalf — a line of official rules in the description costs nothing and settles it.

Download YouTube comments free →