MrBeast Says His $10 Million Video Might Not Perform Well. His Channel Added 4 Million Subscribers Anyway

Sep 17, 2026 · ShinyPull Data Team

MrBeast has uploaded to YouTube more than a thousand times. On Saturday, he's putting one video up that he says outweighs the rest of them combined, at least to him.

The video nobody's seen yet

"This Saturday we upload our biggest YouTube video ever," MrBeast posted Tuesday. "It's been in the works for years, I'm personally over a thousand hours deep into it, spent around $10,000,000 on it, and I literally have dreams about the video lol." The post has already pulled 3.5 million views and 72,000 likes in under a day, the kind of number most creators would call a career highlight for a single tweet about a video nobody has watched yet.

What makes the announcement land differently is the follow-up. An hour later, he posted again: "It prob won't perform that well but it's something I care a lot about and I hope my passion for it shines through." That's not the language of someone chasing a number. Years of work and eight figures of spend, paired with an open acknowledgment that the algorithm might shrug at it anyway, is an unusual thing to say out loud when your job is measured in views.

What he's betting it on

Whatever Saturday's video does, it's landing on a channel that was already the single largest thing on YouTube before he ever mentioned it. MrBeast sits at 517 million subscribers, the most of any of the 7,875 YouTube channels ShinyPull tracks, and the gap to second place isn't close. Over just the last 30 days, the channel added 4 million subscribers and 2.89 billion views, an average pace of roughly 133,000 new subscribers every single day, weekends and slow news cycles included.

None of that growth needed Saturday's video to happen. It's what the channel does on a normal month, powered by however many uploads across the main channel and its family of spinoffs land in a given window. The $10 million video isn't trying to kickstart something stalled, it's landing on top of a curve that's already climbing on its own.

The pattern behind the number

This isn't the first time a MrBeast announcement has decoupled from the channel's growth curve. When he got married in July, ShinyPull's own tracking showed his subscriber pace briefly slow the week of the wedding itself, then climb faster than usual once photos and details became public, an audience responding to the person, not just the content. The same disconnect shows up here in reverse: a huge, deeply personal project going up this weekend, on a channel whose daily numbers were already moving before a single frame of it aired.

That's part of why "it prob won't perform that well" is worth taking at face value rather than reading as false modesty. A channel this size doesn't need any individual upload to work. Whether Saturday's video becomes his biggest hit or falls flat next to a channel average, the 517 million subscribers and the roughly 96 million daily views aren't going anywhere either way.

What Saturday tests

The experiment isn't whether the video is popular, MrBeast videos are near-automatically popular at this scale. It's whether a project built around what he personally cares about, rather than what reliably performs, still finds an audience proportional to the channel's size, or whether it becomes the rare data point where the algorithm and what people want to watch pull in different directions from the subscriber count backing it. Either outcome is a measurable answer, and it's one ShinyPull's daily tracking will show clearly against 30 days of a channel that was already climbing before any of this started.

Why this channel keeps taking swings

MrBeast doesn't need to spend $10 million or a thousand personal hours on anything. The channel would keep adding subscribers on autopilot with far cheaper content, the last 30 days already prove that much. What the Saturday video represents instead is a choice to spend the biggest platform in the world's history on something he wants to make rather than something engineered purely to hold its growth rate. Whether that turns out to be a good trade for the algorithm is a separate question from whether it's a good trade for him, and his own two posts suggest he's already made peace with which one he's optimizing for this time.