Everything is clear fast
They dated for four years. She ended it. They must stay chained for 30 days. If either crosses the line, they both lose $250,000. We understand the whole video before the first minute.
BLUE PANTHERThe brief ↗I studied 68,184 caption lines, 24,200 comments and five years of public results. This site shows exactly what I found.
The best MrBeast videos have a rule we understand right away and people we care about. A big idea gets the click. The people keep us watching.
These two videos work in different ways. In both, the setting puts pressure on the people, and their choices become the real story.
The chain gets our attention. The relationship keeps us watching. We want to know if they will get close again, argue or quit.
They dated for four years. She ended it. They must stay chained for 30 days. If either crosses the line, they both lose $250,000. We understand the whole video before the first minute.
Sleep, privacy, jealousy and temptation all test the same relationship. Every challenge shows us something new about them.
We find out who gets the money and what happens to the relationship. The ending feels complete because both questions are answered.
Use challenges that change a real relationship. Choose people with history together. Make the final prize finish their story.
Everyone understands a grocery store. The surprise is seeing it become a game. Juan then gives us someone to care about.
Random shoppers learn on camera that the last person inside wins $250,000. Their reactions feel real because they did not know what was coming.
Food, jobs, family, teams and boredom all fit inside one place. The store keeps giving people new choices without another big set.
Viewers start asking who deserves to win, not only who is left. The top comments keep coming back to Juan, his family and his choices.
Use places everyone knows. Cast different kinds of people. Find the person viewers care about and keep coming back to them.
These are not bad videos. They are two videos where a few simple changes could make the story easier to follow.

76.8M views · 474.2K per day · 3.18% like rate
The opening gives us schools, food, child labor, unsafe water and ten locations at once. Every new country starts another story, so there is no one person to follow from beginning to end.
Shorten the repeated arrival, explanation and reveal scenes. By school seven, they feel too much like school three.
Start with one student and one deadline. Keep a world map and a 10-school counter on screen. End by returning to that student.
I Built 10 Schools Around The World
I Built 10 Schools Before Time Ran Out
People who watched loved the mission. One top comment calls it “bigger than YouTube.” But it got fewer views than many nearby uploads. Keep the mission and make the story easier to follow.
Open full teardown →
143.4M views · 814.8K per day · 1.26% like rate
The video asks which age is best, but most of the games reward strength and speed. Viewers started talking about unfair rules instead of the people.
Do not make a child and an adult play the same physical game. Give us time to know each age group before they leave.
Use six tests: speed, memory, creativity, teamwork, courage and judgment. Give each age group a fair way to compete.
Do not name one “best age.” Show what each age does best, then end with a mixed-age team.
The comments show the problem clearly: 90K likes on “why did you make the grandma run,” 83K on “skill based matchmaking,” and 47K on the child-versus-adult mismatch.
Open full teardown →The price ladder already works. I would keep every bunker, but make each stop answer the same two questions: would it survive a nuclear blast, and how long could four people live inside? Experts and a clearly labeled 3D simulation handle the dangerous tests. Government bunkers never need to be damaged or occupied.
“This bunker cost one dollar. That mountain hides a billion-dollar bunker. We put every bunker through the same nuclear survival simulation to answer two questions: does it survive the blast, and how many days could four people live inside? One expensive bunker fails much sooner than you think, and at the end we will spend 24 hours inside the best private bunker.”
Same disaster. Same four people. Two results: blast survival and days alive. Every simulation is shown as an estimate, not a real nuclear test.Before every score, Jimmy and his friends write down how many days they think the bunker will last. The person furthest from the expert estimate loses one comfort for the final 24-hour stay. This gives us jokes, guesses and something personal to follow without turning restricted bunkers into an impossible game.
A 3D nuclear blast hits the $1 bunker and the mountain bunker. Two survival clocks start, but the final times stay hidden.
Use the same blast distance, four people and no new supplies. Score structure, air, water, food, power and medical care.
Give each bunker a fast tour, one safe real-world check and one simulation result. Add its survival time to a leaderboard.
An expensive bunker can lose days because of fire risk, old supplies or weak backup systems. The ranking can now surprise us.
At the $1 billion bunker, film only approved areas. Use public facts and a generic model. Never pretend to test classified systems.
Count down the full ranking. Name the strongest bunker and the best private option, then spend 24 hours there during a controlled power-off test.
Tour → clear answerThe original shows what is inside. This version also tells us what every feature means when survival is the goal.
Price → surpriseThe most expensive bunker may win the blast test but lose on food, power or daily life. The leaderboard can move.
Restricted → practicalRun safe checks only in private bunkers. Use approved facts and generic 3D models when access is limited.
List → real endingThe final ranking answers the hook. The 24-hour stay shows what the winning private bunker is actually like.
These are not rules for every video. They are ideas that showed up again and again in the results, edits and comments.
Exes, strangers and families give us questions that a bigger set cannot answer.
The best openings quickly tell us who is playing, what they can win and how they can lose.
A long video works when the teams, rules, supplies or relationships keep changing.
If viewers talk more about unfair rules than the winner, the rules have taken over the story.
A big mission matters more when we meet one person early and return to them at the end.
With a large cast, the edit must quickly show us who to follow and why.
Each new price should change the same goal, person or problem.
Search, sort and open any of the 121 videos. Every page has a score, the opening transcript, key moments, viewer comments and clear ideas for improvement.
25:41#102
22:45#042
12:39#118
36:49#030
15:40#088
42:57#008
12:20#081
41:27#049
14:06#116
28:25#051
26:10#052
17:43#107
22:26#056
17:10#089
14:47#078
16:11#095
21:42#038
10:50#117
37:04#024
37:25#016
57:51#001
17:35#020
21:32#065
16:28#072The Audience Performance Score measures what viewers did. The Creative Diagnosis explains why the idea may have worked and gives clear actions for the next edit or concept.
60Views at this ageHow its views compare with nearby uploads at the same point after release
20Likes at this ageHow its likes compare with nearby uploads after the same amount of time
10Likes per viewHow often a viewer leaves a public like
10Positive commentsHow often sampled comments use clearly positive audience language
Every case page shows the score, evidence and next action for all six categories.
Daily updates: Public views, likes and performance scores refresh every day. The updater reads YouTube when available and uses the open-source Return YouTube Dislike API when YouTube challenges the server request.
Young videos: Launch traffic is unusually strong. Until day 30, the score is pulled toward 50 and shows its evidence confidence and raw early signal.
Creative Diagnosis: Packaging, hook, structure, stakes, characters and comment response each show what to keep and what to change.
Comments: For each video, I stored 100 top comments and 100 sampled comments.
Captions: 119 videos had auto captions. 2 used creator captions because auto captions were not available.
Not available: I cannot see CTR, impressions, watch time or retention. The moments on this site come from the transcript, not private analytics.