retti.aiTeardowns › I Legally Cheated a Marathon
Predicted Retention Teardown

I Legally Cheated a Marathon

By Chris Howett · Sports · 702.4K views · 13:04

I Legally Cheated a Marathon

The teardown in brief

What's working

What's costing attention

The first 30 seconds

Today I'm going to be running a marathon in banned shoes. These shoes are banned by World Athletics because they use technology which makes them too good. The shoe I'm wearing is a beefed-up version of the shoe Sebastian Sawe wore to break the world record and run under 2 hours at the London Marathon, which is the mara

Hook fires in 7 seconds with concept, world record context, and the specific PB target all landing before 25 seconds — strong Tier 1 delivery that holds the high end of the mandatory packaging drop for this format.

Where viewers drop

5:17 — Shoe Tech Context Dump (critical)

The race stops dead for 83 seconds while you explain foam measurements, carbon rings, and a Mercedes-AMG partnership. Viewers clicked for a marathon challenge — they're watching you grind through miles 13–16, heart rate up, invested in the outcome. Then suddenly you're in a shoe review.

Why it matters — This is the riskiest moment in the video structurally — you're cutting the tension at exactly the point where the race should be getting harder, and you're replacing it with something viewers could read on a product page.

7:05 — Mid-Race Sponsor Stop (critical)

62 seconds of sponsor content arrives right as your cramps are starting to build and the race is entering its hardest phase. The transition in is hard — 'that's where Runner, Runner! the automated running coach app...' — and all race momentum stops completely.

Why it matters — The viewer was just told you're starting to struggle. They're leaning in. Then you stop everything to pitch a running app. This gives a large chunk of the audience a clean exit right at the worst possible time — the moment they were most invested.

3:22 — Slow Middle Drift (Miles 5–13) (moderate)

For about 115 seconds, the video cycles through several similar 'I'm doing okay but it's getting harder' check-ins — crowd is big, corners hurt, off pace but fine, subscribe joke, Tower Bridge. Each beat is pleasant but nothing escalates. There's no new information, no new obstacle, and no mounting consequence.

Why it matters — Viewers can sense when a challenge video is treading water. The premise (beat PB in banned shoes) hasn't been threatened yet, the stakes haven't changed since the hook, and there's no new tension to chase. Some viewers will start mentally checking out here even if they don't click away yet.

0:00 — Missing Failure Consequence (moderate)

The hook establishes a clear goal (beat 2:42:08) but never says what happens if the goal isn't achieved. As a result, every mile of struggle is entertaining but not tense — there's no fear, only curiosity.

Why it matters — Stakes persistence across a 13-minute video depends on the viewer having a reason to dread failure, not just hope for success. Right now the emotional texture is 'will he do it?' when it could be 'what does he lose if he doesn't?' That difference determines whether the final mile feels like a climax or just an interesting update.

How the video is built

What any creator can steal

Want this on your own video?

Paste any YouTube URL and Retti maps every drop, spike and plateau to the moment that caused it.

Analyse a video free