retti.aiTeardowns › Every Time I Score, The Defender Gets Stronger!
Predicted Retention Teardown

Every Time I Score, The Defender Gets Stronger!

By Jesser · Sports · 2.6M views · 25:51

Every Time I Score, The Defender Gets Stronger!

The teardown in brief

What's working

What's costing attention

The first 30 seconds

Today, every time we score, the defender gets stronger! We'll be going up against a wide range of opponents, from a middle schooler to a professional bodybuilder to an ex-military beast that can bench press 500 pounds! Me and Johnny have to get through 15 levels. We're starting with 95, we're going off bench press. I'm

Strong Tier 1 delivery — concept ('every time we score the defender gets stronger'), format ('15 levels'), and cast preview (middle schooler to 500lb powerlifter) all land by second 14. The thumbnail/title promise is immediately reaffirmed and the viewer knows exactly what kind of video this is before the first level starts.

Where viewers drop

0:00 — Missing Failure Consequence (critical)

The hook tells you the format — 15 levels, bench press goes up — but never answers the one question every viewer is quietly asking: what happens if Jesser and Johnny can't score? There's a cash prize for the defender, but zero stated consequence for the creator if he fails a level. The viewer has nothing to fear on Jesser's behalf.

Why it matters — Without a failure consequence, every level becomes a fun exhibition rather than a genuine test. You watch hoping he scores, not dreading that he won't — and that emotional ceiling kills the stakes on every single attempt.

9:03 — Repetitive Middle Stretch (Levels 7–11) (moderate)

Between Level 7 (Johnny Da Silva) and Level 11 (Sherman Gay), the video runs through five consecutive opponents in roughly the same rhythm: intro clip → trash talk → gameplay → score or stop → move on. The briefcase twist from Level 8 adds one beat of novelty, but by Level 10 the format feels predictable. You know roughly how each round will end before it begins.

Why it matters — This 8.5-minute stretch covers about a third of the video and is where the curve is going to decay fastest. There's no surprise, no score update that creates tension, and no escalating mechanic to signal 'this level is different.' Viewers who made it through the first half start quietly asking 'how many more rounds?'

1:01 — Cash Prize Stakes Forgotten for 20+ Minutes (moderate)

The cash prize for the defender with most stops is introduced at 1:01 and then essentially dropped until 21:07 when it's briefly referenced. That's a 20-minute gap. The viewer hears it, registers it, and then has no reason to mentally track the stop count throughout the video. When Kevin wins at the end it should feel like a culminating payoff — instead it feels like a reveal we forgot was coming.

Why it matters — Stakes reminders exist to keep the emotional investment active. Every time you show a stop without a scoreboard update or a reference to the prize, you're letting the viewer's investment leak out. The prize is a great structural device — it creates a parallel competition the viewer can root in — but only if they're reminded it exists.

0:00 — Flat Audio Arc — No Contrast for Big Moments (mild)

The audio energy data shows the entire 26-minute video runs at LOUD to VERY_LOUD for 97% of its runtime (only 3% is normal conversational level, zero quiet). This means the biggest moments in the video — scoring against Youngstrong, the weighted ball arc shots, 'Delete that footage' — are at the same energy level as every other moment. There's nothing for the peaks to peak against.

Why it matters — When everything is loud, nothing is loud. The viewer's nervous system calibrates to your baseline within the first two minutes. By minute 10, VERY_LOUD sounds the same as LOUD. The moments that should feel like explosions just feel like more of the same. You're leaving genuine emotional impact on the table on your best material.

How the video is built

What any creator can steal

More teardowns from Jesser

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