Does adding literal garbage make the stew better? We compared days when Zak threw in scraps and leftovers against days with proper ingredients to find out.
24 days sampled
271 days sampled
The verdict: proper ingredients win by 0.4 points on average. Maybe there's a reason we don't usually cook with week-old pizza crusts.
At what point do too many flavors make the stew taste like absolutely nothing? Every chef knows that mixing every color of paint gives you brown. Does the same apply to soup?
The data reveals a complexity sweet spot around 10/10 complexity, where overall ratings peak at 8.3/10. Push past that, and the flavors start canceling each other out. Zak flew too close to the sun.
Does TikTok's algorithm prefer bad soup? Each bubble is an ingredient category, positioned by average views and taste rating. Bigger bubbles mean more data.
Where Zak the Chef and Zak the TikToker are at war. Categories in the top-left are Hidden Gems: great taste, low views. Categories in the bottom-right are the Engagement Trap where TikTok literally rewards him for making worse soup.
Does a thick, murky soup taste better than a clear broth? Each dot is a day in the stew's life, colored by how Zak felt about the result.
The data maps out the broth's texture space. If the soup is too clear, it's boring. If it's too thick, it's sludge. But right in the middle sits the Zone of Perfection where the green dots cluster. That's perpetual soup heaven.
Using AI embeddings, we gave every video a “vibe check” and mapped the entire history of the stew into distinct eras: The Hearty Winter Broth era, The Too Much Acid dark ages, and beyond.
Explore the Topography of Taste →