Logline
Recurring battle. Prof High predicts your favorite strain based on your Spotify top artists, your TikTok For You categories, your zodiac, your Wordle streak. Audience confirms or denies in the comments. The 19,000-strain database becomes a party trick.Concept
Personalization that feels magical. Pick a non-cannabis input - something the audience already shares publicly - and run it through the strain matching algorithm. Out the other end comes a single strain prediction. The audience either confirms it (“how did you know”) or denies it (“not even close”), and the comment section becomes the show. The mechanic is built on the same matching engine that powers the app’s personalized recommendations. The inputs are surface-level on purpose. Spotify top five. TikTok For You categories. Three songs you played this week. Your zodiac. The point is not that astrology predicts your strain. The point is that the model is good enough to find a meaningful pattern in almost anything you give it. This show is the lightest, fastest format in the bible. It runs three times a week. It takes minutes to produce once the matching tool is wired up. It also doubles as a permanent recruitment tool - the comment section is full of people asking us to run the algorithm on their inputs, which generates an endless backlog of episodes.Why It Works
Personalization is the hook
People stop scrolling for content that might be about them. The format is engineered to make every viewer think the next prediction could be theirs.
Comment section is the show
Confirms and denies turn the audience into the cast. The format gets stronger with every comment, not weaker.
Defensible by the model
No competitor has a strain-matching engine trained on 19,000 strains plus user feedback loops. The party trick is only possible because of the underlying tech.
Format
| Beat | Runtime | What Happens |
|---|---|---|
| The matchup card | 0:00-0:05 | ”PROF vs ALGO” graphic. Today’s input type announced. |
| The input | 0:05-0:15 | Show the audience-submitted input on the left side of the screen. |
| The model run | 0:15-0:30 | Quick visual of the matching engine narrowing 19,000 down to one. |
| The prediction | 0:30-0:50 | Strain reveal on the right side. AI strain music plays under it. |
| The proof | 0:50-1:10 | One sentence on why the model picked it. Terpene + High Family receipt. |
| Confirmation card | Days later | Repost with the audience’s confirm or deny pinned. |
Platforms
| Platform | Role |
|---|---|
| TikTok | Primary. The native home of the format. Comments drive everything. |
| Instagram Reels | Mirror. Story polls used to crowdsource the next input type. |
| Web | A permanent matcher tool on the marketing site. The show points there. |
| App | Strain match feature gets a soft mention in every episode. |
Cadence
Three times per week. The format is light enough to sustain that pace, and the audience needs the frequency to feel ambient.Example Episodes
Prof vs Algorithm: I predicted your strain from your top five Spotify artists. Did I nail it? The pilot. Five users submit their top five. The model returns a strain for each. Comments confirm or deny. Prof vs Algorithm: TikTok For You categories give you up. Audience screenshots their FYP categories list. The model maps category clusters to terpene profiles to a strain. Surprisingly accurate. Prof vs Algorithm: pick three songs you played this week, here’s your strain. The fastest version. Lowest input cost for the audience. Highest submission volume. Best for high-cadence weeks. Prof vs Algorithm: zodiac plus birth year, with the data caveat. The cheeky version. Professor High names the limits of the input upfront (“astrology is not a terpene”) and then runs the model anyway. Result is unexpectedly coherent, which is the joke. Prof vs Algorithm: when the audience proves the model wrong (and why). A meta episode. Compiles the best “not even close” comments from previous episodes and explains what the model missed. Shows the algorithm learning in public.Production Notes
Recurring “PROF vs ALGO” matchup graphic at the top of every video, styled like a fight-card title screen. Split-screen layout: input on the left, predicted strain on the right. The strain reveal uses the AI strain music from that strain’s soundtrack as the audio bed - a quiet flex of the music pipeline. Confirmation and denial cards are reposted as a separate piece days later, which doubles the content output per episode.Hashtags & Discovery
#profvsalgorithm #strainmatch #cannabisai #thisiswhyimhigh #foryou #cannabistokSuccess Metrics
Submission volume in the comments (target: 100+ inputs per episode within 24 hours). Confirm-vs-deny ratio (target: 60%+ confirmations, which proves the model). Repost rate of the confirmation cards. Click-through to the matcher tool on web. App installs attributed to the strain match feature.Pillar
Community & Engagement with Product & App Features as the soft demo underneath.Status
concept
Related
Spotify = Your Strain
The original music-input version. Prof vs Algo is the broader recurring framework.
Strain Horoscope
The astrology adjacent format. Different vibe, related input mechanic.