Strain data & methodology: where our strain data comes from and how we check it

What is in the database

The CTU strain database holds 1,184 strain pages (October 22, 2026). Each page lists what sources say about the strain: type, THC (and CBD where it is sourced), terpenes, flavors, commonly reported effects, lineage, breeder, flowering time and grow notes. A dash (—) means no source lists that value yet; we never fill a gap with a guess.

Source: CTU strain dataset, verified Oct 2026 · counts read from the live database on October 22, 2026

Where the data comes from

• Breeders first. When the breeder publishes a value (flowering time, THC, height, yield), it anchors the strain’s figure.
• Then seed banks and public strain listings. We compare several independent sources and store every source with its URL, the value it gives and the date we checked it.
• Polite collection. We respect each site’s robots rules and terms and collect slowly.
• Every strain page lists its sources. The “Checked [date] · X sources” line at the top of the facts opens the full list.

How a value gets on the page

• Two or more sources: we show the range and a typical middle value (the median when there are three or more); clear outliers are dropped and logged.
• One source: shown and labeled “1 source”.
• No source: the field is hidden or shows a dash. Never invented.
• Units are never mixed: indoor yield (per square meter) and outdoor yield (per plant) are kept apart, never averaged together.
• Placeholders removed: generic values from the original import (for example a default 18–22% THC, a default trio of terpenes, default flowering windows and heights, CBD under 1%) are treated as missing, not as data.
• Browse lists (by terpene, flavor, grow trait) are only indexed and linked when at least 12 strains carry real data for them.

How often it is checked

Each strain page shows its own “Checked” date; the current records were last checked in August 2026. We re-check in batches, most-searched strains first. A breeder-first update of grow data (flowering, yield, height, difficulty) is under way, starting with the 200 most-searched strains, then the rest of the database in batches.

What we do not claim

• No medical claims. We do not say any strain treats, cures or helps with a health condition, and we publish no “strains for [condition]” lists.
• Effects are reported, not promised. Words such as “relaxed” or “energetic” summarize what growers and consumers commonly report. Experience varies by person, dose and product.
• THC and terpenes are listed values, not lab results for a product you can buy. Potency, flavor and effects vary by grower, batch and plant.
• A strain website is never shown as a breeder. If the breeder is unknown, we say so.
• Not legal advice. Growing and buying cannabis is legal only in some states; see where cannabis is legal.

Use and corrections

Journalists, researchers and educators may quote figures from the database and from our 2026 strain data report with credit to Cannabis Training University and a link. Found a wrong value? Use “Suggest a missing strain” on the strain database with the strain, the value and your source; nothing is published without a check.