Our Grow a Garden 2 Data Corrections

Weather list — corrected. Our first weather list had Blizzard, Disco, Moonlit, Wet, Chilled and Lightning Storm. All six belong to the original Grow a Garden. We caught it by checking the game's own achievement list, which named events our list did not contain. The list was withheld until it was rebuilt.

Pet ability text — downgraded. We had presented pet abilities as exact in-game wording. Independent checking showed every source traces to the same wiki lineage with no screenshot anywhere. They are now labelled as descriptions, not quotes.

Pet prices — both figures were right. Sources split between 15,000,000 and 5,000,000 for the Raccoon and similarly for three other pets. It was a price cut, undocumented in any patch note. We now publish the current figure and name the old one.

Venom Splitter — a misspelling. No such item exists; it is the Venom Spitter, and it is a crop rather than a pet. The error appears in marketplace listings and at least one large gaming site.

Tornado — not weather. It is the Wind Staff gear's ability. Because everyone files it under weather, nobody had a page for the actual item.

Fall Harvest date. We had 2 August 2026. The official announcement posted at 22:00 UTC on 1 August; sites showing the 2nd are reading a late-evening timestamp in their own timezone.

Price floor and minimum weight — withdrawn. We published a price floor on 71 crops, calculated as 0.9025 × the average value. That constant is exactly 0.95², and it only holds if sale value scales with the square of size. The game's own mechanics page gives the exponent as 2.5. The original check had assumed a floor at 95% of average size and then derived the exponent from that assumption, so it confirmed itself. Both figures are withdrawn until someone reads a real floor in game.

Sell price formula — a term was missing. 74 pages gave the formula without SizeMultiplier. The game's own documentation includes it. Because it defaults to 1 the arithmetic on every worked example still checked out, which is why it survived so long. All pages now carry the full formula.

Shop and feature pages — wrong kind of number. Our shop pages carried a price, a Robux cost and a rarity. A shop has none of those; the values described an item each one stocks, pulled in from an item template. The same applied to pages for systems like Crops and Gears. 31 figures withdrawn.

Guild slot costs — bad arithmetic. We said the cost per slot ‘roughly triples’ across the upgrade range. Against our own table it rises from about 7 Robux to about 100 — roughly thirteenfold. The table was right; the sentence under it was not.

Five pages were published twice. Sprinklers, the Admin Chest and the three shops each existed at two urls with identical content, which splits our own search signal. One copy of each was removed and its links repointed.

Restock wait — wrong on a product page. The Venom Spitter page gave its 0.475% restock chance as “roughly one appearance every 10.8 hours”. Against the game's 5-minute shop timer the wait is 17.5 hours, which our own shop-restock table already stated. The figure understated the wait by 39%, on the page where that comparison decides whether you grind for the item or buy it.

Five pages competed with our own better ones. Feature pages for Crops, Eggs, Gear, Pets and Chests each targeted the same query as the far fuller index for that section — a 30-word chests page against a 442-word crates index. The definitions worth keeping moved to the indexes and the thin pages were retired.

Our own trust pages made claims we could not support. The sources page said badge figures were pulled “continuously” when they are a snapshot taken at build time; it listed live player counts and vote totals as sources we use, which we do not publish anywhere; and it accused a competitor of inflating a value by a specific multiple we have no record of. The methodology page said withheld fields render as “not yet confirmed” when they are simply left off, and that everything outstanding was listed on the unconfirmed page when the largest category — crate and egg drop tables — was not. All corrected. A page claiming rigour is the worst place to be loose.

Every site in this niche has been wrong about these things. The difference is whether it is visible.

A correction log is also the only honest way to claim accuracy: anyone can say their data is verified, but showing what you got wrong and when you fixed it is checkable.