"When's my last frost?" doesn't have one national answer, and the gap between regions is bigger than most people expect. Using Earth Sciences New Zealand's own ground-frost climate records, Wellington's frost risk typically eases by August, while Canterbury commonly holds onto it until December, a four-month spread across a country that isn't even particularly large.
The far north barely enters the conversation at all. Northland sees so little ground frost in an average year that waiting for a "last frost" isn't really a useful planting strategy there, soil temperature and rainfall matter more than cold. Auckland is close behind, clear by September most years, followed by Hawke's Bay and Taranaki, both typically safe from October.
The West Coast clears by October too, similar timing to Hawke's Bay and Taranaki despite the very different climate, before the country's largest single group: Waikato, Bay of Plenty, Gisborne, Manawatū-Whanganui, Nelson, Tasman, Marlborough and Otago all typically clear by November. That's a wider, colder-climate group than the traditional Labour Weekend rule usually gets credit for, Otago among them: the region's reputation for a long, hard winter is real, but its last frost timing itself lands in the same month as Waikato's, even though the frost along the way is considerably heavier.
The West Coast clears by October too, similar timing to Hawke's Bay and Taranaki despite the very different climate, before the country's largest single group: Waikato, Bay of Plenty, Gisborne, Manawatū-Whanganui, Nelson, Tasman, Marlborough and Otago all typically clear by November.
Canterbury runs latest of any region with a genuine frost-free season, often not clearing until December. Southland is the true outlier: the region's own data shows measurable frost risk in every month of the year on average, which means a fixed "last frost" doesn't really apply there the way it does everywhere else, greenhouses and cloches do more work for a Southland garden than calendar timing ever could.
Two things are worth remembering about all of this. First, these are averages built from decades of climate data, not predictions: a mild year can clear frost weeks early, and a cold one can deliver a surprise well past the usual date. Second, the cost of the two mistakes isn't symmetric. Planting a week or two after the average last frost costs you almost nothing. Planting a week or two before it, on the strength of a warm spell, risks losing the entire planting overnight.