Your last and first frost dates are the clock the whole garden runs on - what they mean, why they're a probability and not a promise, and how they decide when each crop can go out.
What you need
Almost every "when do I plant this?" question is really a question about frost. Your last spring frost and first fall frost mark the ends of your growing season, and nearly every planting date in the garden is counted forward or backward from them. Get these two dates and their meaning straight and the rest of the calendar falls into place.
This is the part most sites get wrong. A published "last frost date" is not the day after which frost is impossible - it is a statistical average, the date by which the last freeze has usually passed in most years. There is still a real chance of a frost after it. Published dates are typically drawn at a probability threshold that leaves roughly a one-in-ten to three-in-ten chance of a later freeze.
That is why this planner hands you the odds, not a single date, for your nearest weather station - so you can decide how much risk to take with a tender crop rather than trusting a date that was never a promise. It reads your local station and tells you the range; the choice of how safe to play it is yours.
Here is one station's spring, drawn from its own thirty-year record. The median last-frost date is not a safe date - it is the middle of the odds, so frost still arrives later in half of years. Waiting for the one-in-ten date buys you a season that frost spoils only about one year in ten:
A crop's frost tolerance decides where in that window it can sit. The tender ones - tomato, basil, squash, pepper - have no tolerance at all and cannot go out until the danger has passed:
Mechanism: Ice nucleation in leaf tissue.
Evidence: settled without a trial - climatology.
Sow or set out after the last frost date, or start indoors and transplant once it has passed.
The hardy and half-hardy crops are the opposite: they tolerate cold and can go out before the last frost, on an offset measured from it - which is exactly how the planner schedules them:
Mechanism: Cold hardiness. Hardy crops tolerate hard frost and half-hardy crops tolerate light frost, so they can occupy the ground weeks before frost-tender crops are allowed in. The offset is measured from the last spring frost.
Evidence: confirmed against 4 sources in the research-based literature.
University of Illinois Extension, 'Time Your Vegetable Plantings by Cold Hardiness' (2018-03-26), extension.illinois.edu/blogs/ilriverhort/2018-03-26-time-your-vegetable-plantings-cold-hardiness: very hardy vegetables 'withstand freezing temperatures and hard frosts without injury' (planted earliest in spring); frost-tolerant vegetables 'can withstand light frosts, but not freezing temperatures' (planted a couple of weeks later). LLM-summarised from a read of the page; it establishes the CATEGORIES but states no week offsets.
Colorado State Univ. Extension, GardenNotes #720, 'Vegetable Planting Guide', extension.colostate.edu/resource/vegetable-planting-guide (read 2026-07-25): hardy vegetables go in 'As early as 2-4 weeks before the date of the average last spring frost'; semi-hardy 'As early as 0-2 weeks before the date of the average last spring frost'; tender from seed 'around the date of the average last spring frost'; very tender 'Typically 2+ weeks after'. This is the source that states the offsets in weeks.
Where this rule sits in that range: hardy at 3 weeks is mid-range of CSU's 2-4. Half-hardy at 2 weeks is the EARLIEST EDGE of CSU's 0-2 - the most aggressive planting CSU sanctions, not the middle of it. The rule is inside published guidance at both ends, but it leans early on the half-hardy side and a gardener following it in a cold spring is taking that boundary.
Do not import an offset from another publication's hardiness scheme without checking the category names: CSU has no 'very hardy' tier, so a '4-6 weeks before frost' figure written against a four-tier scheme does not describe CSU's 'hardy'. Categories are not comparable across sources.
And the whole season has to be long enough for a crop to finish. A long-season crop in a short season won't make it to harvest before the fall frost, so the planner checks that too:
Mechanism: Accumulated GDD vs gdd_to_maturity, or DTM vs frost-free days.
Evidence: settled without a trial - GDD models.
Transplant, choose a shorter-DTM cultivar, or refuse the species for this site.
A common tangle: the USDA hardiness zone, the frost dates, and day length answer three different questions and should never be treated as one. Zone tells you whether a perennial survives the winter; frost dates schedule your annual planting; day length drives things like onion bulbing. Most garden tools collapse all three into "your zone," which is wrong - the explainer your zone does not know your frost date walks through why.
Find your two frost dates, treat them as odds rather than a guarantee, and count from them: tender crops after the last one, hardy crops before it on their offset, and every crop finishing before the fall frost. The first-bed walkthrough puts this into practice on a real planting.
Next in Start here → Frost dates vs hardiness zones vs day length
The rules this guide leans on:
Plants in this guide: Basil