Seed Starting Schedule and Tray Calculator

Every seed packet gives the same instruction in the same unhelpful form: start six to eight weeks before the last frost. Which means the arithmetic runs backwards from a date the packet does not know, through a hardening-off period the packet does not mention, and lands on a sowing day that is usually a fortnight earlier than people guess.

A placeholder is filled in for you and it means nothing. The real date comes from your cooperative extension service or from the NOAA record for a station near you, and it is a probability rather than a promise — the commonly published date is the one there is a 50 percent chance of frost after, which is not the date to plant tender crops on.
weeks
Tender crops go out after the frost date and after the soil has warmed. Hardy crops go out before it. The packet or your extension service will say which side this crop sits on.
weeks
Straight off the seed packet. It assumes the packet author had reasonable light and warmth; a cold windowsill in March adds a week or two to it.
days
From the packet, at the temperature the packet assumes. Cold media stretches this considerably, which is what a propagation mat is for.
weeks
If the crop moves from a plug tray into a larger pot part way through. Set to zero if it stays in the same cell until it goes out.
days
Days of gradual exposure before setting out. Counted here as extra days added before the transplant date rather than as part of the grow-on weeks. If your packet weeks already include it, set this to zero so it is not counted twice.
Normally today. Move it to see how the schedule reads from a different day.
plants
%
Germination failure, damping off, weak seedlings culled at pot-up. Twenty percent is not pessimistic for a first attempt with old seed on a windowsill; an experienced grower with fresh seed and bottom heat runs far lower.
seeds
Sowing more than one per cell and thinning to the strongest is how growers cut the cell failure rate. It costs seed and it costs a thinning pass.
cells
in
in
sowings
Set above 1 to repeat the whole schedule at a fixed interval, which is how a continuous harvest is arranged rather than one glut.
days
Seed Starting Schedule Calculator — Sow Dates and TraysBuildFigure

The chain, and where the fortnight goes missing

Working a sowing date out backwards has four links and most people count two of them. The transplant date is set relative to the frost date, not equal to it. Hardening off happens before the transplant date and takes a week or so of its own. The grow-on period from the packet runs before that. Germination sits at the front and is usually a matter of days.

Suppose an extension service gives you a last frost date of May 10, and the crop wants setting out a week after it. That puts transplanting on May 17. Seven days of hardening off backs up to May 10 for the start of that. Six weeks of growing on backs up to March 29. If the packet said start six weeks before last frost, a naive reading would have you sowing on March 29 as well only by coincidence — subtract the hardening period and it moves, and if the crop goes out two weeks after frost rather than one, the whole chain shifts again.

StepOffsetExample with a May 10 frost date
Sowstart of the chainMarch 29
Germination expected+7 daysApril 5
Start hardening off+6 weeks from sowingMay 10
Set out+7 days hardeningMay 17

The calculator counts hardening off as extra days added before setting out rather than as part of the grow-on weeks. Some packet instructions already fold it in. If yours does, set the hardening field to zero rather than counting it twice, or you will sow a week earlier than you meant to and have overgrown plants waiting on the bench.

The frost date is a probability, not a wall

The single date most sources print is typically the one with about a fifty percent chance of a frost occurring after it. Half the years, in other words, it frosts later. Planting a tender crop on that date is a coin flip, and this is the most common way a carefully built schedule fails.

Extension services generally publish the same station at several probabilities — a date after which there is a 10 percent chance of frost, a 50 percent chance, a 90 percent chance. For anything that dies at 32 degrees, the 10 percent date is the one to build the schedule on. For hardy crops going out deliberately before the frost date, the 50 percent figure is fine because a light frost is not the failure mode.

Soil temperature is the second half of the answer and it lags air temperature by weeks. Warm-season crops set into cold soil sit and sulk without dying, losing more time than a late sowing would have cost. A soil thermometer at planting depth for a few mornings settles it better than any date does.

Counting cells, seeds and bench space

Work forwards from the plants you actually want. Forty-eight plants with a 20 percent cell failure rate means 48 divided by 0.8, which is 60 cells, rounded up. At two seeds a cell that is 120 seeds. Sixty cells at 72 cells a tray is one tray, with room to spare — the rounding up to whole trays is often where the real number lands, and it is why growing 48 and growing 68 of something can cost exactly the same in bench space.

A standard 1020 tray is about 21.5 by 11 inches, which is 1.64 square feet. Trays packed edge to edge understate the real bench requirement by around a third once gaps, edges and reaching room are counted, which is why the calculator shows both figures. Bench area is the constraint that actually binds in a small greenhouse in April, and it binds worse than seed cost by an order of magnitude.

Sowing two seeds per cell and thinning to the strongest is the standard way to attack the failure rate rather than the tray count. It doubles the seed and adds a thinning pass, and it is nearly always worth it for anything with expensive seed and marginal germination. For crops where seed is cheap and germination is reliable, one per cell and a slightly higher failure allowance is simpler.

Related arithmetic

What the seedlings sit in is a volume question: the growing media and container volume calculator turns a tray count into bags of mix. What they sit on may need bottom heat, and the propagation heat mat calculator works out whether a mat can hold the media temperature in an unheated house and what it costs to run. What they grow under is the usual reason for leggy seedlings, and the daily light integral calculator is the way to check that. Where they end up going is the raised bed soil calculator and, for the general beginner questions, the vegetable garden starter guide.

Questions people ask

Where do I get my last frost date?

From your cooperative extension service, or from the NOAA climate normals for a weather station near you. This page states no date for anywhere, and the placeholder in the field is arithmetic on today rather than a climate figure. Ask for the table at several probabilities rather than the single headline date, and be aware that station data can be a poor match for your own site — a low spot at the bottom of a slope where cold air pools can run a week or two behind a station three miles away on a ridge.

Does the hardening off period come out of the weeks on the packet or get added to them?

The calculator adds it, so that six weeks of growing plus seven days of hardening is 49 days from sowing to setting out. Packet instructions vary in what they include and rarely say. If yours clearly means the six weeks to cover everything up to planting out, set the hardening field to zero. Getting this wrong in the direction of sowing too early is the more common mistake and gives you root-bound plants waiting on a bench for a week, which costs more than sowing a few days late does.

How much should I allow for seedlings that fail?

It depends far more on your conditions than on the seed. Fresh seed on a heated bench with good light can run under 5 percent cell failure; old seed on a cool windowsill with damping-off pressure can run well over 30. Twenty percent is a reasonable first guess for a home setup and it costs almost nothing to be wrong on the generous side, since spare seedlings are easy to give away and a shortfall in May cannot be fixed. Track your own rate for one season and the guess stops being a guess.

Why do my successions all come ready at once?

Because a later sowing grows faster. Two weeks apart in the tray in early spring can compress to nine or ten days apart at harvest, since the second sowing gets longer days, more light and warmer media through its whole life. The fix is to widen the intervals as the season advances rather than keeping them fixed, and to reverse it in late summer when shortening days stretch each successive crop out again. This calculator spaces successions evenly because that is what you told it to do, not because even spacing is right.

Can I use this for direct sowing rather than transplants?

Partly. Set the grow-on weeks and the hardening days to zero and the sow date collapses onto the planting date, which gives you the right answer for the calendar. What it will not tell you is whether the soil is warm enough to sow into, which for direct sowing is the whole question and is not a date at all. The cell, tray and bench sections stop being relevant, though the plants-wanted and failure-rate arithmetic still tells you roughly how much seed to buy.

Related