In-Season vs. Off-Season Produce: 15 Fruits and Vegetables Priced Month by Month

Search for seasonal produce advice and you will find calendar after calendar. What is in season in April. Twelve months of fruit. Each one explains, reasonably enough, that abundant local supply pushes prices down and scarce out-of-season supply pushes them up. The economics are sound. The measurement is missing.

That absence goes all the way up. The USDA's own Seasonal Produce Guide, published through its Food and Nutrition Service, sorts produce into spring, summer, fall, and winter, notes that availability varies with local growing conditions and weather, and makes no claim about cost or savings anywhere in it. The calendar is federal. The price promise is folk wisdom layered on top.

Meanwhile, price data for many of these items has existed monthly, publicly, and free for more than 40 years. It just sits in plain-text files that almost nobody opens.

Why This Is Worth Measuring Rather Than Repeating

Two things happen when a rule goes unmeasured for long enough. It gets applied where it does not belong, and it gets trusted more than it has earned.

If the seasonal discount on bananas is two cents a pound, a shopper who mentally files bananas alongside strawberries is spending attention on a decision that cannot pay them back. And if navel oranges are dearest in October, the shopper who thinks of fall as citrus season is being actively misled by advice that was never checked. Putting numbers on each item separates the places the rule earns its reputation from the places it is just repeated.

It also puts the stakes in proportion. Produce is one line on a grocery bill that varies enormously by geography, as our state-by-state look at the real cost of eating healthy lays out, and a seasonal effect has to be fairly large before it competes with where you happen to live.

The method is deliberately plain, and anyone with a spreadsheet can rebuild it.

Every price in this analysis comes from the Bureau of Labor Statistics Average Price program, downloaded directly from the BLS flat-file archive as the national U.S. city average series. Annual summary rows were dropped so only true monthly observations remain. Each monthly price was then divided by the Consumer Price Index for All Urban Consumers for that same month and rescaled, which converts nominal dollars into constant dollars and removes the long upward drift of general inflation from inside the window.

From there the index is simple arithmetic. For each item, all the January prices in the window are averaged, then all the February prices, and so on. Each month's average is divided by the average of the 12 monthly figures. An index of 1.00 means that month sits exactly on the item's own annual average. An index of 1.20 means 20% above it, and 0.85 means 15% below.

Why 20 Years and Not Five

The obvious window would be the most recent five years, and that was where this started. It does not hold up.

Iceberg lettuce has no published national average price at all from January 2021 through April 2023, and grapefruit none through May 2023. On a 2021 to 2025 window, each item is left with roughly 30 observations spread across 12 calendar months — two or three per month, and for grapefruit a grand total of one October price. On that window iceberg appears to swing 18.8% with a February low. On 20 years of data it swings 11.7% with a July low. The five-year figure was noise wearing the costume of a finding.

Twenty years lifts most cells to 17 to 20 observations, which is enough to average honestly. It also lets the six discontinued series be compared on equal footing against the live ones. The tradeoff is that older years carry price relationships that may have shifted since, which is why the newer five-year results are reported later as a currency check rather than hidden. If you want the longer view of how grocery prices have moved over decades rather than months, our look at what $50 of groceries bought in 1995 versus today covers that ground.

Here is the anchor table, ranked from the largest seasonal swing to the smallest. Find your item, read across.

And here is the month-by-month grid the ranking comes from. Read 1.00 as the item's own annual average, anything above as a premium, anything below as a discount. A dash means the item was not priced in that month.

Look down the columns rather than across the rows and the shape of the whole table appears. Most of these numbers live between 0.95 and 1.05. The seasonal discount, for the majority of this list, is a rounding error on a grocery bill. For context on where the larger savings in a produce aisle actually hide, our comparison of expensive healthy foods and their affordable swaps works a different angle on the same problem.

Strawberries are the rule's best evidence, and it is not close.

The index puts them at 0.81 in both June and July against 1.35 in December — a 67.1% swing built on 234 monthly observations, with 18 to 20 prices behind every single cell. This is not a thin-data artifact. It is the most seasonal price behavior in the entire federal produce record.

Put it in dollars. At the trailing 12-month average of $2.735 per 12-ounce pint, the index prices December at $3.69 a pint and July at $2.21. That is a gap of $1.48 on the same container, or roughly $4.92 a pound in December against $2.95 a pound in July. Same fruit, same store, same shelf, two different decisions.

And the supply side explains it cleanly. USDA's Economic Research Service reports in its March 2026 Fruit and Tree Nuts Outlook that California grows about 90% of the nation's strawberries and ships almost 90% of its volume between April and October, while Florida covers November through April with about two-thirds of its volume moving in February and March. Lower prices are typical in summer when domestic volume is high, and higher prices in winter when it is not. The December pint is expensive because it is arriving from a much thinner pipeline.

The Second-Biggest Bargain Is One Nobody Can Track Anymore

Thompson seedless grapes swing 53.0%, from 0.79 in September to 1.21 in May. It is the second-largest effect in the table, it rests on 233 observations, and it is unusable as current advice, because BLS stopped publishing the series in September 2021.

That is worth sitting with for a moment. The two clearest seasonal bargains in four decades of federal produce pricing are strawberries and Thompson grapes, and half of that pair went dark five years ago. Anyone who wants to know whether grapes are still 53% cheaper in September than in May has no free, national, monthly source to consult.

This is the part of the table that contradicts the advice rather than confirming it.

Navel oranges are cheapest in March and April, at an index of 0.90, and dearest in September and October, at 1.15 and 1.16. That is a 28.0% swing pointing the wrong way for anyone who associates citrus with fall. USDA places navel harvest between October and March, so the October peak sits precisely at the seam between crop years, when last season's fruit is gone and this season's has barely started. Grapefruit does the same thing at 19.5%, cheapest in March and dearest in October. Lemons do a quieter version at 10.1%, with the same March low and the same early-fall high.

Three citrus items, three identical shapes. For a reader stocking up on oranges in autumn because the produce displays suggest it is the moment, the measured difference is about 40 cents a pound against the spring price.

Potatoes Follow the Warehouse, Not the Field

White potatoes are dearest in August, at an index of 1.05, and cheapest in March and April, at 0.97. The swing is only 7.6%, but its direction is instructive: potato pricing tracks the storage year rather than the growing season. The previous autumn's crop is drawn down month by month, prices drift up through late summer as the last of it clears, and then settle again once new harvest arrives.

The lesson generalizes past potatoes. For any crop that stores well, "in season" at the register means something different from "in season" in the field, and the price calendar belongs to the warehouse. Our breakdown of which grocery stores offer the best nutrition per dollar looks at the other half of that equation, which is where you shop rather than when.

Bananas are the flat line at the bottom of the table. A 2.5% swing across 239 observations, drifting from 0.99 in the autumn months to 1.01 in the first quarter. At 65.5 cents a pound, the entire seasonal effect is about two cents.

Broccoli is the more persuasive case, though, because its data is the densest in the whole exercise: 240 observations, a full 20 in every single calendar month, not one gap. And the answer is still a 9.9% swing. That matters, because the obvious objection to a flat result is that the data must be missing something. Here nothing is missing. The season simply does not move the price much.

The same is true of sweet peppers at 10.8%, lemons at 10.1%, romaine at 13.1%, and iceberg at 11.7%. Both lettuces, incidentally, peak in late autumn rather than deep winter, which is not where most shoppers would guess.

Why the Off Season Filled In

There is a structural reason so much of this table is flat, and it is not a measurement problem.

Between 2007 and 2023, the import share of U.S. fresh fruit consumption rose from 50% to 59%, and for fresh vegetables it climbed from 20% to 35%, according to USDA data on fresh produce import reliance. Imports now play an increasingly important role in keeping fresh fruit and vegetables available year-round. Layer controlled-atmosphere storage on top of that, which lets a crop be released into the market evenly instead of all at harvest, and the off-season scarcity that used to create the price spike has largely been filled in.

Strawberries resisted this because they are too perishable to store and still overwhelmingly domestic. Bananas never had a season at the register in the first place. Most of the list sits somewhere in between, and has been drifting toward flat for decades.

For items with almost no price calendar left, the savings question turns into a preparation question instead — which is the kind of arithmetic we ran on batch-cooked chicken versus deli meat, where the difference comes from how the food is bought and handled rather than when.

Percentages are easy to overweight. Here is the arithmetic on a year of ordinary shopping instead.

The scenario is one unit a week of each item — one pint of strawberries, one pound of everything else — for the nine items BLS still publishes, priced at each item's trailing 12-month average, then shaped by its seasonal index. Against that baseline, we move 12 units a year out of each item's three dearest months and into its three cheapest. That is a quarter of the year's buying shifted, which is about as much as a real household could plausibly manage.

Thirty-three dollars and change, on an $876 annual produce outlay. About 3.8%.

And look at the distribution. Strawberries deliver $14.02 of that $33.35 — 42% of the total from a single item. Bananas deliver 17 cents. Potatoes deliver 70 cents, which is less than the price of the potatoes.

So the honest version of the rule is narrow rather than false. Seasonal timing is worth real money on soft fruit, worth a few dollars a year across the salad and citrus aisles combined, and worth nothing measurable on storage crops and year-round imports. A shopper who already knows that strawberries are a summer buy has captured most of the available benefit and can stop optimizing. For the items where the season does not move the price, other levers do: unit price, store choice, and substitution all outweigh the calendar, which is the terrain our ranking of the best protein per dollar maps in detail.

Every number above comes with limits, and they are worth stating plainly rather than burying.

October 2025 does not exist. Every one of the nine live produce series is blank for that month, and so is the CPI itself. The reason is procedural, not statistical: BLS states it could not collect October 2025 reference period survey data for some surveys due to a lapse in appropriations, with a dash representing the missing value in its database and API. Those months were dropped from the calculation, never estimated or filled in. Strawberries are also missing November 2025, and navel oranges are missing July and November 2025.

Some rows deserve more trust than others. Broccoli, bananas, tomatoes, and potatoes rest on 236 to 240 observations with every cell full. Cherries rest on 29 observations concentrated in three months. Anjou pears show a tidy-looking 19.6% swing that depends on a single August price, a single October price, and no September price at all. The observation column in the ranking table is there so you can discount those rows yourself. BLS makes the same point about its own product, cautioning that average prices estimate price levels rather than price change, and that narrow item definitions mean some averages are calculated from substantially fewer prices than a full index would use.

The five-year picture mostly agrees, with one exception. Recomputed on 2021 through 2025 alone, strawberries still swing 56.7% with a July low and December high, tomatoes 15.0%, potatoes 9.4%, and bananas 2.5% — the same story at slightly different magnitudes. Iceberg and grapefruit are the exception, and only because their recent series are full of holes.

This is a national average, not your store. These are U.S. city average prices, which means they smooth across regions, chains, and formats. A local harvest glut, a chain's loss-leader strategy, or a regional shortage can swamp a 10% seasonal effect entirely. An independent cross-check bears out the level, at least where the pack sizes line up: USDA's national retail report for early September 2026 put advertised bananas at 65 cents a pound against the BLS collected average of 65.2 cents for August, and Roma tomatoes at $1.13 a pound. Those are advertised feature prices from a differently built source, and they land close.

Six items are frozen in time. Grapes, peaches, cherries, pears, sweet peppers, and broccoli all stopped being priced nationally between 2019 and 2021. Their rows describe how those items used to behave, over the window shown, and nothing more recent exists to check them against.