Torch & Anvil

Composting Math: the One Free Layer

Verified: Java 26.3 / Bedrock 26.50 · 2026-09-16 · levels, tiers and formula shared by both editions; one item-list difference noted

Every composter question comes down to a single number: how many of a given item it takes to get one bone meal. The answer is not seven, and it is not seven divided by the item's chance. This guide derives the number the composter calculator uses, prices all five chance tiers, and settles which crops are actually worth feeding to the box.

Seven levels, six coin flips

A composter fills in seven steps. Drop in a compostable item and it either adds a level of compost or does nothing, decided by that item's composting chance. When the seventh level lands, the compost sits for 20 game ticks — one second — and then the bone meal can be taken out, which empties the block back to nothing.

The part that changes the arithmetic is a rule about the first item. When the composter is empty, whatever you add always raises the level, no matter how poor its chance. The Minecraft Wiki records the behavior against MC-196452. So a cycle is not seven gambles; it is one certainty followed by six gambles. Each of those six takes, on average, one divided by the chance. Put together, the average number of items for one bone meal is

items per bone meal = 1 + 6/p

with p the composting chance as a fraction. The wiki writes the same rule as y = 1 + 6/x. The gap between that and the naive seven-over-p is not academic: at 30% the naive figure is 23.3 items and the real one is 21, a difference of about one bone meal in every nine stacks.

What each tier costs

The game sorts every compostable item into five chances, and nothing sits between them. That makes the whole economy a five-row table.

Composting chanceItems per bone mealBone meal per stack of 64Items for 64 bone meal
30%21.003.051,344
50%13.004.92832
65%10.236.26655
85%8.067.94516
100%7.009.14448

Read the last column as the shopping list. A full stack of bone meal, 64 of it, is 448 cakes or 1,344 saplings, and the whole spread between best and worst is only three times — which is why the practical question is rarely "what is the best item" but "what do I have too much of".

The two crops that break the pattern

The wiki notes that smaller pieces produce more compost than the blocks they make, and the melon is the clean case. Slices compost at 50% and the block at 65%, so the block looks like the better item until the crafting cost is counted. Nine slices make one melon block. Those nine slices are worth 9 ÷ 13 = 0.69 bone meal; the block they become is worth 1 ÷ 10.23 = 0.10. Crafting melons up before composting them throws away about seven eighths of their value.

Cookies run the same logic the other way and are the best trade in the game. One recipe takes 2 wheat and 1 cocoa beans and returns 8 cookies, which compost at 85%. Those cookies are worth 8 ÷ 8.06 = 0.99 bone meal, against 2 ÷ 10.23 = 0.20 for composting the wheat as it came. One crafting step multiplies the return about fivefold, and cocoa beans are renewable. This is the exception the wiki calls out by name, and it is the only case where crafting before composting pays.

A wheat farm, priced

Take an ordinary wheat farm producing a stack of wheat and a stack of seeds per harvest. The wheat is 65%: 64 ÷ 10.23 = 6.26 bone meal. The seeds are 30%: 64 ÷ 21 = 3.05. The harvest is therefore worth about 9.3 bone meal if everything goes in the box, and two thirds of that comes from the wheat even though the seeds arrive in the same quantity. If the farm feeds a villager trade instead, the seeds are the half worth composting, because the wheat has somewhere better to be.

What this does not model

Every figure here is an average over many cycles. A single composter run is six coin flips and can land anywhere: a 30% item can fill a composter in seven items or eat forty, and no amount of arithmetic changes that. The tool reports expected values because those are what a farm's throughput actually follows over a session, not what any one box will do.

Two items are deliberately left untiered. The straw bed, new in the 26.3 drop, is listed at 65% by the wiki and at 85% by minecraftmaps.com, so this site gives it no tier until they agree. The grass block is compostable at 30% in Bedrock only, which the wiki records as MCPE-140266, so it is called out rather than folded into a shared list. One more discrepancy is worth knowing about: the wiki's own wheat seeds page states that a stack of seeds yields 2.74 bone meal, which is 64 × 0.3 ÷ 7 — the naive calculation that forgets the free first layer. Its composter page gives 3.05, and that is the figure this site publishes, because it is the one the composter page's own formula and table produce.

Sources

The seven levels, the one-second ready delay, the always-lands first item and its bug reference, the five chance tiers with their item lists, the formula y = 1 + 6/x and the published per-stack yields are all on the Minecraft Wiki's composter page, read 2026-09-16. Every tier and every derived figure was checked against the composter calculator at minecraftmaps.com/tools/composter-calculator on the same day, which publishes the same five chances and the same items-per-bone-meal figures. The naive 2.74 figure quoted above is from the wiki's wheat seeds page. The melon, cookie and wheat-farm arithmetic is this site's, built on those chances, and every number printed on this page and on the tool is asserted in tests/mirror_composter-calculator.py, which re-derives the formula from an exact expectation calculation rather than assuming it.