How to Balance a Minecraft Server Shop Economy

I built a cobblestone generator on a test world back in June to check whether a design I'd found was as quick as the video claimed, and I sat there for an hour with a hopper counter watching the number climb. The video had promised 28,600 blocks an hour and I counted a bit over 19,000, which was still more cobblestone than any reasonable person needs, and somewhere around the forty minute mark I started thinking about a server I'd looked at a few weeks earlier where the admin shop bought cobblestone from players at six cents a block.

The figure that matters there isn't the six cents, it's the six cents multiplied by nineteen thousand, which comes out at $1,140 an hour for a build that cost me twenty minutes and a bucket of lava. Nobody sets a price like that on purpose, and it happens because the price gets set against what the item feels worth to a player holding one in their hand, and never against what a farm feeding it in bulk will do to the server over a weekend.

The price list I'm working from in this post is a composite, stitched together from a few shop configs and rounded into something typical, so don't read it as any real server's numbers. It's close enough to be useful, because the mistakes in it are the mistakes that keep turning up. Everything below is a way of checking your own list rather than a set of prices to copy off mine.

The only sum that matters

Every sell price in a server shop is a wage, whether the admin who set it was thinking in those terms or not. A player opens the shop, works out which item pays best per hour of their time, and then builds whatever machine produces that item fastest. The sum you want is the sell price multiplied by what the best known farm for that item puts out in an hour, and the answer is the hourly income you've quietly offered to anybody willing to spend an evening following a redstone tutorial.

Cobblestone at six cents a block makes a decent worked example. A basic AFK cobblestone farm does around 3,000 blocks an hour, which pays $180 and looks perfectly sane on a spreadsheet. The easy build I tested claims 28,600, and even at the 19,000 I measured it paid a bit over $1,100 an hour. The multi-unit designs people post run up to 296,000 items an hour, which at six cents a block works out at $17,760, on a server where an iron pickaxe from the adventurer shop costs a hundred.

What the farms put out

I went through the sellable items on that composite list and paired each one with the fastest farm I could find a credible rate for, which took an evening and left me a bit rattled. I started with the farms your players are most likely to build, since those are the ones that turn up on a server first. Two of the rates below I built and counted myself, the cobblestone one and the witch farm, and the rest come from tutorial claims and the wiki, so treat the top ends as optimistic rather than typical. Even the conservative ends are enough to make the point.

Item	Sell price	Automated output per hour	Pays per hour
Carrot / potato	$0.05	54,000+ with a bonemeal micro farm	$2,700
Obsidian	$0.75	1,600 to 17,280	$1,200 to $12,960
Cobblestone	$0.06	3,000 to 296,000	$180 to $17,760
Redstone	$0.18	up to 8,000 from a witch farm	$1,440
Sand / gravel	$0.09	10,000+ from a gravity block duper	$900
Iron ingot	$1.00	400 typical, 608 for a large modular build	$400 to $608

I sorted that table by what each item pays rather than by what it costs, which is not how any shop config I've ever opened is laid out.

Iron is the item every admin worries about, and it sits at the bottom of that table for a reason that has nothing to do with its price. Iron golems can't spawn faster than one every 35 seconds no matter how many villagers you press into service, and each golem drops three to five ingots, so about 400 an hour is a ceiling that no clever design gets past. A large modular build reaches roughly 608 by running several spawning cells side by side, and beyond that you're building a second farm rather than a faster one. An iron farm is a bounded machine, so a dollar an ingot is a figure you can reason about.

Cobblestone, at the other end of that table, has no ceiling at all. A cobblestone generator is lava meeting water, and the only limits are how many you can pack into a chunk and how much of the server's tick budget you're allowed to eat before somebody notices. The same goes for crop farms, where the bonemeal designs skip the growth timer completely, and for gravity block dupers, which still work in 1.21 and turn sand into something with no supply constraint whatsoever. A villager trading hall sits on the same economy from another angle, since emeralds arrive there on a schedule nobody set either.

That difference between a capped machine and an uncapped one is the thing to price against, far more than an item's rarity or how valuable it feels in the hand. Setting a shop price without checking the farm rate is a bit like pricing an all-you-can-eat buffet without asking who's coming, and on a Minecraft server the answer is always somebody with a hopper clock and a free Saturday, which is how a server ends up with its first millionaire on day four.

The witch farm on that list caught me out completely. I built one over a weekend expecting a gunpowder and glowstone machine, and the redstone output was the part I hadn't thought through, because witches drop four to eight dust each and a decent perimeter kills enough of them to push past 8,000 dust an hour. At eighteen cents that comes to $1,440 an hour from a farm most players build for entirely unrelated reasons. Redstone reads like a mining item on a shop list, and it stopped being one when the drop rates changed.

Amethyst is the item that composite list gets wrong in a way I didn't spot until late in the evening. Budding amethyst regrows its clusters forever, so it belongs with the renewable items rather than sitting alongside the ores in the miner's shop, and any price arrived at by comparing it to quartz or lapis has been compared against the wrong thing. The automatic farms are slower than mining by hand, since a piston breaks a cluster for two shards where a pickaxe gets four, so the problem here arrives slowly rather than all at once. Give it a few months on a busy server and the supply is still unbounded.

Quartz

Nether quartz ore doesn't regenerate the way budding amethyst does, and the only renewable source anywhere in the game is piglin bartering, where quartz turns up on roughly 4.36% of trades. That puts a hard bound on supply, because the quartz on a server is a function of how much gold its gold farms produce, and gold farms need a player-dealt kill before any ingots drop at all. I went looking for a quartz farm that gets around this and came back empty handed. Thirty-five cents is defensible in a way almost nothing else in that miner's shop manages, and quartz is the one item on the list I'd leave exactly where it is.

Crafting chains

The other way a shop leaks money has nothing to do with farms. If a player can buy items from the shop, craft them into something else and sell the result back for more than the ingredients cost, the shop turns into a machine for making money without anybody leaving spawn. The test is one line of arithmetic: take the sell price of the output, multiply it by how many the recipe makes, and compare that to the total buy price of everything going in. Any recipe where the first number comes out larger is a hole.

I ran that check across the composite list and couldn't find a single profitable loop, which wasn't the result I'd expected, and two things are doing the work. Most crafted outputs carry no sell price at all, so glass, brick, terracotta, concrete and cake are dead ends by design, and the buy prices sit at four to six times the sell prices on everything else, which absorbs whatever margin a recipe might otherwise have produced. That spread does a lot of quiet work, and it's the first thing I'd look at on a list that isn't holding up.

The near misses are more instructive than the holes. Wheat sells for eight cents, so three wheat is twenty-four cents, and a loaf of bread made from exactly those three sells back for twenty-five. One cent of margin sounds like nothing at all, but a Crafter wired to a villager wheat farm converts every loaf automatically and forever, and a four percent uplift on an infinite supply is still infinite. I tested it on my own world for an hour to be sure I hadn't misread the recipe, and the balance climbed exactly as slowly and as inevitably as the arithmetic said it would.

Sand buys for forty-seven cents on that list and glass has no sell price at all, which is the only thing standing between the shop and somebody's furnace array, because the moment glass gets a sell price above forty-seven cents the loop opens up. Admins add sell prices to no-sell items all the time, usually because a player asked politely and the item looked harmless sitting there on its own. Every no-sell flag on a crafted item is load-bearing even when nothing else in the shop suggests it, which is why glass is the line I'd watch hardest.

There's a bucket in the adventurer shop at fifty dollars, and it's the most consequential item on the whole list, because a bucket is the entire barrier to entry for both the cobblestone generator and the obsidian farm. That adventurer section is buy-only, which makes it a money sink, and sinks are the half of this problem I'm not covering here. Saddles and name tags and ender chests pull money out of the economy once per player and then stop doing anything, so a shop that only drains money when somebody buys their first ender chest has stopped draining by month three.

The traps that aren't design problems

Some of the worst leaks I've seen weren't priced wrong at all, they were matched wrong in the config file of whichever shop plugin the server is running. Coal and charcoal are separate items with separate IDs, and they come from completely different places, one out of a pickaxe and one out of a furnace full of logs, which is a distinction that matters far more to a config file than it ever did to a player. If your shop matches loosely enough to buy charcoal at coal's price, a tree farm and a row of furnaces becomes a coal mine that never runs dry, and coal at forty cents is among the better paying items on the list.

Raw copper and copper ingots go wrong in the opposite direction. A copper ore block drops two to five raw copper before Fortune enters into it, so a config pricing raw copper as though it were a finished ingot pays several times over for one block mined. Stone and cobblestone split the same way the moment somebody brings a silk touch pickaxe to a stone generator. I check item IDs in a shop config before I look at a single price now, because a matching error costs more than a pricing error and takes about ninety seconds to find.

Fortune III

Every sell price on a mining item wants setting against Fortune III rather than the base drop, because nobody mining for profit is swinging an unenchanted pickaxe. Lapis drops four to nine per ore normally and averages around 14.7 with Fortune III, redstone goes from four or five up to roughly seven, and amethyst clusters give four shards to a plain pickaxe against closer to 8.8 enchanted. An admin who prices lapis by mining a few ore with an iron pickaxe and writing down what falls out has underpriced it threefold for everybody who matters.

---

None of this tells you what to charge, and I've deliberately not put a corrected price list at the bottom, partly because the right figure depends on how much automation your server tolerates and partly because I'd be guessing. A server that patches gravity block dupers and caps hopper throughput is a different economy from one that doesn't, and the same sand price is fine on the first and ruinous on the second.

I still can't explain the cobblestone number I counted. The design claimed 28,600 an hour and I measured a bit over 19,000 across three separate hour-long runs, and I checked the obvious culprits, hopper backup and render distance and whether the chunk was staying loaded, without turning up anything wrong. It might be that the tutorial rate was recorded in conditions I wasn't reproducing, or it might be something about how my test world was ticking, and I asked in a redstone Discord and got two answers that contradicted each other.

The part I haven't sorted out is keeping any of this current. Every price is a bet on the fastest known farm for that item, and somebody posts a quicker design every few months, so a list that balanced fine in spring is quietly leaking by September. I don't have a process for spotting that beyond occasionally watching a farm video and getting an uneasy feeling, which is not a system by any definition, and I've been meaning to build myself a spreadsheet for about two years.

Keep your pumpkin lit and your diamonds close. - MrPumpkin