How Squishy Dumpling Values Are Set, and Why Your Trades Still Feel Unfair

Most arguments in the trading hub are not really about a dumpling. They are about a number that two players are holding in their heads, each with a different origin story. One of you is pricing off what you paid last week. The other is pricing off what someone offered them an hour ago. Neither number is wrong, exactly, and that is the problem.

This post is the honest version of what sits behind the trade calculator: where the numbers come from, what they are good at, and the places where they will quietly mislead you if you treat them as gospel.

Rarity bands do most of the work

Every squishy in our dataset lives in one of seven bands: Common, Uncommon, Rare, Epic, Apocalypse, Animal, and Heart. The band is not decorative. It is the single strongest predictor of what a dumpling is worth, and the gaps between bands are much bigger than the gaps inside them.

Read that last column again, because it is the part people miss. The multiplier between bands shrinks as you climb. Common to Uncommon roughly triples your value. Animal to Heart lands between 1.6x and 2.1x depending which ends of the two ranges you compare. The ladder gets flatter near the top, which means the last few rungs cost far more input value per rung than the first few.

These multipliers are arithmetic on our own value table, nothing more. They describe our model of the game, not a measurement of it.

Inside a band, the spread is narrow. Two Commons sit at 5 and 6. Two Hearts sit at 900 and 1100. If someone insists that their same-band dumpling is worth double yours, the value model does not support that, and you should ask what they are actually pricing.

The x3.5 mutation premium is a floor, not a ceiling

A mutated squishy is valued at 3.5x its normal estimate throughout this site. A Plain Pork at 5 becomes about 18 mutated. A Cupid Dumpling at 1100 becomes about 3850. That multiplier is applied consistently so that a trade with mutations on both sides still nets out sensibly.

The number to hold on to is the absolute premium, not the multiplier. A mutated Common adds around 13 points of value. A mutated Heart adds close to 2750. Same multiplier, wildly different consequences. This is why a mutated low-tier item is a fine sweetener and a mutated high-tier item is basically the whole trade.

Practical rule: never let a mutated top-band squishy be the thing you throw in to round out an offer. It is not a rounding item. It is the offer.

Demand weighting is why two calculators disagree

We run two verdict engines on purpose, and they answer different questions. Both use the same 5 percent fair band. The trade calculator compares raw totals and stops there. The WFL checker first multiplies each item by its demand label, rising items at 1.15x and cooling items at 0.90x, and only then compares.

Because the threshold is identical, any disagreement between the two comes entirely from demand weighting, which makes the disagreement itself informative. If you are handing over items we have labelled as rising and receiving ones we have labelled as cooling, the raw totals can match while the weighted totals do not. The two multipliers sit 25 percentage points apart, which is more than enough to flip a verdict on its own.

Worth being precise about what that demand label is: it is a hand-maintained field in our dataset, our read of which way an item is moving. It is not measured trade volume, because no such feed exists for this game. Treat it as an opinion we are willing to show our working for, not a statistic.

Three reasons a fair trade still feels wrong

1. You are pricing effort, the calculator is pricing outcome

If a squishy took you a long grind to obtain, your internal price includes that grind. The value model does not care how you got the item. It only cares what the item is. This is not the calculator being cold. It is the reason it is useful: the other player does not owe you your farming time back, and a tool that tried to price your effort would give a different answer to every player.

2. Quantity scales linearly, the ladder does not

Multi-item offers are where the arithmetic catches people out, in both directions. Ten Commons is 50 to 60 points, which is genuinely close to a Rare at 45 to 52, so that offer is more reasonable than it looks. The same ten Commons against an Epic at 120 to 140 is roughly half, and against a Heart-tier squishy it is a rounding error. A pile grows in a straight line while the ladder does not, so the bigger the item on the other side, the worse volume gets. This is exactly the sum you should be doing before you accept, not after.

3. Fair is a band, not a point

A verdict of fair does not mean the sides are identical. It means the gap fell inside a tolerance, and on the trade calculator that tolerance is 5 percent. On a trade worth 1000, the band still allows a 50 point difference. Repeat that often enough while always taking the wrong end of it and you will bleed value while every single trade reads as fair.

What this model cannot tell you

Being straight about the limits, because they are larger than most value sites will admit. This is a hand-maintained table of 14 squishies. It is not a price feed, there is no telemetry behind it, and nobody, including us, can see what trades actually clear at. Any value list for this game, ours included, is a model of what a sensible trade looks like rather than a record of what trades happened.

• Updates move the ground. A change to how an item is obtained can reprice it faster than any value list reacts. Nobody publishes patch notes for this game, so our update log does not pretend to track what changed: it tracks the signals that tell us to go and re-check.
• The high end is the least certain. The top-band numbers are the ones we are least confident in, simply because there is less to go on up there. Weight them accordingly.
• Nothing here models a specific person. If a trader is chasing one exact squishy to finish their set, they will pay above any list, and no calculator predicts that.

Use the numbers as a guardrail against the trades that are obviously bad, which is what a value model is genuinely good for. Do not use them as a hammer to win an argument. The player across from you has their own model, and being 4 percent right about a fair-band trade is not worth burning a trading partner over.

Where to go from here

If a specific offer is in front of you right now, run it through the WFL checker rather than eyeballing it, then look up anything you did not recognise on the values list. Two minutes of checking beats a week of regret.