The trade up contract in CS2 takes ten skins of the same rarity and combines them into one skin of the next rarity up. That is the whole idea in a sentence, but the interesting part is the output float math, the collection weighting, and whether the contract actually makes you money. Let me walk you through all of it the way I would explain it to a friend loading their first contract.
Last updated: July 9, 2026. Verified against current CS2 mechanics, July 2026.
The 10:1 mechanic and rarity step-up
Every contract needs exactly ten input skins, and every input must be the same rarity tier. You cannot mix a Mil-Spec with a Restricted. The single output lands one tier higher.
The CS2 rarity ladder for weapon skins goes like this:
| Rarity | Colour | Trades up to |
|---|---|---|
| Consumer Grade | White | Industrial |
| Industrial Grade | Light blue | Mil-Spec |
| Mil-Spec | Blue | Restricted |
| Restricted | Purple | Classified |
| Classified | Pink | Covert |
| Covert | Red | Nothing (top tier) |
Covert is the ceiling. You cannot trade up Coverts into knives or gloves, and you cannot feed knives, gloves, or Contraband items into a contract at all. If you want to browse where each tier sits and what it costs, our Covert rarity hub and the full skins catalog are the fastest way to price a tier.
Which collections mix, and how the output is chosen
Here is where people get tripped up. Your ten inputs can come from different collections, as long as they share a rarity. What changes is the pool of possible outputs.
Each input skin "votes" for its own collection. When the contract resolves, the game picks an output from the next-tier skins of one of the represented collections, weighted by how many inputs came from that collection.
- Feed 10 skins from one collection, and 100 percent of outputs come from that collection's next tier.
- Feed 7 from Collection A and 3 from Collection B, and roughly 70 percent of outcomes draw from A, 30 percent from B.
- Within the chosen collection, the specific skin is picked at random from all skins at that next tier.
That second point matters a lot. If a collection has one expensive skin and four cheap ones at the target tier, your odds of hitting the good one are only one in five even when that collection "wins" the roll. Use the collections browser to see exactly how many skins sit at each tier before you build a contract.
How output float is calculated
This is the part collectors obsess over, and rightly so. The output float is not just the average of your input floats. Each input is first normalised against its own skin's wear range, then averaged, then mapped onto the output skin's range.
The wear scale runs 0.00 to 1.00 and maps onto the familiar conditions:
| Condition | Float range |
|---|---|
| Factory New | 0.00 to 0.07 |
| Minimal Wear | 0.07 to 0.15 |
| Field-Tested | 0.15 to 0.38 |
| Well-Worn | 0.38 to 0.45 |
| Battle-Scarred | 0.45 to 1.00 |
The calculation runs in three steps:
- Normalise each input. For every input, compute
(actual float - skin min) / (skin max - skin min). This gives a 0 to 1 value relative to that specific skin's caps. - Average the normalised values across all ten inputs.
- Map onto the output. Output float =
average × (output max - output min) + output min.
The catch: many skins do not use the full 0.00 to 1.00 range. A skin capped at 0.00 to 0.50 behaves very differently in the formula than one that runs the full range. Two inputs with the same raw float can normalise to completely different values. That is why hand-picking low-float inputs on capped skins is the classic way to force a Factory New or Minimal Wear output. Our trade-up calculator does this normalisation for you so you are not doing algebra in a spreadsheet.
StatTrak rules
StatTrak has its own hard rule: all ten inputs must be StatTrak, and the output will be StatTrak. You cannot mix StatTrak and non-StatTrak in the same contract, and a normal-input contract can never produce a StatTrak output.
A few things worth knowing:
- StatTrak trade-ups draw from the same collections and float math as normal ones. The only difference is the StatTrak tag carries through.
- Souvenir skins cannot be used in trade-up contracts at all.
- StatTrak outputs start their kill counter at zero regardless of the counters on your inputs.
Because StatTrak inputs cost more, the StatTrak contract only makes sense when the target StatTrak skin carries a meaningful premium over its non-StatTrak twin. Sometimes it does, often it does not.
Profitability basics
A trade up contract is profitable when the expected value of the output exceeds the total cost of your ten inputs plus any market fees on the eventual sale. Expected value is simply each possible output's price multiplied by its probability, summed across every outcome.
The honest reality: most naive trade-ups lose money. The profitable ones tend to share a few traits.
- Cheap, low-float inputs on capped skins that force a high-value condition on the output.
- A favourable output pool where the expensive outcomes are more likely, or where even the "bad" outcomes still beat your input cost.
- Collections with a wide price gap between input tier and output tier.
Remember to account for fees when you sell the output. What you net after a marketplace takes its cut is the number that matters, and our fee calculator shows the real payout across each market. For finding the price gaps that make contracts viable in the first place, deals and the markets overview are where I start.
Bringing it together
The trade up contract rewards patience and arithmetic, not luck alone. Pick inputs with intent, understand the output pool, and let the float math work for you rather than against you. When you are ready to build one, price your inputs across every market on our skins catalog, hunt low floats and rare patterns in the pattern finder, and run the numbers in the trade-up calculator before you commit a single skin.
Remy, SkinWise