The Complete CS2 Trade-Up Guide (2026)
A CS2 trade-up turns 10 same-rarity skins into 1 skin of the next rarity, with the output decided by collection weighting and priced by the exact output float. This guide is the map: it explains the full chain, mechanic, float math, real profitability, fees, NPV/IRR, and listing depth, and links to a focused deep-dive for each piece.
To see the theory applied to live inventory, browse current profitable CS2 trade-ups or model your own 10-skin contract in the CS2 trade-up calculator. Everything below tells you what to look at and why it matters before you spend a cent.
Start Here: How Trade-Ups Actually Work
Feed 10 weapon skins of the same rarity into a contract and you get back 1 skin of the next tier up: Mil-Spec to Restricted, Restricted to Classified, Classified to Covert, and so on. Knives and gloves are the exception, those use 5 Covert inputs and return an item from the matching case's special pool.
The output is randomly drawn from the next-tier skins in every collection your inputs touch, weighted by how many inputs came from each collection. Seven Fracture inputs plus three Prisma inputs means a 70% chance of a Fracture output and 30% of a Prisma output. That weighting, not luck, is the lever you control.
New to the mechanic? Start with how CS2 trade-ups work for the 10-skin, float, and odds breakdown, then read the beginner walkthrough in how do CS2 trade-ups work.
The Float Math That Decides Your Output
Output float is not the average of your input floats. It is the average of each input's adjusted float, mapped onto the output skin's own range:
adjusted_float = (input_float - input_min) / (input_max - input_min)
output_float = (avg_adjusted_float * (output_max - output_min)) + output_min
The output's min and max come from the output skin, not the inputs. An AWP Asiimov starts at 0.18, so even a perfect set of inputs cannot make it Factory New. A Glock-18 Fade caps at 0.08, so it is Factory New no matter what. A skin at 0.0699 versus 0.0701 can differ in price by two to five times, even though the two are visually identical.
The full ranges, the adjusted-float formula, and worked examples live in the CS2 float values guide.
Are Trade-Ups Actually Profitable?
Some are. Most that look profitable in a generic calculator are not, because the calculator quotes prices for floats that do not exist at that price. The honest answer comes from checking real listings: a contract a theory engine valued at $2,778 profit came in at $99 once we priced the actual buyable inputs.
For the data-backed version of this answer, read are CS2 trade-ups actually profitable, and for the original side-by-side teardown see theory vs reality.
Why Calculators Lie (The Wedge)
A theoretical calculator assumes an ideal input float, often 0.005, that no cheap listing carries. The real cheapest listings sit at 0.04 to 0.06, which pushes the output from Factory New into Minimal Wear and erases the margin. The sticker price is fiction and the float is fiction; only live listings tell you the truth.
The complete breakdown of how this happens, and why it is the core problem CSAlpha was built to solve, is in why trade-up calculators lie.
Marketplace Fees Compress Every Margin
Fees decide whether a thin trade-up survives. CSFloat charges 2% to sellers and 2.8% + $0.30 to buyers; DMarket charges 2% seller / 2.5% buyer; Skinport charges 8% seller / 0% buyer. The optimal play is usually buy inputs on Skinport (0% buyer fee), sell the output on CSFloat or DMarket (2% seller fee). The $0.30 CSFloat flat fee is a hidden 6 to 18% tax on cheap Mil-Spec inputs.
Exact buyer and seller percentages, with the math on a real $95 contract, are in the CSFloat, DMarket & Skinport fees breakdown.
Probability, Expected Value, and Chance-to-Profit
When inputs span multiple collections, your output is a weighted lottery. Expected value blends every possible output by its probability; chance-to-profit tells you how often the contract clears your cost. A trade-up can have positive EV and still lose money most of the time if one rare high-value output carries the average.
Work through an $80 example in CS2 trade-up probability and expected value.
NPV vs IRR: Ranking by Real Return
Two profitable contracts are not equal if one ties up $400 and the other $40. NPV measures the depth-adjusted dollar profit you can actually realize; IRR expresses return as a rate so you can compare contracts of different sizes and capital lock-up. These are the metrics CSAlpha ranks by, and the advanced NPV/IRR/Budget filters are the gated Pro/Elite feature.
The technical definitions and how they tie to the locked filter are in NPV vs IRR for CS2 trade-ups, explained.
Listing Depth: One Cheap Ask Is Not a Strategy
A single $9 listing does not mean you can buy ten at $9. Buy the cheapest and the next ask is $11, then $13, depth decay. A trade-up that pencils out on the lowest ask can be a loss once you fill all 10 inputs at real, climbing prices. Depth is the difference between a quoted profit and an executable one.
The full concept, with a worked depth-decay example, is in reading listing depth.
Knife and Glove Collections
Knife trade-ups use 5 Covert inputs and pay out from a case's knife/glove pool. The right collection balances affordable inputs, a premium knife pool, and odds that are not diluted into worthless outcomes. The wrong one is a coin flip with a brutal downside.
The 2026 data-ranked shortlist is in 7 best CS2 knife trade-up collections.
Tools: Calculator and Simulator
Use the calculator to test custom inputs, substitutions, and float targets for your own ideas; use the live trade-up table to browse marketplace-backed contracts that already include fee and float math. The difference between the two, and what a trustworthy tool must show, is covered in the calculator guide and the best trade-up simulator.
The Bottom Line
Trade-ups are deterministic math wrapped around one source of randomness: which collection's output you hit. The float calculation has no hidden RNG. Every losing trade-up traces back to the same handful of mistakes, fabricated input floats, average prices, ignored fees, and assuming a single cheap listing equals real depth. Read the linked deep-dives, price against live listings, and you stop guessing.
FAQ
How many skins do you need for a CS2 trade-up?
Standard trade-ups need 10 skins of the same rarity and return 1 skin of the next rarity. Knife and glove trade-ups need 5 Covert skins and return an item from the matching case's special pool.
Are CS2 trade-ups profitable in 2026?
Some are, but most contracts that look profitable in a generic calculator are not once you price real, buyable inputs at their actual floats and subtract marketplace fees. Profitable contracts are found by checking live listings, not theoretical recipes.
How is the output float calculated?
Each input's float is normalized within its own min to max range, the 10 adjusted values are averaged, and that average is mapped onto the output skin's float range. The output's min and max come from the output skin, not the inputs.
What makes a trade-up lose money?
The usual causes are input floats that do not exist at the quoted price, average prices instead of real listing prices, ignored buyer and seller fees, and assuming one cheap listing means you can buy all 10 inputs at that price (listing depth).
Published 2026-06-18 by CSAlpha Team.