CS2 market guides and tools · educational only · prices are estimates — double-check before you act · not financial advice

Trade Up Guide

CS2 Trade Up Guide: Formula, Odds, Knives and Gloves

Updated for the modern CS2 contract rules, normalized float formula, and the 5 Covert skin knife/glove trade up.

Open calculator

What a trade up contract does

A normal CS2 trade up exchanges 10 weapon skins of the same rarity for 1 weapon skin of the next higher rarity. The result comes from a collection represented by one of your inputs, so the collections you add define the result pool.

The contract is not a price exploit by itself. You are paying for a probability-weighted outcome, and the market usually prices popular input skins around the value of the results they can create.

  • Normal contracts use 10 inputs: Consumer to Industrial, Industrial to Mil-Spec, Mil-Spec to Restricted, Restricted to Classified, and Classified to Covert.
  • You cannot mix StatTrak and non-StatTrak inputs in one contract.
  • Souvenir items, knives, gloves, and items without a valid next trade-up tier are not normal trade-up inputs.
  • All inputs should be checked by exact market name, condition, float, StatTrak status, and collection before buying.
CS2 trade up contract inputs
CS2 trade up contract inputs

How outcome odds work

For regular 10 item contracts, each input collection contributes weight to the result pool. If 7 inputs are from Collection A and 3 are from Collection B, the contract first gives Collection A 70% total weight and Collection B 30% total weight. Inside each collection, that collection weight is split between its eligible next-rarity outcomes.

This is why "one expensive target plus filler" can still be a bad contract. If the target collection has many possible outputs, each individual output gets only a slice of that collection weight.

Outcome chance = collection input share / eligible outputs in that collection

Example: 7/10 inputs from a collection with 2 possible outputs = 35% for each of those outputs

Your expected value depends on every possible output, not only the best one.

The current float formula

The important change is that float should be handled as a normalized position inside each input skin range, not just as a raw average. This matters most when inputs or outputs have restricted float ranges, such as finishes that can never reach the full 0.00 to 1.00 range.

First normalize every input to a 0 to 1 position inside its own min/max range. Then average those normalized values. Finally map the average normalized value into the output item range.

  • Factory New is below 0.07, Minimal Wear is 0.07 to 0.15, Field-Tested is 0.15 to 0.38, Well-Worn is 0.38 to 0.45, and Battle-Scarred is 0.45 to 1.00.
  • A low raw float is not always enough. What matters is where that float sits inside the input skin's own possible range.
  • Before buying inputs, calculate the worst acceptable input float for every target output condition.

Normalized input = (Input float - Input min) / (Input max - Input min)

Average normalized = sum(normalized inputs) / input count

Output float = Output min + Average normalized * (Output max - Output min)

Knife and glove trade ups

The October 2025 CS2 update extended trade up contracts to Covert quality. This is the big new rule: 5 Covert weapon skins can now produce a rare special item.

With 5 StatTrak Covert inputs, the result is a StatTrak knife from a collection represented by the inputs. With 5 regular Covert inputs, the result is a regular knife or a pair of regular gloves from a collection represented by the inputs. Gloves do not have StatTrak versions.

This is still a trade up, not a sure-profit button. You can control the broad eligible pool by choosing the right Covert inputs, but you cannot choose the exact finish, pattern index, or final float.

  • Use exactly 5 Covert weapon skins for the knife/glove contract.
  • Do not mix StatTrak and regular Covert skins.
  • Use case-collection Coverts that actually map to a knife or glove rare-special pool.
  • Regular Covert inputs can return knives or gloves; StatTrak Covert inputs return StatTrak knives.
  • Pattern-based premiums, such as Blue Gems or high Fade percentages, are not controllable through the contract.
Finished CS2 trade up contract
Finished CS2 trade up contract

How to judge profit

A profitable-looking contract can still be negative EV once fees, spreads, float premiums, and bad outcomes are included. Calculate the contract as if you were repeating it many times, not as if you only care about the jackpot.

Use real buy prices for inputs and realistic sell prices for outputs. Steam, Buff163, Skinport, DMarket, Market.CSGO and other supported markets can differ a lot, and the best listing price is not always the price you can actually sell for.

  • Include platform fees and withdrawal costs.
  • Use current prices, not old screenshots or stale averages.
  • Check liquidity. A rare high listing does not mean there is real demand at that price.
  • Avoid overpaying for pattern or sticker value on inputs unless the output math still works without that overpay.

EV = sum(outcome chance * sell price after fees) - total input cost

Profit on one attempt = actual output sell price after fees - total input cost

Practical checklist before you click Proceed

The safest trade up process is boring on purpose: define the target pool, check every possible result, verify float math, then compare EV against the real input cost.

  • Pick the target collection or case pool first.
  • List all eligible outputs and their realistic sell prices.
  • Calculate normalized input floats and predicted output conditions.
  • Check whether the contract is normal 10-item or Covert 5-item.
  • Confirm StatTrak status, trade holds, and market fees.
  • Only proceed if the downside is acceptable, not just because the best hit is attractive.

Useful references