CS2 Skin Liquidity by Category, and How Far Marketplaces Disagree on Price
Which categories of CS2 skin actually trade, measured with CSDB’s published liquidity formula over 30-day sales, active listings and spread — and how much the 19 marketplaces in one dated snapshot disagree on the same item. Market characteristics, not advice: see the market overview for the current snapshot and the price tracker for individual items.
Data as of: Published: 2026-09-11
n = 2,062 skins with a liquidity score. 8,566 price rows (one per skin and wear) enter the spread analysis; 412 rows from the weekly phase feed or with fewer than two comparable markets are excluded.
See also: CS2 tools, build your crosshair, pro settings, lineups, guides, maps, console commands, pro crosshair codes.
Key findings
- Pistols are the most liquid category (median score 80, n = 425; median 2,718 sales in 30 days) and Knives the least (median score 53, n = 584; median 64 sales).
- 49% of all 2,062 scored skins are “very liquid” on CSDB’s formula; only 0% are illiquid — the catalogue is priced items, not the long tail of unlisted ones.
- Marketplaces disagree far more than a single price suggests: across 8,566 price rows quoted on two or more markets, the median gap between the cheapest and dearest market is 53.2%; only 1% of rows are within 10% and 67% are 30% or more apart.
- Dispersion is a price-band effect: Under $1 items have a median spread of 161% (n = 2,552rows), falling to 25.1% in the $100–$1,000 band (n = 1,606).
Liquidity by category
| Category | Skins | Median score | Very liquid | Liquid | Moderate | Illiquid | Median 30d sales | Median listings | Median spread |
|---|---|---|---|---|---|---|---|---|---|
| Pistols | 425 | 80 | 68% | 25% | 8% | 0% | 2,718 | 5,635 | 59.4% |
| Rifles | 471 | 79 | 67% | 24% | 9% | 0% | 2,028 | 5,786 | 55% |
| SMGs | 293 | 79 | 66% | 26% | 7% | 0% | 2,926 | 4,442 | 92.5% |
| Heavy | 210 | 77 | 61% | 30% | 7% | 1% | 2,105 | 3,673 | 100% |
| Gloves | 72 | 70 | 50% | 19% | 31% | 0% | 337 | 3,747 | 29.7% |
| Knives | 584 | 53 | 6% | 60% | 34% | 0% | 64 | 1,459 | 25.5% |
Cross-marketplace price dispersion
| Cheapest-market price | Price rows | Median spread | P75 spread | Within 10% | Median markets |
|---|---|---|---|---|---|
| Under $1 | 2,552 | 161% | 328% | 0% | 12 |
| $1–$10 | 1,991 | 54.3% | 125.5% | 1% | 13 |
| $10–$100 | 2,242 | 36.8% | 84% | 1% | 12 |
| $100–$1,000 | 1,606 | 25.1% | 57.9% | 1% | 12 |
| $1,000+ | 175 | 29.3% | 60.9% | 5% | 11 |
Overall: median 53.2%, upper quartile 146.9%. A wide spread on a cheap item is usually one market quoting a few cents above the others; on an expensive item it is usually a real regional or fee difference.
Methodology
- Universe: every skin in the committed price snapshot (2,062); 2,062 have enough activity to score (sales or listings above zero).
- Liquidity score = 55% velocity (log-scaled 30-day sales) + 25% depth (log-scaled active listings) + 20% tightness (1 − spread ÷ 30%, neutral when no spread is measurable). Labels: 70+ very liquid, 45+ liquid, 20+ moderate, below 20 illiquid. Sales and listings are summed across every wear and variant of the skin.
- Spread is computed per price row (one skin, one wear) from the daily feed only: (highest − lowest) ÷ lowest across markets quoting above zero, Steam Market excluded because its price is gross of Valve’s cut. Rows with fewer than two comparable quotes are dropped. Price band is the row’s cheapest quote.
- Categories below 30 scored skins and price bands below 30 rows are not published. Medians are linear-interpolated.
Sources: SteamWebAPI bulk price feed (daily snapshot), CSDB liquidity score v2 (published formula). Computation: src/lib/research/skin-liquidity.ts. General rules: Data Lab methodology.
Limitations
- One snapshot, one day. Liquidity is a 30-day window ending at the snapshot; spread is that morning’s quotes. Neither is a trend, and the page says nothing about where any price goes next.
- Spread is max-minus-min over every quoting market (usually about 12), so a single outlier quote sets it. It measures disagreement, not the deal available to any one buyer, and it ignores fees, withdrawal limits and regional currency conversion.
- Doppler and Gamma phase rows come from a weekly feed and are excluded from the spread analysis; their parent skins remain in the liquidity scores.
- “Listings” and “sales” are what the feed reports across the markets it covers; markets outside the feed are invisible here.
- The catalogue is skins with a market price, not every item in the game; unlisted items are absent, which makes the whole set look more liquid than the game’s item pool.
Source: CSDB.gg, "CS2 Skin Liquidity by Category and Cross-Marketplace Price Dispersion", data as of 14 Sep 2026, accessed [date]. https://csdb.gg/data/skin-liquidity/