CS2 Skin Market Report October 2026: 53 of 55 Items Fell, and Two "Risers" That Weren't

Yes, and by more than in August. Measured like-for-like (the same 55 items, CSFloat's lowest listing at each date), the market lost ground in every window I can compare:

The 22 August row needs one sentence of context. Between 22 August and 13 September prices roughly paused: my median index for most groups ticked up slightly over those three weeks before the October snapshot came in lower. So the decline was not one steady slide. It was a summer dip, a short pause, and then a second and larger leg down.

It also hit almost everything. Of the 55 items, 21 fell by more than 10% in the thirty days, 20 fell between 5% and 10%, and 12 fell by less than 5%. A median that moves while a few items crash can mislead you; this one moved because nearly every item drifted down together. That is the pattern I described in why CS2 skin prices drop: when the whole board falls together, the cause is usually something that hit everyone, not one skin's story.

Price level mattered more than item type. Splitting the 55 items by their 31 July price:

• Under $50 (18 items): median −14.7% since 31 July, one riser.
• $50 to $1,000 (27 items): median −12.8%.
• $1,000 and up (10 items): median −4.7%.

Expensive items held up best and cheap ones fell hardest, the same ordering I saw in August, only wider. One caution before reading too much into that: the top band is ten items, mostly Dragon Lores and knives, so a single quiet listing can move its median.

Cases, by a distance. Here is the thirty-day median by category, from the chart above:

Inside the case group, the cheap cases did all the falling and the expensive ones held. Since 31 July, CSFloat's floor on the Revolution Case went from $0.26 to $0.16 (−38.5%), the Recoil Case from $0.31 to $0.20 (−35.5%), and the Kilowatt Case from $0.15 to $0.11 (−26.7%). Over the same 74 days the Chroma 3 Case moved from $3.06 to $3.00 (−2.0%), the Gamma 2 Case from $2.80 to $2.64 (−5.7%), and the Fever Case actually rose, $0.52 to $0.54. Same item type, opposite behaviour. Price level, not "cases", is what separated them. The CS2 case prices guide explains why sub-dollar cases behave this way. Its live ladder currently shows the Revolution Case at $0.14 and the Chroma 3 Case at $3.00, as of the 13 October 2026 refresh.

Pistols surprised me. The Glock-18 Water Elemental in Factory New fell from $51.99 to $40.20 in thirty days, a 22.7% drop and the steepest of any gun I track. It was $61.00 on 31 July. The Desert Eagle Printstream (Factory New) was the other large pistol move, $72.37 to $61.00 on CSFloat (−15.7%). With only eight pistols in the set, two items like these are enough to pull the category median down, so I treat "pistols fell 11.5%" as "two pistols fell hard and the rest drifted".

Knives fell at about the market's pace, with no sign of catching up. The Karambit Fade (Factory New) went from $1,773.91 to $1,560.34 on CSFloat, −12.0% in the month. The AWP Dragon Lore (Factory New) slid from $11,022.55 to $10,385, −5.8%, a smaller fall in percentage terms, consistent with the price-level pattern above. Rifles fell least as a category, though the FT AK-47 Redline (my usual barometer, more below) still lost 5.7%.

Mostly not, and this is the section I would most like other trackers to copy. If you rank items by "best price anywhere", two items look like October's big winners: the Karambit Fade in Minimal Wear at +32.6% and the USP-S Cortex (Field-Tested) at +33.6%. Both rises come from a change in which marketplaces were listing, not from anyone paying more.

• Karambit Fade (Minimal Wear). On 13 September the cheapest listing was on CSFloat at $1,900. On 13 October my feed returned no CSFloat listing for it at all, only market.csgo at $2,520. On 13 September market.csgo had it at $2,800. So the one venue I can compare across both dates got cheaper by 10%, and the apparent jump is the floor venue disappearing.
• USP-S Cortex (Field-Tested). On 13 September the floor was a $2.35 Skinport listing. Skinport was no longer in my feed on 13 October. CSFloat, present both times, moved from $3.36 to $3.14, a 6.5% fall.

The genuine risers on a like-for-like CSFloat basis were two, and both were small: the M4A4 Asiimov (Well-Worn), $116.00 to $120.68 (+4.0%), and the Souvenir AWP Dragon Lore (Field-Tested), $5,799.93 to $5,951.63 (+2.6%). Two risers out of 55 is the honest count. A headline built on "best price anywhere" would have shown ten risers and a median of only −4.7% for the same thirty days, a gentler story than the data supports.

The general lesson: a "lowest price across markets" series changes every time the list of markets changes. If one cheap venue drops out, every item it used to be cheapest on seems to rise overnight. If you keep your own CS2 skin price tracker, store each venue's price separately and compare like with like, or your biggest "winners" will be bookkeeping. I go deeper on reading series like this in CS2 skin price history.

Two of the five entries I can grade with my own data; two I cannot, and I would rather say so than fill the gap with guesses. Each row below repeats what I said I would watch in August and what my data shows now.

The knife row is the one that changed my mind. In August I described knives as a laggard that might catch up once the spread between venues narrowed. Instead the spread widened: sellers on the dearer venue kept their asking prices while the CSFloat floor fell. That is not what a category about to rally looks like. The spread figures use only the two venues present at all three dates, and the 13 October set has one fewer knife because CSFloat returned no listing for the Karambit Fade in Minimal Wear.

The Redline row is the one I am least comfortable with. I picked it because I thought volume would tell me more than price, and my feed cannot see volume. Next time I will list watch items I can actually measure.

Nothing was recorded, and here is why. This site refreshes its prices automatically every three days. From 22 September to 10 October every scheduled refresh was rejected by the data-quality gate I built into it: the price API returned market values that failed validation, so the script kept the previous data and exited with an error rather than publishing bad numbers. The 13 October run was the first one to pass. That leaves a thirty-day hole between the 13 September and 13 October snapshots, and no day-by-day path inside it.

Two consequences matter for reading this report:

1. I can tell you where prices ended, not how they got there. The −7.4% month could have been a steady slide or a sharp drop on one day. I do not know, and I am not going to infer it.
2. The list of venues changed during the gap. On 13 September my feed returned listings from CSFloat, market.csgo, CS.MONEY and Skinport. On 13 October it returned CSFloat, CS.MONEY and market.csgo, with no Skinport. The lowest listing came from CSFloat for 48 of 56 items, CS.MONEY for 6 and market.csgo for 2, against 28, 7 and 1 on 13 September plus 20 from Skinport.

That venue change is why every headline figure here uses CSFloat alone: it is the one venue present at all four dates (31 July, 22 August, 13 September, 13 October), covering 55 of my 56 items. The 56th, the Karambit Fade in Minimal Wear, had no CSFloat listing on 13 October. The four CSFloat snapshots are saved as a pinned file in the repository, taken from the price-history commits, so the charts on this page are rebuilt from fixed numbers and cannot drift when the live feed changes. The live prices elsewhere on the site come from the latest refresh (13 October 2026).

A lowest listing is an asking price, not a completed sale. It tells you what you could have bought at, at that moment, on that venue. It does not tell you how many sold, and with only three venues it says nothing about Buff163 or the Steam Community Market, which my feed does not cover at all.

A broad, even decline like this one tells you something about the market's direction and nothing about any single skin. Here is how I am using it, and how I would suggest you do.

If you are buying to use the skin, a falling market is simply good news. The FT AK-47 Redline is cheaper than in July; at the latest refresh its lowest listing was $24.80 (13 October 2026). Check the float before you pay (how to check a CS2 float), and buy on the venue that is cheapest that day. On 13 October the median gap between the cheapest and dearest venue for the same item was 13.9% across my 55 multi-venue items, so the price checker is worth a look before you pay.

If you are holding skins as a store of value, the like-for-like numbers are the ones to remember: −11.9% for the median item in 74 days, and losses in every category. Items above $1,000 lost least, but "lost least" still means lost. Nothing in my data shows a floor forming yet. October's thirty-day fall was larger than August's.

If you sell regularly, watch the venue spreads rather than headline prices. When the dearer venue keeps its asking prices while the cheapest one falls, as with knives this autumn, listings priced off the dearer venue sit unsold. My guide to selling CS2 skins covers fee maths by venue.

What I will not do is call a bottom or a rebound from four snapshots. The data says the market fell, broadly, and cheap items fell most. It does not say what happens next, and nobody's spreadsheet does. The investing hub collects the methods I use to keep my own tracking honest, and the knives hub has per-knife pages if the spread finding matters to something you own.

One rule carries forward to the next report: if the refresh fails again, I will publish the gap dates alongside the numbers, as I have done here, rather than draw a straight line across a month of missing data.