CS2 Market Notes: The Bounce Stayed Selective

This week was the first recent check that did not read like another clean step lower.

Across 34,202 comparable references after excluding Sticker Slabs and Cologne Ranked stickers, the median move from August 28 to September 4 was 0.00%. That does not mean every item was flat. It means the middle stopped falling enough for the weekly median to land on the line.

Breadth improved too. Using the same plus-or-minus 0.25% noise band, 14,558 references moved higher, 11,162 moved lower, and 8,482 stayed roughly flat. That is a better balance than the prior two notes, where fallers kept doing most of the work.

The catch is that the bounce was selective. StatTrak weapons and the four Season 5 red weapons had a strong week. Cologne Foils were still weak, lenticular stickers were mixed, and several broad categories looked flat rather than clearly recovered.

This is not financial advice. CSBeacon is not a marketplace, broker, or trading advisor. These notes are meant to help CS2 inventory owners understand what changed and how much confidence the available data deserves.

The middle finally stabilized

The cleanest shift this week is in breadth. More references rose than fell, and the broad median stopped printing another negative weekly move.

That matters because August had become a grind: small declines in the middle, bigger weakness in certain premium and event-linked pockets, and enough red breadth to keep the market feeling heavy. September opened with a calmer middle.

Among items with positive comparable aggregate references on August 28 and September 4:

The table is not a victory lap. Knives, gloves, containers, charms, souvenirs, and broad stickers all landed at a flat median. That is stabilization, not strength.

The stronger read came from StatTrak weapons and ordinary weapons. StatTrak had the cleanest category-level move, with a +4.60% median and more than twice as many higher references as lower ones. General weapons were less dramatic, but the +0.68% median still broke from the softer August pattern.

For inventory owners, that distinction matters. A week can look better overall because the high-volume middle stops leaking, even while specific collections or finishes are still under pressure.

Armory reds had a real bounce

The four Season 5 red weapon finishes were the clearest item group improvement in this snapshot.

Across AK-47 AUTOEXEC, AWP Sovereign Flame, Glock-18 Ghost Protocol, and AK-47 Consequence of the Jinn, 15 of the 20 standard wear references moved higher from August 28 to September 4. The median move was +6.56%.

Glock-18 Ghost Protocol led the group. Factory New rose from about $472.50 to $560.24, roughly +18.6%. Minimal Wear, Field-Tested, and Well-Worn were also up double digits. AK-47 Consequence of the Jinn Battle-Scarred rose from about $53.53 to $69.72, roughly +30.2%.

The weaker side was concentrated in AWP Sovereign Flame. Factory New fell from about $469.00 to $440.86, roughly -6.0%, and Battle-Scarred fell about -5.5%. That keeps the group from being a simple all-clear signal.

Cologne stayed split by finish

Original Cologne 2026 stickers stopped falling at the broad median, but the finish split still matters.

The original Cologne set had 792 comparable references this week: 193 Foils, 193 Holos, 193 Golds, and 213 remaining originals. Its median move was 0.00%. Underneath that flat headline, Foils were still weak at about -5.88%, with 115 lower and 43 higher. Holos were slightly positive at about +0.39%, while Golds were stronger at about +1.50%.

That is a more useful read than saying Cologne was flat. The category stabilized because Holos, Golds, and the remaining originals offset Foil weakness. If you hold or track Cologne stickers, the finish matters more than the event label.

Ranked Cologne remained too young and internally noisy to blend into that read. Across 571 comparable Ranked references, the median was also 0.00%, but 277 sat inside the noise band and the finish groups were mostly flat by median.

Valve has already said the Cologne shop remains available through September 29. That still gives the market time to digest supply, especially for event items that are tied directly to player and team demand.

Lenticulars were still uneven

Season 5 lenticular stickers did not join the Armory weapon bounce cleanly.

Across 25 comparable lenticular references, the median move was about -1.40%. Fifteen moved lower, nine moved higher, and one was flat. That is not a collapse, but it is still negative breadth.

The split was wide. Partyboy rose from about $2.27 to $2.49, roughly +9.7%, and Pepper Gauge rose about +8.7%. Digital Dashboard also rose from about $9.97 to $10.30, keeping its lead inside the group. On the other side, BullHit fell about -13.0%, Electric Avenue fell about -12.4%, and Econometer fell about -7.3%.

The practical lesson is the same one this category keeps giving: newer sticker groups can look stable at the collection level while individual references move in opposite directions. A single median is useful, but it is not enough.

Public indexes told a mixed story

Outside CSBeacon's item-reference snapshot, the public backdrop also looked mixed rather than decisively bullish.

Steam Charts showed a last-30-days average near 817,000 players, with its displayed last-30-days change at -0.95%, and a 24-hour peak above 1.18 million when checked. That is a softer player-count backdrop than earlier in the year, but still a very large active game.

The Steam Dashboard (steamdb.com) CS2 skin statistics page labeled its broader category index as a 29-day window ending September 4 and showed it down 4.1%, with gloves, weapon skins, knives, and stickers all negative at the median. CS2.SH's All Skins Index, using its own basket and weighting, showed +0.2% over seven days but -2.2% over 30 days as of September 3.

Those numbers can all be true at the same time because they use different baskets, windows, weighting, and flat bands; Steam Dashboard uses a plus-or-minus 1% flat band, while CSBeacon uses plus-or-minus 0.25% for this weekly breadth read. For the evergreen version of that problem, see Why CS2 Skin Price Trackers Disagree. The short version for this week: short-window stabilization does not erase the weaker 30-day backdrop.

Valve update context did not add a new item-market shock this week. The market still appears to be processing the larger summer themes: Cologne supply, Ranked sticker repricing, Season 5 items, and a softer broad backdrop.

Closing read

This was the best-looking weekly snapshot in a few checks, but it was not a clean green market.

The broad median stopped falling. Higher references outnumbered lower references. StatTrak weapons had a strong median week, and Armory reds finally produced a convincing bounce after several messy snapshots.

At the same time, Cologne Foils kept sliding, lenticular stickers were still negative at the median, and several major categories were merely flat. The market improved by becoming less synchronized to the downside.

That is still useful. For a CS2 inventory owner, the question this week is not whether everything recovered. It is whether your own inventory sits in the groups that bounced, the groups that flattened, or the groups still leaking lower.

Source and data notes

Market movement compares CSBeacon's stored global pricing snapshots from August 28 and September 4, 2026. Broad category medians include only items with positive comparable aggregate references on both dates. Moves within plus or minus 0.25% are treated as roughly flat for breadth counts.

Beginning with this note, comparisons use CSBeacon's current market-listing model for both endpoints. Prices quoted for August 28 are therefore on that basis and will differ from overlapping prices in the August 28 note, which used the prior valuation model. Week-over-week moves shown here are like-for-like.

Sticker Slabs were excluded from all sticker calculations and highlighted movement because they are separate, thinly listed items that can distort sticker comparisons. Cologne Ranked stickers were excluded from broad direction claims because their current references remain young and internally noisy enough to deserve separate treatment.

Steam Charts, Steam Dashboard (steamdb.com) market statistics, CS2.SH index data, and Valve announcements were reviewed on September 4, 2026 for public market and game context. Steam Dashboard and CS2.SH display their own snapshot or as-of dates, which do not have to match CSBeacon's Friday-to-Friday comparison window. Their figures are context, not proof of causation for weekly item-reference movement.

Prices are aggregate market-listing references processed by CSBeacon, not confirmed sale prices or guaranteed cash values. Thin listings, unusual variants, naming aliases, limited liquidity, and changing source coverage can create extreme apparent movement.