Last week, the slide got broader again. This week, it got quieter.
Across 33,484 comparable references after excluding Sticker Slabs and Cologne Ranked stickers, the median move from August 21 to August 28 was about -0.35%. That is still red, but it is a smaller decline than last week's -0.82% clean median.
Breadth stayed negative. Using the same plus-or-minus 0.25% noise band, 17,023 references moved lower, 11,125 moved higher, and 5,336 stayed roughly flat. The middle stopped falling as hard, but the market did not suddenly turn green.
The cleaner category read was also less one-note than the prior two weeks. Charms were weakest at the median. StatTrak weapons, knives, souvenirs, gloves, containers, and original Cologne stickers leaned lower. Ordinary stickers and general weapons were roughly flat at the median, even though their internal breadth still leaned negative.
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 broad slide slowed
A smaller median decline matters because the last few Friday checks were moving in the wrong direction. August 14 was almost flat at the middle, August 21 weakened again, and August 28 sits between those two snapshots.
Among items with positive comparable aggregate references on August 21 and August 28:
| Category | Comparable refs | Median move | Lower | Higher |
|---|---|---|---|---|
| Charms | 1,005 | -2.01% | 594 | 270 |
| StatTrak weapons | 3,373 | -1.13% | 1,933 | 1,127 |
| Knives | 5,364 | -0.94% | 3,180 | 1,613 |
| Souvenirs | 3,956 | -0.58% | 2,110 | 1,152 |
| Gloves | 410 | -0.52% | 218 | 158 |
| Containers | 471 | -0.38% | 245 | 165 |
| Stickers, excluding Ranked and Slabs | 9,849 | 0.00% | 4,793 | 3,189 |
| Weapons and items | 8,933 | 0.00% | 3,867 | 3,421 |
1,005
-2.01%
594
270
3,373
-1.13%
1,933
1,127
5,364
-0.94%
3,180
1,613
3,956
-0.58%
2,110
1,152
410
-0.52%
218
158
471
-0.38%
245
165
9,849
0.00%
4,793
3,189
8,933
0.00%
3,867
3,421
Charms are the easiest weak spot to name this week, but they are also a newer and thinner category than knives or gloves. That makes the direction worth noting without pretending every charm reference has mature liquidity behind it.
The higher-confidence premium read is more moderate: knives, StatTrak weapons, gloves, and souvenirs were all lower at the median, but none repeated last week's glove-sized drop. Gloves in particular cooled from a clear bearish signal to a small median decline with more mixed breadth.
That is why this reads as calmer, not bullish. A flat median in general weapons does not erase negative breadth in premium categories, and a smaller weekly decline does not prove buyers are back in control.
Armory reds stopped being one-way
The four Season 5 red weapon finishes no longer look like a simple launch-premium unwind every week.
Across AK-47 AUTOEXEC, AWP Sovereign Flame, Glock-18 Ghost Protocol, and AK-47 Consequence of the Jinn, 11 of the 20 standard wear references moved lower, five moved higher, and four stayed inside the noise band. The median move rounded to -0.35% on the rechecked comparison.
That is a big change from August 21, when 19 of 20 standard wear references declined. This week still had red, but it was no longer the whole story.
- AWP | Sovereign Flame Factory New rose from about $466.03 to $518.29, roughly +11%.
- AK-47 | AUTOEXEC Minimal Wear rose from about $286.89 to $307.70, roughly +7%.
- Glock-18 | Ghost Protocol Factory New fell from about $567.98 to $538.45, roughly -5%.
- AK-47 | Consequence of the Jinn Well-Worn fell from about $85.96 to $81.73, roughly -5%.
The practical read is that the easy launch-premium story is getting less clean. Some high-end wears bounced while cheaper wears kept slipping. That can happen when supply is still settling, trade holds are still working through the market, and buyers are choosier about which exact wear they want.
For inventory owners, this is the moment where item-level tracking matters more than collection-level vibes. A user holding Sovereign Flame Factory New had a very different week from someone holding Ghost Protocol Factory New, even though both belong to the same Season 5 red bucket.
Cologne originals kept leaking lower
Original Cologne 2026 stickers were still negative, just less violently than the messier Ranked Series.
Excluding Ranked stickers and Sticker Slabs, original Cologne stickers had 792 comparable references and a median move of about -1.15%. The subgroup split was 193 Foils, 193 Holos, 193 Golds, and 213 remaining original references. Foils were the weakest subgroup at about -3.70%, Holos fell about -2.35%, Golds fell about -1.10%, and the remaining originals were flat at the median.
That is a normal weak-market signal: not every sticker down, not every finish equally hit, but enough median pressure to matter.
Ranked Cologne remains harder to use as a clean signal. The comparable Ranked set had a median move of about -5.00% across 561 entries: 193 Foils, 192 Holos, and 176 Golds. Foils and Holos were both around -10% at the median while Golds were roughly flat, but the raw individual moves still include too many extreme jumps and reversals to treat the category like mature price discovery.
Valve's own framing still matters here: Ranked stickers were introduced as a lower-priced companion to the original Event Series, and the Cologne shop remains available through September 29. That gives buyers more time and more supply paths than a normal closed event would have.
The useful conclusion is narrow: original Cologne stickers are still drifting lower, especially Foils and Holos. Ranked stickers are still young enough that their weekly medians deserve a warning label.
Lenticulars split instead of sliding together
The Season 5 lenticular sticker group no longer moved as one block either.
Partyboy was the weakest tracked lenticular this week, falling from about $3.53 to $2.83, roughly -20%. Poo Cocktail Supreme, Pineapple On Pizza, Pepper Gauge, Burnout, and Digital Dashboard all moved lower too, but their declines were smaller.
NITRO! rose from about $4.96 to $5.19, roughly +5%, and Electric Avenue rose from about $2.56 to $2.60. That is not enough to call the group strong, but it does break the clean all-red pattern from earlier checks.
Digital Dashboard remains the leader in this small group at about $10.51. The interesting part is that its lead has held while its price has not: it is still clearly above the others, but it has drifted lower for several Friday checks after the July pop.
The public backdrop was mixed
The market data was redder than the broad public context this week.
Steam Charts showed a 30-day average of about 820,000 players when this post was verified, down about 3.89% from July. Concurrents at the time of writing were still above one million, against a 24-hour peak near 1.25 million, so this is weakness inside a very large game rather than evidence that Counter-Strike vanished overnight.
CSMarketCap showed an estimated CS2 skin-market value near $6.26 billion, about 28.2% below its modeled all-time high. Its displayed seven-day move was about +1.8%, while the 30-day move was about -7.4%; that window does not need to match CSBeacon's Friday-to-Friday snapshot exactly. For the evergreen version of this problem, see Why CS2 Skin Price Trackers Disagree. Aggregate public indexes and stored item-reference comparisons can disagree over short windows.
Valve's August 24 update was not an economy shock. It focused on map scripting and related tooling, while the August 19 update was mostly maps and a gameplay fix. So this week's price movement does not need a new Valve-market event to explain it. Sometimes the story is simply a market digesting the last several changes.
For the longer player-count argument, see Is CS2 Dying?. The short version for these notes is the same as last week: player count is context, not a complete explanation for every item move.
Closing read
This was a calmer week, not a clean turn.
The clean comparable set fell about 0.35% at the median, less than August 21's 0.82% decline. Fallers still outnumbered risers, and the weaker categories still included inventory-relevant areas like charms, StatTrak weapons, knives, souvenirs, gloves, and original Cologne stickers.
The biggest change was shape. Armory reds stopped being almost universally red. Gloves cooled from the clearest bearish signal to a smaller decline. Original Cologne kept leaking lower while Ranked Cologne remained unstable. Lenticulars split instead of sliding together.
If August 21 said pressure was broad again, August 28 says the pressure is still there but less synchronized. That is a better market than last week. It is not yet a clean recovery.
Source and data notes
Market movement compares CSBeacon's stored global pricing snapshots from August 21 and August 28, 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.
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.
Valve's Cologne Ranked Series announcement is the primary source for the Ranked Series and its lower fixed-price structure. Valve's Season 5 announcement confirms the Armory weapon and sticker collections discussed here. Valve's August 24 update was checked for current-week game context and did not introduce a new market-facing item change.
Steam Charts player-count data and CSMarketCap's market-cap estimate were checked as public context on August 28, 2026. CSMarketCap displayed an estimated market cap near $6.26 billion, 28.2% below its modeled all-time high, with a seven-day move of about +1.8%. Those public datasets are context, not proof of causation for the weekly price 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.