Which CS2 case has the best ROI to open in 2026?

No active case has positive expected value to open as of 2026 — every active case returns less than the cost of the case + key on average. The closest-to-break-even cases are the older discontinued ones with high knife floors (Operation Bravo, Operation Wildfire, Operation Vanguard) because their knives clear $1,000+. But those cases themselves trade at $100-500, so the trade is "hold and keep" rather than "open."

Which CS2 cases are the best stores of value?

Discontinued cases with low circulating supply. The strongest holds historically have been Operation Bravo, Operation Vanguard, Operation Wildfire, Winter Offensive Weapon Case, eSports 2014 Summer, and Operation Phoenix — all under a few hundred thousand units circulating. Supply only goes down (unboxing destroys the case to produce a skin), so price tends to drift up over years assuming demand holds.

Did the Armory update change CS2 case ROI math?

Yes. Post-Armory cases use collection-weighted probability — within a knife drop, each case's rare-special pool has its own internal weighting rather than a flat equal-split across all knives. This means cheaper-knife cases converge faster to a lower expected return than the legacy flat model predicted, and the math is now per-case rather than per-tier.

How do you actually calculate case EV?

For each drop in the case, multiply its drop probability by its post-fee Steam Market price, then sum. Subtract the cost of (case + key + Steam fees on the sell-side). The standard CS:GO/CS2 drop probabilities are 79.92% mil-spec, 15.98% restricted, 3.2% classified, 0.64% covert, 0.26% knife — but within each tier, individual skin probability + StatTrak™ multipliers split that share further. See our /tools/csroi/ recommendation for a live calculator.

What's the source for these supply numbers?

Supply estimates come from CS:STONKS, refreshed 2026-05-18. CS:STONKS reconstructs circulating supply from Steam Market listings + estimated unbox rate; the numbers are best-effort estimates, not Valve-published figures (Valve doesn't publish per-case supply). Marked "incomplete data" where the sample is too thin for confident estimates.

Best CS2 Cases to Open in 2026 (ROI Math)

The uncomfortable answer first: opening cases is a losing bet on average. Every active CS2 case returns less than (case + key) cost over enough openings — typically 30 to 55 cents on the dollar. The interesting question is which cases store value if you hold them: low-supply discontinued cases have appreciated steadily for years, while the latest drops trend down for months after release. This guide covers the math, the supply data, and what to actually do with the cases sitting in your inventory.

The supply table at a glance

Tracking 43 CS2 weapon cases with circulating supply estimates from CS:STONKS, updated 2026-05-18. Across the tracked set, roughly 1467.6M cases remain in circulation and roughly 2278.7M cases have been opened (and destroyed) over the franchise’s history.

Supply matters because unboxing is destructive: when someone opens a case to roll for a skin, the case is consumed. Combined with the fact that no new copies of discontinued cases drop in match-end rewards, supply only moves in one direction — down. That’s the entire mechanic behind "store of value" for low-supply cases.

Lowest-supply cases (best stores of value)

Sorted by current circulating supply ascending. These are the cases that have historically appreciated the most year-over-year:

Winter Offensive Weapon Case — about 323K circulating, ~14.7M already unboxed · Base Grade
CS:GO Weapon Case 2 — about 1.8M circulating, ~4.5M already unboxed · Base Grade
Chroma Case — about 2.0M circulating, ~32.9M already unboxed · Base Grade
Huntsman Weapon Case — about 3.2M circulating, ~24.8M already unboxed · Base Grade
Operation Phoenix Weapon Case — about 3.8M circulating, ~60.3M already unboxed · Base Grade
Falchion Case — about 4.3M circulating, ~54.4M already unboxed · Base Grade
Horizon Case — about 11.9M circulating, ~44.1M already unboxed · Base Grade
Operation Breakout Weapon Case — about 14.4M circulating, ~66.3M already unboxed · Base Grade
Operation Vanguard Weapon Case — about 16.0M circulating, ~20.8M already unboxed · Base Grade
Shadow Case — about 16.3M circulating, ~39.1M already unboxed · Base Grade
Glove Case — about 18.3M circulating, ~60.1M already unboxed · Base Grade
Spectrum Case — about 18.8M circulating, ~57.4M already unboxed · Base Grade

Full sortable list, with images, drop totals, and unbox counts, lives at /cases/ — sort by remaining, dropped, or unboxed; click any case to jump to its market page.

The math: why open EV is negative

The standard CS2 drop probabilities haven’t changed:

79.92% Mil-Spec (blue) skin
15.98% Restricted (purple)
3.20% Classified (pink)
0.64% Covert (red)
0.26% Rare Special (knife or — post-2024 — gloves on some cases)

On top of these tier probabilities, every drop has a flat ~10% chance of being StatTrak™. The Armory update (post-knife change) replaced the legacy flat per-knife distribution with a per-case collection-weighted model: cases whose knife collection is cheaper (say, mostly Bayonets and Flips) return less on the rare-special hit than cases with expensive collections (Karambits, Butterflies, the modern Skeleton-tier knives).

Plug in the math and almost every active case sits between -30% and -55% ROI per opening. For the live per-case calculation we recommend CSROI; for the full post-Armory model walkthrough see The Complete Guide to CS2 Case ROI in 2026.

So which cases should you actually buy / hold?

Three honest strategies, ordered from lowest-risk to most-speculative:

Hold discontinued, low-supply cases. The classics — Operation Bravo, Vanguard, Wildfire, Phoenix, Winter Offensive, eSports 2014 Summer. They’ve historically tracked +5 to +20% per year against USD. Boring, slow, and the safest bet for "skin savings account."
Buy new cases on release week and hold 18 months. New cases trade at a premium during the week of release (high opening volume), then plummet over the next 60 days as supply floods in. Buy after the unbox cliff (typically week 8 onwards) and hold; the drop rate reduces materially after 12-18 months and price recovers. Higher variance than holding classics.
Open cases for entertainment, not profit. If you enjoy unboxing as a game in itself, budget for it the way you’d budget for lottery tickets — small known outlay, asymmetric upside. The 0.26% knife shot is fun because it’s rare. Just don’t convince yourself you’re investing.

The trap: high-volume new cases

A case in active drop rotation is the worst long-term hold, because supply keeps growing. The Kilowatt case dropped tens of millions of copies in its first year; the Gallery, Limited Edition, Fracture, and Recoil cases all sit at 10M+ circulating supply. Even when they later go discontinued, the absolute floor is set by the supply ceiling. Compare to Operation Bravo, which has fewer than a million ever printed and trades around a hundred dollars per case.

The supply column in the /cases/ table is the single most important sort key for this — anything above 5M circulating is a high-volume case and should be treated as a flow item, not a store of value.

The Steam fee gotcha

Steam Market charges a 13% combined fee on every sale (10% Steam + 5% CS2 publisher fee, applied to the buyer-paid price). When you do open a case and pull a Mil-Spec, the "$8 sale" returns about $6.96 to your Steam Wallet. For ROI math always use the post-fee number — that’s the actual amount you can spend on the next case. CSROI builds this in by default.

About the data

Supply estimates come from CS:STONKS, last refreshed 2026-05-18. Some older cases are marked as having incomplete data — usually the cases too old to reconstruct circulating supply confidently from Steam Market listings. We re-import on each data refresh; the /cases/ page shows the live numbers along with the source link.