November’s useful finding is the data gap
We captured the existing CSFloat cache on November 12, 2026 and calculated returns for 42 standard weapon cases. All 42 had more than 95% of outcome probability priced; only three had every modeled outcome priced. Those three also passed the two-hour source-freshness check at capture: Snakebite, Recoil and Revolution. Their modeled returns were 40.66%, 30.01% and 26.67%, respectively. The remaining 39 cannot be compared as complete results, so this dataset cannot support a definitive best-case claim across the entire catalog.
This is original analysis of cached data, not an audit of paid openings and not proof of actual resale proceeds. Some source quotes date back months. Rather than hide those gaps, we publish the full snapshot with source dates, coverage and a downloadable dataset.
Dated snapshot, not live prices or a profit recommendation
Captured 2026-11-12 11:03:17 UTC from cached CSFloat listing asks via OpenSkin. Source dates can be months older. Only 3 cases meet both complete-price and two-hour freshness checks. Missing quotes make RTP a lower bound and expected loss an upper bound; provisional rows cannot establish the best case.
| Case | Opening | EV / known value | RTP | Expected loss | Priced odds | Oldest quote |
|---|---|---|---|---|---|---|
| Snakebite CaseComplete & recent | $2.75 | $1.12 | 40.66% | $1.63 | 100.0000% | 2026-11-12 |
| Recoil CaseComplete & recent | $2.68 | $0.80 | 30.01% | $1.88 | 100.0000% | 2026-11-12 |
| Revolution CaseComplete & recent | $2.63 | $0.70 | 26.67% | $1.93 | 100.0000% | 2026-11-12 |
Why almost-complete prices can still mislead
A missing expensive knife can carry a meaningful share of expected value despite representing very little probability. For incomplete cases, the table shows the sum of known prices, not a complete EV. “At least” return and “at most” loss describe bounds on that model, not forecasts of what a player earns.
RTP is not profit
Return divides probability-weighted item value by the cost of both case and key. A 60% return is a −40% net ROI before fees. A $2.49 key is the report’s default assumption; case costs come from the same marketplace. Listing asks can exceed obtainable sale proceeds, and missing or old quotes add uncertainty.
No month-on-month trend without history
We have no prior comparable monthly sample, so we do not claim returns rose or fell in November. A future report needs another dated capture under the same rules. A cache retrieval date is not the date of the underlying market quote.
Use the data without chasing
Read the full case-return table and methodology, check the published rarity odds, or use the free simulator without spending money. If you want a particular skin, comparing its purchase price is more useful than chasing a rare outcome.
Sources: our dated OpenSkin/CSFloat cache capture, ByMykel’s community case metadata and the disclosed calculator model. No revenue estimate, paid-session claim or current-market recommendation is inferred from this snapshot.
