The First Independent CS2 Case Opening ROI Study

Across all 8,931 cases, the average ROI comes out to 89.80%. In other words, the average case returns roughly 90 cents for every dollar spent, and the platform keeps the rest.

And this isn’t a case of a few weird outliers distorting the picture.

The median ROI is 90.00%, and the standard deviation is 2.63, which means most cases sit fairly close to that 90% center.

Funny enough, that awful return is actually not far off from the real ROI of opening an official Clutch case yourself.

That leaves an 80-percentage-point spread between the best and worst public cases available right now.

Out of all the cases analyzed, not one breaks even. None even crossed the 98% ROI mark.

This shouldn't be surprising to anyone who understands the nature of the business, but it's worth stating plainly because the question comes up constantly: no, there is no "hidden gem" case that actually pays out more than it costs. Every case on every platform is mathematically designed so the house wins over time.

The only question is how much:

Out of nearly 9,000 cases, just 326 cases (3.7%) have an ROI above 95%. That’s the near-fair tier, where the house edge drops below 5%.

On the other hand, 14 cases in the dataset take more than 20% of every dollar spent.

The distribution looks like this:

This was one of the strongest findings in the dataset.

Out of 8,931 cases, only 326 (just 3.65%) have a very good ROI: above 95%.

Here’s where those near-fair cases are located:

A lot of newer players assume expensive cases should offer better value.

The logic sounds reasonable: if you’re paying more, the site should give you better odds to justify it, but the data says otherwise:

The correlation between price and ROI is basically zero. If anything, the most expensive cases are slightly worse on average than the cheapest ones.

What price changes is not value. It changes volatility and how the session feels in the short run.

Cheap cases often rely on one or two jackpot items carrying a big part of the theoretical return. On paper the odds may look fine, but in practice you might need hundreds or thousands of openings before that jackpot lands. Most players never get close or have the will to even try.

On the other hand, expensive cases usually feel smoother. Wins are spread out more evenly, so sessions look less brutal and it’s easier to walk away thinking the value was decent over just a few sessions.

But that smoother experience can be misleading. The math underneath usually isn’t better, and sometimes it’s slightly worse if you play more.

Across the full dataset, frozen sites carry a 1.18 percentage point lower average house edge than dynamic ones.

That doesn’t necessarily mean frozen sites are more generous. It means they are more structurally predictable.

When a site locks its loot table, the margin is baked in. You can calculate the exact EV and know it will stay there until the site changes something.

Dynamic sites create more room for margin drift because prices are constantly moving and most players are not recalculating odds every time they open a case.

That does not automatically make dynamic pricing shady or scammy. Some dynamic platforms, with CSGORoll being the best example, keep the movement relatively tight. But as a general rule, frozen tables are easier to verify and more player-friendly over time.

First, check the individual case, not just the platform average.

The difference between a 4% edge case and a 14% edge case on the same site is often bigger than the difference between entire platforms.

Second, don’t get hypnotized by case price. Expensive does not mean better value.

Third, use the data before opening anything on any site. For the first time in the case opening space, every publicly available case from every site included in this study can be tested directly in the CaseLab simulator.

You can inspect the full loot table, item values, odds, and session stats, then simulate thousands of openings in seconds with the CaseLab simulator. In practice, it works like a case opening calculator built for real long-run analysis.

This study is not an endorsement of any platform listed, and it is not a recommendation to gamble on any of them.

Case opening is gambling and carries risk by default.

The purpose of this page is not to tell people where to deposit. It is to give the CS2 gambling community something that did not really exist before: a single independent view of what case opening actually costs across the industry.

Raw house edge is only one part of the picture. Transparency, provably fair implementation, community feedback, payout issues, public warning signs, and similar factors matter too. The goal of this page is to encourage more caution, more independent research, and better-informed decisions in a space where transparency is often weak.

A site can have decent odds and still raise serious concerns elsewhere. Always look deeper before putting money into any gambling or other money-related activity.

This is a data study, nothing more and nothing less.

RTP, ROI, and EV are often used almost interchangeably in case opening, but the main point is simple: they reflect the math over time, not what happens in a few opens.