Expected value (EV) is the average amount a case returns per opening, over many openings. You calculate it by multiplying each item's value by its chance of dropping, then adding the results. If the EV is lower than the case price, the difference is what you lose on average every time you open it.
That one number cuts through the marketing around a case. The big knife image on the case thumbnail does not tell you what it pays; the odds table does.
The formula
For every item in the case:
- Take the item's value in dollars.
- Multiply it by its drop chance, written as a decimal (2% = 0.02).
- Add up the results for all items.
EV = (value₁ × chance₁) + (value₂ × chance₂) + … + (valueₙ × chanceₙ)
Then compare it to the price:
- Return to player (RTP) = EV ÷ case price
- House edge = 1 − RTP
- Average loss per opening = case price − EV
All drop chances must add up to 100%. If they don't, the odds table is incomplete and the result is meaningless.
A worked example
Here is a fictional $10.00 case with six items. The numbers are invented for the example; the method is the same for any real case.
| Item | Value | Drop chance | Value × chance |
|---|---|---|---|
| Item A | $250.00 | 0.50% | $1.25 |
| Item B | $60.00 | 2.00% | $1.20 |
| Item C | $20.00 | 7.50% | $1.50 |
| Item D | $9.00 | 20.00% | $1.80 |
| Item E | $4.00 | 30.00% | $1.20 |
| Item F | $1.40 | 40.00% | $0.56 |
| Total | 100.00% | $7.51 |
The case has an EV of $7.51 and costs $10.00:
- RTP = 7.51 ÷ 10.00 = 75.1%
- House edge = 24.9%
- Average loss = $2.49 per opening
Over 100 openings ($1,000 spent), you should expect to end up with roughly $751 in item value. That is an average, not a promise: your real result could be much lower or much higher.
What the average hides
Look at the same case from the point of view of a single opening:
| Outcome | Chance | Result vs $10.00 price |
|---|---|---|
| Item A or B (profit) | 2.5% | +$50.00 to +$240.00 |
| Item C (profit) | 7.5% | +$10.00 |
| Item D (small loss) | 20% | −$1.00 |
| Item E or F (big loss) | 70% | −$6.00 to −$8.60 |
Seven openings in ten lose more than half the case price. The EV of $7.51 is pulled up by rare items most people never hit. This is why expected value and "what usually happens" are very different numbers, and why chasing the rare item gets expensive.
How rare is "rare"? The chance of hitting Item A at least once:
| Openings | Spent | Chance of at least one Item A |
|---|---|---|
| 10 | $100 | 4.9% |
| 50 | $500 | 22.2% |
| 100 | $1,000 | 39.4% |
| 200 | $2,000 | 63.3% |
Formula: 1 − (1 − 0.005)ⁿ. Even after $2,000 of openings, more than a third of players have still not seen it.
Three mistakes that make EV numbers wrong
1. Using the wrong item values
The EV is only as good as the values you plug in. A site's own listed value may be higher than what you could actually sell the item for. For a realistic number, use what the item sells for on a market you would really use, after fees. If selling costs 10–15% in fees, your real EV drops by about the same amount.
2. Ignoring withdrawal conditions
An item you cannot withdraw is not worth its listed value to you. Minimum withdrawal amounts, wagering requirements and trade restrictions all reduce the real return, sometimes to zero for small wins. See what "free" really means on free cases for how conditions change the math.
3. Thinking a streak changes the odds
If each opening is independent, ten misses in a row do not make the next opening more likely to hit. The EV per opening stays the same no matter what happened before. A site that is provably fair lets you check that results were fixed in advance, but it does not make the odds any better.
How to use EV in practice
- Comparing two cases: a higher RTP means a slower average loss. It does not mean the case is a good bet.
- Checking a site: if the full odds table is missing, or the chances don't add up to 100%, you can't calculate EV. Treat that as a red flag.
- Setting a budget: price minus EV, times the number of openings, gives you the expected cost of a session. Our example: 50 openings × $2.49 = about $125 expected loss.
For how these numbers work in battles, where several players split or take the combined value, see how CS2 case battles work. For rarity tiers and drop rates, see CS2 case odds explained.
Frequently asked questions
What is a good expected value for a CS2 case?
Any case with an expected value below its price loses money on average, and almost every case on every site is priced that way. A smaller gap means a slower average loss, not a profit.
Can a case have an expected value above its price?
In principle, yes, if a site misprices it or item values move after the case is set up. In practice, sites adjust prices, and the item values used in the calculation may not be what you can actually sell for.
Does expected value tell me what I will get from one opening?
No. It is an average over a very large number of openings. A single opening usually returns far less than the expected value, with a small chance of returning far more.
Where do I find the odds for a case?
Reputable sites publish a probability for every item in a case. If a site does not show the full odds table, you cannot calculate the expected value, and that is a warning sign in itself.
18+ only. Set a budget you can afford to lose and stop at your limit. Support: Gambling Therapy · BeGambleAware.