How many cases does it take to get a knife?
There is no fixed number. Every case is a fresh roll with a 0.26% chance of a knife or gloves, and past cases do not change the next one. What you can work out is the chance of at least one knife after a set number of cases.
With the published odds, you pass a 50% chance at 267 cases. You need 885 cases for 90%, 1,151 cases for 95% and 1,769 cases for 99%. Even then, 1 person in 100 would still have no knife.
The calculator above does this math for any number of cases, and adds the cost at the case and key prices you enter.
How is the chance calculated?
The chance of missing on one case is 1 - p, where p is the drop chance. Each case is independent, so the chance of missing on all N cases is (1 - p) multiplied by itself N times. The chance of at least one hit is what is left:
chance = 1 - (1 - p)^N
To find how many cases you need for a target chance T, the formula is turned around:
cases = ln(1 - T) / ln(1 - p), rounded up.
The expected number of hits is simpler. It is N × p. After 385 cases you expect 1 knife or gloves on average, but the chance of getting at least one is only 63.3%. Some people get two, and many get none.
What drop rates does the calculator use?
It uses the case odds Valve published in 2017, which still apply to CS2 weapon cases. You can see the full table and the source on our CS2 drop rates page.
| What you want | Chance per case | About 1 in |
|---|---|---|
| Knife or gloves | 0.26% | 385 |
| Covert or better | 0.90% | 111 |
| Covert skin | 0.64% | 156 |
| Classified skin | 3.2% | 31 |
"Covert or better" adds the Covert chance and the knife or gloves chance together. It answers the question "what are my odds of a red or gold pull?"
How many cases for each chance?
This table shows the number of cases needed to reach each chance of at least one hit.
| Chance | Knife or gloves | Covert or better | Covert skin | Classified skin |
|---|---|---|---|---|
| 50% | 267 | 77 | 108 | 22 |
| 90% | 885 | 255 | 359 | 71 |
| 95% | 1,151 | 332 | 467 | 93 |
| 99% | 1,769 | 510 | 718 | 142 |
The jump from 95% to 99% is large for every target. That last few percent is the expensive part, because each extra case only adds a small amount of chance.
What does it cost?
Each case needs a key, so the cost per case is the case price plus the key price. Pick a case and the case price starts at its Skinport lowest listing from our price snapshot. If nothing was listed, it starts at the 7-day median sale instead. The key price starts at $2.49, the price of a key bought inside the game in USD. Valve shows that price only in the game, so our source is a secondary one. You can type over both prices.
On the 24 Sep 2026 snapshot, the Fever Case had a Skinport lowest listing of $0.53. With a $2.49 key, each case costs $3.02. That means:
| Target | Cases | Cost at $3.02 per case |
|---|---|---|
| 50% chance of a knife or gloves | 267 | $806.34 |
| 90% chance | 885 | $2,672.70 |
| 95% chance | 1,151 | $3,476.02 |
| 99% chance | 1,769 | $5,342.38 |
The calculator also shows the cost per expected hit. That is the price of one case divided by the chance per case. At $3.02 per case it is about $1,162 per knife or gloves on average.
Worked example: 100 cases
You open 100 cases and want a knife or gloves. The chance of missing on one case is 0.9974. Multiply that by itself 100 times and you get about 0.7708.
So the chance of at least one knife or gloves is 1 minus 0.7708, which is about 22.9%. You would expect 0.26 hits on average. With Fever Cases at $0.53 (Skinport lowest listing, 24 Sep 2026) and $2.49 keys, the 100 cases cost $302.
In plain words, more than 3 out of 4 people who open 100 cases get no knife or gloves at all.
Worked example: aiming for a Covert
You want any red skin, and you plan to open 50 cases. The chance per case for a Covert skin is 0.64%. The chance of missing all 50 is 0.9936 to the power of 50, which is about 0.7254.
That gives about a 27.5% chance of at least one Covert. If you count knives and gloves as a win too, the chance per case goes up to 0.9% and the chance over 50 cases goes up to about 36.4%.
Does opening more cases make a knife more likely next time?
No. This is the most common mistake with case odds. A case does not "remember" that you missed. After 300 empty cases, the next case still has the same 0.26% chance.
The chance of at least one knife grows as you plan more cases, because you get more tries. But no single case gets better. This is why the calculator shows the chance for a plan of N cases, not a countdown to a sure knife.
What this calculator does not include
It does not include the value of what you unbox. Most pulls are Mil-Spec skins that are worth far less than a case and key. A full expected value needs live prices for every item, which we plan to add in a later tool.
It also does not cover case-opening websites. Those sites set their own odds and prices, so these numbers do not apply to them. If you want to understand the math those sites use, read house edge explained or try the house edge calculator.
When should you use this tool?
Use it before you spend money on cases, to see what a knife really costs on average. It is also useful to settle arguments about odds, since you can share the link with your exact numbers.
If you just want the fun of opening without the cost, try the free case opening simulator. It uses the same drop rates.
