CS2 Odds of Getting Factory New: How Rare FN Really Is and Why It Depends on the Skin
Updated · 
Written by Lena Ortiz ·
Pulling a Factory New skin feels special because it is rare, but how rare depends on the skin far more than most people think. For a typical skin with the full float range, the chance is about 3%. For a Doppler knife it's about 96%. For an AWP Asiimov it's zero, because that skin can't exist in Factory New at all.
This guide explains where that 3% comes from, how float caps change it, and how to work out the real odds for the exact skin you're chasing. We also cover StatTrak and Covert maths, how many cases a Factory New pull usually takes, what changed for trade-ups in October 2025, and how wear works on third-party case sites. Valve doesn't publish wear odds, so the numbers here come from a large community analysis of real items plus our own calculations. We didn't open test cases for this page.
Reminder: unboxing is gambling with real money. Only open cases where it's legal for you and if you're of legal age.
Wear and float in one minute
Every CS2 skin has a float value between 0 and 1. The lower the float, the cleaner the skin. The float decides the wear name you see:
| Wear | Float range | Width of range |
|---|---|---|
| Factory New (FN) | 0.00 to 0.07 | 0.07 |
| Minimal Wear (MW) | 0.07 to 0.15 | 0.08 |
| Field-Tested (FT) | 0.15 to 0.38 | 0.23 |
| Well-Worn (WW) | 0.38 to 0.45 | 0.07 |
| Battle-Scarred (BS) | 0.45 to 1.00 | 0.55 |
If you need a refresher on floats, our float value guide and float chart cover the basics. This page is only about one question: when a new skin is generated, what's the chance it lands in Factory New?
The obvious guess is "7%", because Factory New covers 7% of the 0 to 1 scale. That guess is wrong, and the reason is the most useful thing to understand about wear odds.
Where the 3% comes from
Valve has never published how floats are generated. What we have instead is a large data project. CSFloat's developer (Step7750) analysed roughly 250 million real items and found that floats aren't spread evenly across 0 to 1. Instead, the game first picks a wear bracket with fixed weights, then picks a float evenly inside that bracket.
The weights the analysis found are:
| Bracket | Share of generated floats | Share of the 0-1 scale |
|---|---|---|
| 0.00 to 0.07 (FN) | about 3% | 7% |
| 0.07 to 0.15 (MW) | about 24% | 8% |
| 0.15 to 0.38 (FT) | about 33% | 23% |
| 0.38 to 0.45 (WW) | about 24% | 7% |
| 0.45 to 1.00 (BS) | about 16% | 55% |
Two things jump out. Factory New is under-weighted, at about 3% instead of 7%. And Battle-Scarred, which covers more than half the scale, only gets about 16% of floats. Minimal Wear and Well-Worn, each tiny on the scale, get about a quarter each.
There's a second step that matters even more. The number picked in the first step is a "raw" float between 0 and 1. The game then squeezes it into the skin's own float range using this formula:
final float = raw float × (max − min) + min
For a full-range skin (0.00 to 1.00), the final float equals the raw float, so the 3% holds. For any skin with a narrower range, the same raw number lands somewhere different, and the wear odds change.
These figures come from community data, not Valve. They've held up well across years of trading and are the basis most float tools use, but treat them as a very good estimate rather than an official rule.
Factory New odds depend on the skin's float range
Here's the maths in practice. To find a skin's Factory New chance, work out which raw floats produce a final float under 0.07, then add up the weights for that part of the raw scale.
Full range: 0.00 to 1.00
Raw and final floats are the same, so Factory New is simply the first bracket: about 3%. Plenty of skins work like this, and it's the number most people mean when they ask about Factory New odds.
Doppler and Fade knives: 0.00 to 0.08
The final float is raw × 0.08. A final float under 0.07 needs a raw float under 0.875. That covers the whole FN, MW, FT and WW brackets (3 + 24 + 33 + 24 = 84%) plus part of the BS bracket: (0.875 − 0.45) ÷ 0.55 of 16%, which is about 12.4%. Total: about 96.4% Factory New, with the rest Minimal Wear.
That's why you almost never see a Minimal Wear Doppler. It's not that Valve "made Dopplers clean"; the range is just so narrow that most raw floats map under 0.07.
A range of 0.00 to 0.50
Final float = raw × 0.50. Under 0.07 needs a raw float under 0.14. That's the full FN bracket (3%) plus most of the MW bracket: (0.14 − 0.07) ÷ 0.08 of 24%, which is 21%. Total: about 24% Factory New. A half-range skin is about eight times as likely to come out Factory New as a full-range one.
A range of 0.00 to 0.70
Under 0.07 needs a raw float under 0.10, so 3% plus (0.03 ÷ 0.08) × 24% = 9%. Total: about 12% Factory New.
Older skins with a 0.06 to 0.80 range
Many older skins use a 0.06 to 0.80 range. Their final float is raw × 0.74 + 0.06. To land under 0.07, the raw float must be under about 0.0135. That's only a sliver of the FN bracket: 0.0135 ÷ 0.07 × 3%, which is about 0.58%. Roughly one in every 170 of these skins comes out Factory New, and almost all of those sit between 0.06 and 0.07.
Skins that can't be Factory New
If the minimum float is above 0.07, Factory New is impossible:
- AK-47 Redline (0.10 to 0.70): no Factory New. Its Minimal Wear chance is only about 7%, because a final float under 0.15 needs a raw float under about 0.083.
- AWP Asiimov (0.18 to 1.00): no Factory New and no Minimal Wear. Its best wear is Field-Tested, at about 40.5%.
Quick reference
| Float range | Example | Factory New chance |
|---|---|---|
| 0.00 to 0.08 | Doppler, Fade knives | about 96% |
| 0.00 to 0.50 | half-range skins | about 24% |
| 0.00 to 0.70 | many newer skins | about 12% |
| 0.00 to 1.00 | full-range skins, Case Hardened | about 3% |
| 0.06 to 0.80 | many older skins | about 0.6% |
| min above 0.07 | Redline, Asiimov | 0% |
Always check the exact range for the skin you want before doing the maths. Float ranges are listed on most skin databases and on our float chart. A difference of 0.01 at the bottom of the range can change the Factory New chance a lot.
See every item's drop chance on Clash.gg + claim free cases
The full calculation: rarity × wear × StatTrak
Factory New is only one part of the odds for a specific pull. To get the chance of the exact item you're after from an official case, multiply three things:
- The rarity chance. Valve's disclosed case odds are about 79.92% Mil-Spec, 15.98% Restricted, 3.20% Classified, 0.64% Covert and 0.26% for a rare special item (knife or gloves). Our case odds guide covers these in detail.
- The chance of that specific skin within the rarity. Skins of the same rarity in a case are generally treated as equally likely, so if a case has two Coverts, each gets half of the 0.64%.
- The wear chance for that skin's float range, from the tables above.
Add StatTrak on top if you want it: about 10% of drops from cases that include StatTrak versions come out StatTrak.
Worked examples
Any full-range Covert in Factory New:
0.64% × 3% = 0.0192%, or about 1 in 5,200 openings.
One specific full-range Covert in a case with two Coverts:
0.32% × 3% = 0.0096%, or about 1 in 10,400.
StatTrak Factory New, any full-range Covert:
0.64% × 3% × 10% = 0.00192%, or about 1 in 52,000.
A Doppler knife in Factory New, from a case with Doppler knives:
the knife chance is 0.26% across every knife and finish in the case, so you'd first need the share of that pool that's Doppler. Once you have a Doppler, it's about 96% to be Factory New. In other words, for Dopplers the hard part is the knife, not the wear.
A full-range knife finish like Case Hardened in Factory New:
the 0.26% knife pool, times that finish's share of the pool, times about 3%. Even before you split by knife type and finish, 0.26% × 3% is about 1 in 12,800 openings. Split it by finish and it gets much longer.
How many cases does a Factory New pull take?
People often say "1 in 5,200, so I need 5,200 cases". That's not how it works. Each opening is independent, so even after 5,200 openings there's a decent chance you still haven't hit.
For a chance of p per opening, the chance of at least one hit in n openings is 1 − (1 − p)ⁿ. Using the "any full-range Covert in Factory New" chance of 0.0192%:
| Openings | Chance of at least one FN Covert |
|---|---|
| 1,000 | about 17.5% |
| 3,610 | about 50% |
| 5,208 | about 63% |
| 12,000 | about 90% |
For comparison, getting any Covert at all (0.64%) reaches a 50% chance after about 108 openings. Adding "Factory New" on top multiplies the work by roughly 33.
Put a price on that and the point gets clearer. Even at a low key-plus-case cost per opening, thousands of openings cost far more than buying the Factory New skin on the market. Unboxing is entertainment, not a way to get a specific item cheaply. Our case odds and EV tool can show what a case returns on average.
Low floats inside Factory New
Factory New isn't one thing. Inside 0.00 to 0.07, a 0.001 float can be worth much more than a 0.069. Because the game picks evenly inside the bracket, the chances for a full-range skin are easy to work out:
- Under 0.01: about 3% × (0.01 ÷ 0.07) = 0.43%, or about 1 in 230 of all drops of that skin.
- Under 0.001: about 0.043%, or about 1 in 2,300.
- Under 0.0001: about 1 in 23,000.
On a capped range the maths changes again. On a 0.00 to 0.08 Doppler, a final float under 0.01 needs a raw float under 0.125, which is the FN bracket (3%) plus (0.055 ÷ 0.08) × 24% from MW, about 19.5%. Low floats are common on Dopplers and very rare on full-range skins, which is one reason low-float premiums vary so much from skin to skin.
Why so few skins come out Factory New
The weighting looks strange at first. Why make Factory New only 3% when it covers 7% of the scale? Nobody outside Valve knows the reasoning for sure, but the effect is clear: Factory New and very low floats stay rare, which keeps them valuable on the market.
It also explains a few things players notice:
- Field-Tested is the most common wear for full-range skins, at about a third of drops.
- Well-Worn turns up more than you'd expect for such a narrow range, at about a quarter of drops.
- Battle-Scarred is less common than its huge range suggests, at about 16%.
Trade-ups: the controllable route to Factory New
Unboxing gives you a fixed chance. Trade-up contracts let you steer the wear, because the output float depends on the floats you put in.
In October 2025 Valve changed the trade-up float formula and added Covert-to-knife contracts. Here's how it works now, as explained by HLTV's patch coverage and trade-up tool sites:
- Each input float is turned into a 0 to 1 value within its own range: (float − min) ÷ (max − min).
- Those values are averaged.
- The average is mapped onto each possible output's range: output = out min + average × (out max − out min).
For a regular trade-up you use 10 inputs. For the new Covert contract you use 5 Coverts for a knife or gloves (or 5 StatTrak Coverts for a StatTrak knife), from the collections of the items you put in.
What this means for Factory New
- Full-range output (0.00 to 1.00): you need an average normalised input under 0.07. Full-range inputs under 0.07 do it, which means Factory New inputs.
- Output range 0.06 to 0.80: you need an average normalised input under about 0.0135. That's a very low float on every input.
- Output range 0.00 to 0.08 (for example Doppler or Fade knives from a Covert contract): almost any average under 0.875 gives Factory New, so even well-used Coverts can produce a Factory New knife of those finishes.
The change also killed an old trick. Before October 2025, capped inputs with a low raw float (for example a 0.06 on a 0.06 to 0.80 skin) counted as 0.06 in the average. Now that same 0.06 counts as 0.00 after scaling, while a 0.30 on a 0.00 to 0.40 skin counts as 0.75. The same float can be "cheap" or "expensive" for the contract depending on its range.
Trade-ups aren't free money either. Low-float inputs cost more, and the output is still random among the possible results. Our trade-up contract guide and trade-up calculator help you check the outputs and floats before you commit.
Open cases and play on CSGOEmpire + free case with FREECASES4U
Factory New on case opening sites
Third-party case sites work differently from official cases. On sites like Clash.gg and CSGOEmpire, a case is a list of specific items, each already tied to a wear, and each with its own drop percentage set by the site. So the Factory New chance on a site case is simply the total percentage of the Factory New items in that case.
That has a few consequences:
- The CSFloat 3% doesn't apply. The site has already chosen which wears are in the case and how likely each one is.
- You can read the odds directly. Look at the item list and add up the percentages of the Factory New entries. Many cases include cheap Factory New filler items, so "FN" on a site case doesn't mean "rare".
- The price is what matters. A Factory New item in a site case is valued at a market price, and the case price is set so the house keeps an edge on average. Check what the case returns on average, not just the shiny top item.
- Provably fair rolls let you check that the roll wasn't changed after you opened, but they don't change the odds or the house edge.
On Clash.gg you can see each item's percentage in the case before you open. Our Clash.gg trade-up review also explains its own upgrade-style game, which works on site odds rather than Valve's floats.
StatTrak and Factory New
StatTrak and wear are separate rolls. Being StatTrak doesn't change your float odds. From official cases with StatTrak versions, about 10% of drops are StatTrak, and the wear then follows the same float maths as above. A StatTrak Factory New of a full-range skin is about 3% × 10% = 0.3% of all drops of that skin, or about 1 in 330.
Whether StatTrak is worth paying extra for is a separate question. We cover it in is StatTrak worth it.
Myths about Factory New odds
"Some times of day give better floats." There's no evidence for this. Floats come from the game's random generation, and nothing in the community data shows a link to time, day or server.
"Opening lots of cases in a row raises your chance." No. Each opening is independent. A long run of Battle-Scarred drops doesn't make the next one more likely to be Factory New.
"Factory New means flawless." Not always. Some skins show visible wear or pattern effects even at low floats, and some Field-Tested skins look nearly identical to Factory New. Check screenshots before paying a Factory New premium.
"All Factory New skins have the same value." Float, pattern and stickers all matter. A 0.069 and a 0.001 of the same skin can be priced very differently.
"Valve publishes wear odds in China." Valve's China disclosure covers rarity odds, not wear. Wear odds are community estimates.
How to work out the Factory New odds for any skin
Here's the process in five steps:
- Find the skin's float range. Skin databases and our float chart list min and max floats.
- Work out the raw float limit. For Factory New, raw limit = (0.07 − min) ÷ (max − min). If the min is above 0.07, the answer is 0%.
- Add up the bracket weights below that raw limit. 3% for 0 to 0.07, 24% for 0.07 to 0.15, 33% for 0.15 to 0.38, 24% for 0.38 to 0.45, 16% for 0.45 to 1. For a partial bracket, take the matching fraction of its weight.
- Multiply by the rarity chance and the skin's share of that rarity in the case.
- Multiply by 10% if you need StatTrak.
Example: a skin with a 0.00 to 0.60 range. Raw limit = 0.07 ÷ 0.60 = 0.1167. That's 3% plus (0.0467 ÷ 0.08) × 24% = 14%. Total about 17% Factory New. If it's a Classified skin in a case with four Classifieds, the chance from one opening is 3.20% ÷ 4 × 17%, about 0.136%, or about 1 in 735.
Quick recap
- For a full-range skin, the chance of Factory New is about 3%, not 7%, based on CSFloat's analysis of about 250 million items.
- The skin's float range changes everything: about 96% for Doppler and Fade knives, about 0.6% on a 0.06 to 0.80 range, and 0% if the minimum is above 0.07.
- Multiply rarity × skin share × wear × StatTrak to get the odds of the exact item. A full-range Factory New Covert is about 1 in 5,200 openings.
- Even 1 in 5,200 only gives you about a 63% chance after 5,200 openings.
- Trade-ups give you control over wear, and the October 2025 change made input ranges matter much more.
- On case sites, wear is fixed per item and the odds are listed, so read the case before you open it.
18+ only. These sites use real money and real skins. If opening cases stops being fun, read our Responsible gambling page and take a break.