How CS2 Odds Are Calculated, Step by Step
Updated · 
Written by Milan Kovac ·
Every CS2 price you see is the end of a short chain: a rating, a win chance, some map and format adjustments, a margin, then a lot of nudging as money and news come in. Bookmakers don't publish their models, but the logic behind them is well known, and once you understand it you can do a rough version yourself, bro. That's the point of this guide: not to copy any one book, but to show how a price is built so you can tell when one looks off.
We'll go step by step with one worked example, using standard formulas and numbers we computed ourselves. Nothing here is a quote from a specific bookmaker.
Usual reminder: betting is entertainment money. The margin is built into every price, and most bettors lose over time.
Step 1: rate the teams
The starting point is a number for each team's strength. The classic method is an Elo-style rating:
- Every team has a rating.
- After each match, the winner takes points from the loser.
- An upset moves more points than an expected result.
- Over time, ratings settle at levels that predict results reasonably well.
Real models add more on top: recent form weighted more than old results, map-level ratings, player-level data, opponent strength, LAN or online. But the core idea is the same: a rating gap, turned into a chance.
From rating gap to win chance
The standard Elo formula converts a rating gap into an expected score: 1 / (1 + 10^(-gap / 400)).
| Rating gap | Stronger team's expected score | Fair odds (stronger) |
|---|---|---|
| 0 | 50.0% | 2.00 |
| 25 | 53.6% | 1.87 |
| 50 | 57.1% | 1.75 |
| 100 | 64.0% | 1.56 |
| 150 | 70.3% | 1.42 |
| 200 | 76.0% | 1.32 |
| 300 | 84.9% | 1.18 |
Fair odds are just 1 divided by the chance. They're what a price would be with no margin at all.
Step 2: adjust each map
CS2 isn't played on "a map"; it's played on specific maps chosen in the veto. A team with an overall 100-point edge might be much stronger on its own pick and weaker on the opponent's. Good models carry a rating per map, or an adjustment per map, and price each one separately.
Our worked example: Team A has a 100-point overall edge. We give it +60 on its own pick, -120 on Team B's pick (B's comfort map), and no change on the decider.
| Map | Effective gap | Team A chance | Fair odds A | Fair odds B |
|---|---|---|---|---|
| Map 1 (A's pick) | +160 | 71.5% | 1.40 | 3.51 |
| Map 2 (B's pick) | -20 | 47.1% | 2.12 | 1.89 |
| Decider | +100 | 64.0% | 1.56 | 2.78 |
Notice that Team A is actually the underdog on Map 2. If a book priced every map off the overall 64%, Map 2 would be badly wrong.
Claim free cases on Clash.gg, no code needed
Step 3: combine maps into a series price
For a best-of-three, Team A wins the series by winning Maps 1 and 2, or by splitting them and winning the decider:
- A wins 2-0: 0.715 x 0.471
- A wins 2-1 via win-lose-win: 0.715 x 0.529 x 0.640
- A wins 2-1 via lose-win-win: 0.285 x 0.471 x 0.640
Add them up and Team A's series chance is 66.5%, fair odds 1.50, with Team B at 2.99. That's a bit higher than the plain 64% overall number, because Team A's big edge on its own pick and on the decider outweighs its small deficit on Map 2.
Before the veto is known, books have to guess the likely maps. After the veto, a good model reprices. That's one reason prices can move just before a match starts. Our CS2 handicap betting guide shows how map spreads are priced from the same per-map chances.
Step 4: add the bookmaker margin
Fair odds total exactly 100% when you add up the implied chances. A bookmaker shortens the prices so the total is above 100%; the extra is its margin (also called overround or vig). How it spreads that margin matters.
Method A: spread evenly (proportional)
Scale both chances up by the same factor. With a 5% margin on our 66.5% / 33.5% series:
- Team A: 1.43
- Team B: 2.84
Method B: load the underdog (power-style)
Some pricing approaches put more of the margin on the longer price. On the same series with the same 5% total:
- Team A: 1.46
- Team B: 2.75
The total margin is identical, but the underdog bettor pays more of it. The effect gets bigger with lopsided matches. Here's an 85% / 15% match with a 6% margin:
| Method | Favourite price | Favourite implied | Underdog price | Underdog implied |
|---|---|---|---|---|
| Fair (no margin) | 1.18 | 85.0% | 6.67 | 15.0% |
| Spread evenly | 1.110 | 90.1% | 6.29 | 15.9% |
| Underdog-loaded | 1.152 | 86.8% | 5.21 | 19.2% |
In the underdog-loaded version, someone backing the 15% team is paying for a 19.2% chance. That's one reason big underdogs are often poor value even when they look generous. It's also a known pattern across betting called the favourite-longshot bias.
Our CS2 betting odds explained guide shows how to find the total margin on any market and strip it out.
Step 5: adjust for news
Ratings are built on past results. News changes the present:
- Stand-ins and roster changes. If a stand-in costs Team A 60 rating points, the overall gap drops from 100 to 40 and the expected score falls from 64.0% to 55.7%. That's a big price change from one player.
- Format. Bo1s are more random than Bo3s, so the same rating gap gives a smaller edge in a Bo1 series price.
- Venue. LAN vs online can matter for some teams.
- Motivation and schedule. Already-qualified teams, tired teams, late-night slots.
Books usually react fast to big public news. Smaller news, like a player playing through illness, may only show up as a price drift.
Step 6: move the line with the money
Once the market opens, the price becomes a conversation between the book and its customers. Two forces move it:
Liability
Say a book opens Team A at 1.80 and Team B at 2.05, then takes $8,000 on A and $2,000 on B. If A wins, it pays out $14,400 against $10,000 taken, a $4,400 loss. If B wins, it pays $4,100 and keeps $5,900. That's lopsided risk, so the book shortens A and lengthens B to attract money on B, for example to 1.72 / 2.18.
Sharp money
Some customers have a history of beating the closing price. When they bet, books often move the line more, because that money carries information. Big early moves on a CS2 match can mean professional bettors disagree with the opening price.
That's why the price at match start (the closing line) is usually more accurate than the opening price. If you regularly get better odds than the close, that's a good sign your reads are sound.
CSGOEmpire match betting + free case with FREECASES4U
Live odds: the same chain, faster
In-play prices follow the same logic, updated round by round: current score, economy, side (CT or T), which map is next, and a model of how often teams come back from each situation. Live models react instantly to the scoreboard but can miss context, such as a team's T side being much stronger on a specific map. Our CS2 live betting guide covers where in-play prices tend to overreact.
How to build your own rough price
You don't need a full model to use this:
- Estimate each map for both teams based on recent results on that map with the current rosters.
- Combine them into a series chance (or use an overall chance for a Bo1).
- Convert to fair odds: 1 / chance.
- Remove the book's margin to see its fair price.
- Compare. Only bet where your fair price is clearly better than the book's fair price, not just a hair better.
Our esports odds calculator does the conversion and margin work for you.
Why prices differ between books
- Different models weight form, maps and players differently.
- Different margins on the same match.
- Different risk positions. A book with too much money on one side moves its price; another might not.
- Different speed. Some react to news in seconds; others are slower.
That's why comparing prices is so useful. A slightly better price on every bet adds up.
Common mistakes when reading odds
- Treating odds as predictions. They're a chance plus margin, shaped by money.
- Ignoring the veto. Series prices before the veto are partly guesses.
- Backing big underdogs for "value". The margin is often heaviest there.
- Chasing line moves late. By the time a move is obvious, the value may be gone.
Where to compare CS2 odds
- Crypto esports books such as Cloudbet list CS2 match winner, map, handicap and totals markets, so you can compare the whole chain from map to series.
- Skin sites with match betting, like CSGOEmpire match betting, offer match winner markets on skin balance.
- Check two or three prices before every bet.
Check that betting is legal where you live and that you meet the legal age.
Stay in control
Understanding odds doesn't remove the margin. Keep stakes small, never chase losses and take breaks. Our responsible gambling page has free, confidential help.
Quick summary
- CS2 prices start from team ratings; a 100-point Elo gap is about 64.0% before any margin.
- Good models price each map from the veto, then combine maps into a series chance (66.5% in our example).
- Books add a margin, sometimes loading more of it onto underdogs (a 15% team priced as 19.2% in our example).
- News and money then move the line; closing prices are usually the most accurate.
- Build a rough price yourself, strip the book's margin, and bet only clear gaps.