Three Formats, Same Information
If you've ever pulled odds data from different sources, you've probably noticed the formats don't match. European sites show 2.50. UK sources say 3/2. American data reads +150. Confusing? Sure. But here's the thing—they're all saying the exact same thing.
For anyone doing sports data analysis, understanding these conversions isn't optional. It's foundational. At OddsFlow, our AI models process odds from markets worldwide, so format conversion is something we deal with constantly.
Let me break down each format and show you how they connect.
Decimal Odds: The Data-Friendly Format
If you're building models or doing any kind of quantitative analysis, decimal odds are your friend. They're mathematically clean and convert directly to probability.
How they work: The number represents total return per unit. Odds of 2.50 means you'd get 2.50 back for every 1 unit—so 1.50 profit plus your original stake.
| Decimal | Total Return (per $1) | Profit | Implied Probability |
|---|
| 1.50 | $1.50 | $0.50 | 66.7% |
| 2.00 | $2.00 | $1.00 | 50.0% |
| 3.00 | $3.00 | $2.00 | 33.3% |
| 5.00 | $5.00 | $4.00 | 20.0% |
Probability = 1 / Decimal Odds
2.50 odds = 1 / 2.50 = 0.40 = 40%
This is why decimal is the standard for analytics. One simple division gets you to probability.
Fractional Odds: The Traditional Format
You'll see fractional odds in UK data sources and older datasets. They show profit relative to stake—so 5/2 means 5 units profit for every 2 units staked.
| Fractional | Decimal | Probability |
|---|
| 1/2 | 1.50 | 66.7% |
| 1/1 (Evens) | 2.00 | 50.0% |
| 3/2 | 2.50 | 40.0% |
| 2/1 | 3.00 | 33.3% |
| 4/1 | 5.00 | 20.0% |
Decimal = (Numerator / Denominator) + 1
5/2 = (5 / 2) + 1 = 2.5 + 1 = 3.50
For analysis purposes, I always convert fractional to decimal immediately. It makes everything easier downstream.
American Odds: The Plus/Minus System
American odds look weird if you're not used to them. They use positive and negative numbers anchored around $100.
Positive odds (+150): Shows profit on a $100 stake. +150 means $150 profit.
Negative odds (-200): Shows how much you'd stake to profit $100. -200 means you'd need to stake $200.
| American | Decimal | Probability |
|---|
| -200 | 1.50 | 66.7% |
| +100 | 2.00 | 50.0% |
| +150 | 2.50 | 40.0% |
| +200 | 3.00 | 33.3% |
| +400 | 5.00 | 20.0% |
If positive: Decimal = (American / 100) + 1
+150 = (150 / 100) + 1 = 2.50
If negative: Decimal = (100 / |American|) + 1
-200 = (100 / 200) + 1 = 1.50
The Master Conversion Table
Keep this handy when you're working with multi-source data:
| Decimal | Fractional | American | Probability |
|---|
| 1.25 | 1/4 | -400 | 80.0% |
| 1.50 | 1/2 | -200 | 66.7% |
| 1.80 | 4/5 | -125 | 55.6% |
| 2.00 | 1/1 | +100 | 50.0% |
| 2.50 | 3/2 | +150 | 40.0% |
| 3.00 | 2/1 | +200 | 33.3% |
| 4.00 | 3/1 | +300 | 25.0% |
| 5.00 | 4/1 | +400 | 20.0% |
| 10.00 | 9/1 | +900 | 10.0% |
Why This Matters for AI Analysis
At OddsFlow, we aggregate odds data from markets around the world. That means handling all three formats constantly. Our preprocessing pipeline converts everything to decimal (and then to implied probability) before any analysis happens.
Why decimal? Because it's the cleanest path to what we actually care about: the probability estimate embedded in the price.
When you're comparing odds across different bookmakers or tracking how prices move over time, consistent formatting is essential. A model that can't properly convert between formats will produce garbage outputs.
Quick Takeaways
Every format encodes the same underlying probability—they're just different ways of expressing it. For any serious data work, decimal is the way to go. It converts cleanly to probability and makes mathematical operations straightforward.
If you're building your own analysis tools, standardize on decimal early in your pipeline. Your future self will thank you.
📖 Related: Implied Probability Explained • What Are Football Odds
*OddsFlow provides AI-powered sports analysis for educational and informational purposes.*