Every fixture on BetSignals carries two separate numbers: a percentage and a star rating. They look like they might be two versions of the same thing. They are not. The percentage tells you how likely an outcome is. The star rating tells you how much betting value it carries. This guide explains how each is worked out, why they often disagree, and how to use them together.
---
Two numbers, two jobs
The percentage answers one question: how often will this happen? A strong home favourite might be 70% to win. That is a high-likelihood outcome, and high-likelihood outcomes are what you want as legs in an accumulator.
The star rating answers a completely different question: is the price on offer better than it should be? A 70% favourite is only a good bet if the odds pay out more than a 70% chance deserves. Very often they do not, because the market has already priced that favourite correctly.
Keep the two jobs separate in your head:
- Percentage = likelihood. Best for building accumulators from probable results.
- Star rating = value. Best for finding single bets where the price is wrong.
---
The percentage: how likely is the outcome
Every match is run through the BetSignals model, which produces three win probabilities: home, draw and away. They add up to 100%. The outcome with the highest probability is the model's primary selection, and that figure is the headline percentage you see on the fixture.
A higher percentage means the model thinks the result is more likely. Nothing more, nothing less. It is a statement about the match, not about the odds. That is exactly what you want when you are choosing accumulator legs: you are stacking outcomes that are likely to land, and the percentage ranks them for you.
---
From probability to fair odds
A probability can be turned into odds. This gives you the fair price: the odds that exactly reflect the model's view, before any bookmaker margin is added. The sum is simple:
Fair odds = 1 divided by the probability
So a 50% chance is fair odds of 2.00 (1 divided by 0.50). A 40% chance is 2.50. A 25% chance is 4.00. The smaller the probability, the bigger the fair price. This is the same idea as implied probability, run in reverse.
The fair price is the bridge between the two numbers. The percentage sets the fair price, and the fair price is what we compare against the market to find value.
---
Value: comparing the fair price to the market
A selection carries value when the bookmaker's odds are longer than the fair odds. In other words, the market is paying you more than the model thinks the chance is worth. This is the foundation of every value bet.
Take a worked example. Say the model makes an away side 45% to win. The fair odds are 1 divided by 0.45, which is 2.22. Now look at what the bookmaker is offering:
- At 2.50, the price is longer than 2.22. You are being paid more than a 45% chance deserves. That is value.
- At 1.90, the price is shorter than 2.22. You are being underpaid. There is no value, however likely the result is.
The size of that gap is the edge, and it is what drives long-term return. The star rating is built directly on top of this comparison.
---
The star rating: how much value
The star rating is worked out in two steps.
Step 1: do the two models agree? Every match is checked by two independently built models, including the attacking and defending Elo model. If they point at different outcomes, you get a red cross and no signal, whatever the percentages say. A cross means only one thing: the models disagree, so there is no selection. Sit on your hands.
Step 2: is there value, and how strong is it? If the models agree, the pick is graded by value, not by how likely it is to win:
- One star is an agreed selection with no value. The market has priced it fairly, or too short. Most picks land here, and that is normal.
- Two to five stars mean the selection carries value, graded by how strong that value is.
---
The away ladder, and the home cap
Home and away run on deliberately different scales, and this is where the model's real behaviour shows through.
Away selections carry the full ladder. Value on a shorter-priced away side is the highest conviction, because shorter prices are easier to get right. As the price lengthens, the value becomes more speculative and the rating steps down.
Home selections cap at two stars. Home results are priced very efficiently by the market, so genuine value on a home side is rare and, when it appears, weak. A home pick is therefore never more than two stars, however likely the home win is.
| Rating | What it means |
| --- | --- |
| ★★★★★ | Away value at odds up to 2.50 |
| ★★★★ | Away value at odds 2.50 to 3.50 |
| ★★★ | Away value at odds 3.50 to 5.00 |
| ★★ | Away value above 5.00 (higher variance), or any home value |
| ★ | An agreed pick with no value |
| ✕ | The two models disagree, no selection |
---
Why a strong favourite can be a low star
This is the point that catches most people out, so it is worth stating plainly.
A team can be very likely to win and still be a poor bet. A home favourite at 75% to win is a strong result for an accumulator, but if the market is offering odds that already reflect that 75%, there is no value in backing it on its own. It will show a high percentage and a low star rating, and both are correct. They are simply answering different questions.
The reverse also happens. A mid-priced away side might be only 45% to win, a modest percentage, yet be trading at odds that imply a smaller chance than that. That is where the stars light up, because that is where the value is.
---
Seeing it on the Performance page
The Performance page proves the two numbers behave differently. For home and for away, it shows two tables side by side:
- By win probability. This buckets picks by likelihood. You will see that high-probability bands win often, which is exactly what you want from accumulator legs.
- By star rating. This buckets the same picks by value, and tracks the profit or loss from a flat one-pound stake on each. This is where you see which star levels have actually returned a profit.
Read together, the two tables tell the whole story: likelihood keeps you winning legs, value keeps you making money. They are not the same thing, and the page shows you both.
---
How to use the two numbers
- Use the percentage when you are building an accumulator. You want likely outcomes, and the percentage ranks them.
- Use the star rating when you are looking for a single value bet. More stars means a stronger price advantage, graded by side and by odds.
- Treat a high percentage with a low star as a good acca leg that is not worth backing on its own.
- Treat a cross as a stand-aside: the models do not agree, so there is no selection.
The percentage tells you what is likely. The star rating tells you what is worth backing. The data surfaces what the models see. The decision is always yours.