Each market we model has its own page, and each page carries only the fixtures whose shape fits that market. A fixture appears on the both-teams-to-score page because the weaker side is still expected to score, not because we have run a BTTS opinion on every match of the day.
That is why a fixture you are looking for may not appear on the page you expect. If the data does not support a call in that market, the fixture is simply not there.
The match result family — 1X2, double chance and the draw — answers who wins. The goal lines answer how many. Both teams to score answers whether both attacks function, independently of the result. Correct score asks for both at once and prices accordingly.
If you are unsure which to look at, start from the question you actually have a view on. A view about whether a match will be open belongs in the goal markets; a view about which side is stronger belongs in the result markets. Forcing one into the other is where most avoidable losses come from.
Every selection shows the model probability for that outcome in that fixture, a confidence band for readability, and the price the pick was published at. The probability is about that fixture only. It is never a claim about how often our selections win.
Each market page carries a written explanation of what the market is, how we model it, when it tends to be worth backing and the mistakes readers most often make with it.
A fixture appears on exactly one market page, because our procedure publishes one market per match. That is a deliberate constraint rather than a limitation of the data — we could generate a number for every market on every fixture, and so could anyone else. The question is whether the number would mean anything.
Publishing everywhere is what produces the familiar wall of predictions that covers every outcome and commits to none. It looks comprehensive and it is unfalsifiable, which is precisely why it is common.
So if you are looking for a specific fixture and it is not on the page you expected, the fixture is on the page its shape fits — or, where the feed had no usable data, it is not published at all.
The result family answers who wins. 1X2 names one outcome of three; double chance covers two at a shorter price; the draw isolates the outcome readers discount most often.
The goal lines answer how many, on a ladder: over 1.5, over 2.5 and over 3.5 with their mirrors at under 1.5, under 2.5 and under 3.5. The right rung is the one where the gap between our probability and the price is widest, not the one with the most attractive number.
Both teams to score answers whether both attacks function, independently of the result. Correct score asks for the result and the goal count together. The first-half lines, over 0.5 and over 1.5, settle at the interval and are modelled on first-half data rather than scaled-down full-match figures.
The most reliable way to use these pages is backwards from what you actually think. A view about which side is stronger belongs in the result markets. A view about whether a match will be open belongs in the goal lines. A view about both attacks functioning belongs in both teams to score. Forcing one kind of view into the wrong market is where most avoidable losses come from.
If you have no particular view and are looking for where the numbers disagree with the market, start at value bets instead — it surfaces the gap directly rather than asking you to find it.
Every market page carries a written explanation of what the market is, how we model it, when it tends to be worth backing, and the mistakes readers most often make with it. The methodology page covers the procedure that sits behind all of them.