Most football betting arguments are really arguments about which numbers to trust. Possession, shots, xG, league position — they all describe a match, but only some of them tell you anything about the next one. This piece separates the stats that predict from the stats that decorate broadcast graphics. If you're still getting comfortable with how football odds are priced in the first place, read the football betting guide first.
What does xG actually tell you — and when does it lie?
Expected goals (xG) is simpler than its reputation. Every shot gets a value based on how often shots like it — same distance, same angle, same body part, same kind of assist — have been scored historically. A tap-in from two yards might be worth 0.9. A hopeful strike from 30 yards might be worth 0.03. Add up a team's shot values and you get a measure of the quality of chances it created, separate from whether anyone finished them.
Why bother? Because goals are rare. A typical match produces between two and three of them, which means results swing on a deflection or a goalkeeper's good afternoon. Chances are far more frequent than goals, so xG builds its picture from much more data. Over a full season, a team's xG difference tells you more about how it will perform next than its goal difference or its league position does.
The limits matter just as much. Single-match xG can mislead badly: 2.1 xG built from twenty long shots describes a worse afternoon than the number suggests, while 1.4 xG built on two clean one-on-ones is a different match entirely. The model doesn't know the score state — teams protecting a lead stop creating on purpose — and it treats a packed penalty area much like an open one. Over three or four matches, xG is barely better than noise. Give it ten or more before you lean on it.
If possession doesn't predict results, what does?
Possession is a style measurement, not a quality measurement. A team can hold 65% of the ball by passing between its centre-backs while the opponent sits in a low block, perfectly comfortable. That sterile control wins nothing, and betting markets stopped being fooled by it years ago, even if television panels haven't.
The numbers that carry signal measure penetration: shots, shots on target, touches inside the penalty area, completed passes into the box. A team producing those week after week is doing something repeatable. A team dominating the ball without producing them is often just losing slowly.
Finishing is the other trap. Teams that score well above their xG for a stretch — every half-chance flying in — tend to drift back towards the model, and so do teams whose forwards can't buy a goal for two months. Finishing runs regress; chance creation persists. A handful of elite finishers beat their xG year after year, but the margin is small and the list is far shorter than fans believe.
| Stat | What it really measures | When it becomes reliable |
|---|---|---|
| Goals / league position | Results, including luck | Half a season or more |
| Possession % | Style of play | Never predictive on its own |
| Shots and box entries | Sustained pressure | Around 6–10 matches |
| xG difference | Chance quality, both ways | 10+ matches; a season is better |
| Goals minus xG | Mostly finishing luck | Assume it fades |
How much is home advantage still worth?
Home advantage is real, but it is not what it was. In the early decades of league football, home sides won roughly half of all matches. That share has fallen steadily across every major European league, decade after decade, and now sits closer to the low-to-mid 40s in most of them. Better pitches, easier travel and professional preparation have all chipped away at it.
The matches played in empty stadiums in 2020 and 2021 settled an old argument: a meaningful slice of home advantage lives in the crowd, and a good chunk of that runs through the referee. Without fans, home teams won less often, and the usual gap in cards and decisions between home and away sides shrank.
None of this means you should bet against home teams on principle. Bookmakers price home advantage into every match and update it faster than folk wisdom does. The mistake to avoid is the old reflex — never back against the home side — which was built on a version of football that no longer exists.
Why does the early-season league table lie?
Six matches into a season, the table is close to worthless as a predictor. The sample is tiny, fixture difficulty is wildly uneven — one team has faced three title contenders, another has faced three promoted sides — and finishing luck hasn't had time to wash out. Every season produces an early surprise leader whose underlying numbers were ordinary all along, and a slow-starting giant whose xG said the goals were coming.
Context across leagues matters too. Average goals per match differ noticeably between leagues, and those gaps persist season after season. A team's scoring record means little until you know the league baseline it was produced in — exactly the adjustment that matters in over/under goals betting.
Before backing a team on reputation or table position, a short checklist:
- xG difference over the last 10+ matches, not the last three
- Shot and box-entry volume: rising, flat, or declining?
- Fixture strength: who were those results actually against?
- Goals minus xG: is this team riding a finishing streak that will fade?
- League context: what does that scoring record mean at this league's baseline?
None of these stats produce certainties. They shift probabilities a few percentage points at a time, and a few points is all an edge ever is. Turning that into profit is a staking problem as much as a reading problem — covered in football bet sizing and bankroll strategy — and the fundamentals of football betting sit underneath all of it.