What each column measures

Every number on the StatGol panel, what it measures and — above all — what it does not mean. That last part tends to be missing from every explanation of live football statistics, and it is the reason this page exists.

How the number gets here

StatGol reads a live data source every 30 seconds and computes on top of it. It is worth knowing which is which, because it changes what you can ask of each number.

Match counts — score, shots, shots on target, attacks, dangerous attacks, corners, possession, cards, fouls, offsides, substitutions, xG, odds and the raw minute-by-minute dominance index — come from watching the match. StatGol does not watch the game itself: if a shot is counted wrong at the source, it arrives wrong here. That is a stated limit, not an excuse, and it is why this page says of each number what it can carry.

Computed inside the app: the per-minute rhythm, the pressure graph, the pressure over the last 15 minutes, the acceleration index, the per-half split of dangerous attacks and shots on target, and every rule about when to hide a number — which is the most laborious part and the least visible.

The full formulas are on the methodology page, with the measurements behind them and a mistake of our own published with the correction next to it.

Before the entries

Every cell holds two numbers

The top one is the home team, the bottom one the away team. Where there is a single number, it belongs to the match, not to a team.

A dash is not a failure

A dash instead of a number means one of three things, and in all of them the panel would rather show nothing than show something that does not hold up:

• The league does not have that statistic. Smaller leagues arrive without xG, without dangerous attacks or without shots. When a zero comes from a league with no coverage, the panel turns it into a dash too: zero and “not measured” are different things.
• Little has happened in the match so far, and the calculation would divide by a number too small to mean anything.
• The window has just crossed half-time. That is the case right below.

The rule at the start of each half

The “last 5 minutes” and “last 15 minutes” versions of some statistics are truncated at the start of each half: the reference point never goes back before the whistle of the current half.

At 47 minutes, therefore, the 15-minute window does not hold fifteen minutes — it holds the two that have run since half-time, exactly like the 5-minute one.

That is why the panel does not publish metrics derived from these windows before 6′ and 16′ in the first half, and before 51′ and 61′ in the second. It applies to attacks, dangerous attacks, shots on target, shots off target and corners. The measurement behind the rule is on the methodology page.

Attack

Attacks

Attacking moves recorded without any judgement of danger. It works as a denominator: many attacks with few dangerous attacks describes a team that reaches the opposition half and stops there.

What it does not mean. It measures neither possession nor control. A counter-attack and a move that dies in midfield count the same.

Dangerous attacks

The move classified as having reached the edge of the box or ended in a shot. It is the most used column on the panel, and it shows who is pressing better than possession does.

The smaller number underneath is the rhythm: dangerous attacks from both teams added up, divided by the minutes played. It lets you compare matches at different points.

What it does not mean. It is not a shot, not a chance and not a probability of a goal. A team can pile up forty dangerous attacks without a single shot. Much of the market treats this number as a predictor; StatGol publishes no measure of how much it predicts, because it does not measure that.

In depth: what dangerous attacks are, how they are classified and why the rate says more than the total.

Pressure (the graph)

The difference in dominance between the teams, minute by minute. The reading arrives every 30 seconds with one value per team; the app averages home minus away per minute and draws it: the bar rises when the home team is closer to goal and falls when it is the away team. The dots mark the goals.

Above 6 the bar saturates on purpose: the graph is there to read the match, not to record records.

What it does not mean. A tall bar is not an imminent goal. It describes what has just happened on the pitch.

Pressure (last 15′)

The average of that same index, per team, over the last 15 minutes of match clock. It shows who is on top now, not over the whole match. Half-time does not count, because the average uses the clock on the readings themselves, not the real clock.

What it does not mean. It is not possession, and the two numbers do not add up to a hundred. They are two independent indices: both teams can be pressing.

Possession

The percentage of possession for each team.

What it does not mean. Possession is not pressure. A team passing the ball around its own half piles up possession without producing anything — and the pressure graph shows that while this column does not.

Acceleration (5′ / 15′)

The rhythm of the recent window divided by the rhythm of the whole match. There are two numbers: the 5-minute window and the 15-minute one. 1.0 means the match is keeping its own rhythm; 1.5, fifty per cent above; 0.6, forty per cent below.

The calculation, in full. Activity at each end is counted as dangerous attacks plus 3 times the shots on target — a shot on target weighs 3 dangerous attacks, and that weight is a choice StatGol made, not a constant of nature. The rhythm of the window is then divided by the rhythm of the match.

When it does not appear. With less than 8 of accumulated activity, already weighted, the index is not published: dividing by a very small number produces enormous swings from a single incident. It also stays hidden while the match has fewer minutes than the window, and during the minutes covered by the start-of-half rule.

What it does not mean. High acceleration describes what has just happened, not what is coming. A match can speed up because one team needs the goal, and just as easily because the other stopped marking.

Shots

Shots

Total shots: on target, off target and blocked.

On target

Shots that forced a save or went in.

What it does not mean. It does not measure the quality of the chance. A thirty-yard shot into the keeper's hands counts the same as a header from the six-yard box. For quality, the column is xG.

In depth: what counts as a shot on target, why hitting the woodwork does not, and what accuracy hides.

On target (last 5′) and On target (last 15′)

The same, restricted to the clock window.

What it does not mean. In the first minutes of each half these numbers exist, but truncated — which is why the panel hides them instead of showing them correct and misleading. That is the start-of-half rule.

Off target and Accuracy

Shots that did not go on target, and the ratio of shots on target to total shots.

What it does not mean. High accuracy from few shots is not quality: it is a small sample. Two shots, both on target, make a hundred per cent.

xG (expected goals)

What a team's shots would be worth in goals, given the quality of the chances: the sum of the goal probability of each shot. An xG of 1.8 means the chances created would, on average, yield almost two goals. Not every league has this data.

Here the model matters. xG arrives already calculated, with the model of whoever calculates it. Different models give different numbers for the same match, and that variation is nobody's error.

What it does not mean. xG is not a forecast of the result. It is a description of the quality of the chances that have already happened. A team on xG 2.5 with no goals is not “owed” a goal: xG has no memory and collects no debt.

In depth: what xG is, how the number is built and why different sources give different values.

Corners

Corners and Corners (last 15′)

Corners taken, over the match and in the recent window. The window follows the start-of-half rule.

What it does not mean. A corner is a consequence of pressure, not a measure of it. A team that crosses a lot piles up corners without creating chances.

By half

Columns switched off by default, separating the first half from the second.

Dangerous attacks and shots on target by half are computed by the app, by following the match minute by minute. That is why there is one case where they do not exist: a match that enters the list already in the second half — the panel did not see the first, and the cell shows a dash saying so. Goals and corners by half do not have this problem.

What it does not mean. A busier second half than first is a fact about those two halves, not a trend for the minutes ahead.

Timeline

Goals and cards in minute order, drawn on the line.

Disallowed goal. An incident recorded as a cancelled goal, which does not enter the score. When the source flags the cancellation, the panel marks the point with a grey cross on a dashed stem, never with a dot, so it cannot be confused with a valid goal. In practice that flag is rare: we measured a full day of matches and it never came once, so the usual case is that the disallowed incident simply never enters the score.

A stated limitation. Some leagues arrive without incident coverage, where the score shows a goal and the timeline carries none. In those cases the panel would rather not answer “0 goals in the last 15′” than invent a number.

Pre-Live and in-play odds

Pre-Live

The odds from before kick-off, at the last price available. They serve as context for what the market expected of the match. “In-play” is the current price.

Goals line and corners line

The market's reference number, for example 2.5 goals. It serves as context for what was expected of the match, not as a recommendation.

Asian handicap

The advantage the market gave one of the teams. Context as well.

What it does not mean. It is not an odds comparison and it is no use for finding the best price — the prices come from one bookmaker, with another as backup when a price is missing. There is no bookmaker link anywhere on the site, and StatGol takes commission from none of them.

Recent form

The last 5 matches of each team: points, goals scored and conceded, the percentage of matches over 1.5, 2.5 and 3.5 goals, and the percentage in which both teams scored.

What it does not mean. Five matches is a small sample, and the opponents were not the same. A percentage over five matches moves twenty points with one result. These numbers describe the recent past and say nothing about this match — they are here as context, like the odds.

Discipline

Yellow cards, fouls, offsides and substitutions, per team.

A red card is not a column: it appears next to the team name, because it changes the match and has to be seen without looking for it.

Lit cell

The cell that lights up when a match meets a condition you set in the My rules menu. It is not a recommendation of anything: it is your own number asking for your attention.

Rules that depend on the 5- or 15-minute windows do not light up at the start of each half, while the window is not yet as long as its name says.

What no column here means

This section closes the page and is the most important part of it.

• No column predicts anything. All of them describe what has already happened. There is not a single number on this screen that estimates what is going to happen — not even xG, the one most often mistaken for a forecast.
• No column is a betting signal. StatGol gives no tips, publishes no strike rate and recommends no bet. When a cell lights up, it is because you set the condition in the My rules menu. The rule is yours.
• Nothing here is per player. Everything is per team.
• There is no history for backtesting. The panel shows what is happening now.
• There is a delay. Between the incident on the pitch and the screen there is somewhere between a few seconds and a minute, counting the recording of the incident and the reading cycles. StatGol is not a live stream and is no substitute for watching the match.

To start using the panel, see the help page. For the formulas and the measurements behind them, the methodology.