Why These Markets Exist and Who They Are For
Corners and cards are football's second-order markets. They are not predictions of scoring, they are predictions of how a match was played, and that makes them behave differently in both good and unhelpful ways.
The behaviour that makes them interesting is that they are strongly connected to the shape of a match rather than the quality of the teams. A team defending a one-goal lead at 70 minutes will produce more corners than it did when it was attacking, and a team going behind will produce more cards. These are conditional facts that are visible pre-match too, which is where the modelling value sits.
The behaviour that makes them frustrating is the margin. Both markets are less efficiently traded than the result, both have thinner liquidity, and both attract a specific kind of bettor: the accumulator builder who likes short prices on things that feel like trivia. Those short prices are what funds the margin.
That combination, interesting inputs and poor pricing, means corners and cards reward you for patience rather than cleverness. Treat them as a side market, a research field, and an in-play tool. Our football betting strategies page covers the wider framework, and the best betting sites comparison is worth checking for which books actually offer a full corners board.
Corner Totals: The Core Market
The main corner market is a two-way over and under on a line, usually ending in .5 so there is no push. Nine and a half is the standard line on a mainstream fixture, with ten and a half available at most books.
Counting is the first thing to get right. A corner is awarded when the ball, having last touched a defender, crosses the goal line without entering the goal. It is counted from the moment it is awarded, not from the moment it is taken. A corner that is short and results in a goalmouth scramble and another corner is two corners. Most data providers count it the same way, but not all, and a provider that counts taken corners rather than awarded corners will systematically under-report by about 0.4 per match.
Corners are counted at the time they are awarded, so a corner in the third minute of added time still counts, and a corner that is awarded and then the match is abandoned still counts on most sportsbooks because the count at abandonment stands. The 2018-19 season had several matches abandoned with corners already taken and the resulting over 9.5 settled, which surprised a lot of people who expected a void.
Corner and Card Prices With the Margin Exposed
Illustrative over-prices and the implied probability once the margin is stripped out. The two right-hand columns are the ones worth doing in a spreadsheet, because they are what you compare against your own estimate.
| Market | Typical over price | Implied prob. (naive) | Two-way margin | Break-even after margin |
|---|---|---|---|---|
| Match result (1X2) | 2.10 / 3.40 / 3.60 | 27.2% | ~4% | 25.4% |
| Corners over 9.5 | 1.85 | 54.1% | ~7% | 50.6% |
| Corners over 10.5 | 2.05 | 48.8% | ~7% | 45.4% |
| Corners under 12.5 | 1.75 | 57.1% | ~7% | 53.5% |
| Corner handicap -1.5 | 1.95 | 51.3% | ~6% | 48.2% |
| Total cards over 4.5 | 1.72 | 58.1% | ~9% | 53.4% |
| Total cards over 5.5 | 2.30 | 43.5% | ~9% | 39.4% |
| Half-time corners over 4.5 | 1.95 | 51.3% | ~10% | 46.1% |
| Booking points over 34.5 | 1.80 | 55.6% | ~9% | 50.7% |
| Cards in first half over 1.5 | 1.75 | 57.1% | ~10% | 51.6% |
| Home most corners | 1.85 | 54.1% | ~6% | 50.8% |
Corner Handicaps and Why They Are the Better Market
Corner handicaps ask which team will take more corners, with lines typically running from -6.5 to +6.5. They are the sharpest product in the corners family and almost nobody bothers with them.
The reason is structural. A corner total forces you to predict the sum of two teams’ corner generation, which means you have to model both attacks and both defences at once. A corner handicap asks a relative question, and relative questions are much easier to forecast because team strength differences cancel out to a degree. A team that is stronger in every dimension will create more corners and concede fewer, and the handicap already knows that.
Settlement follows the standard rules, with push at the integer line and half-win or half-loss at the quarter line. A -1.5 handicap needs a two-corner win. A -1 line pushes on exactly one, wins on two or more, loses on zero. Our handicap guide covers the mechanics, and the Asian handicap page covers the quarter-line splits.
There is a real edge in the relationship between the result market and the corner handicap. When a home favourite is priced 1.40 and their corner handicap is -3.5 at 1.90, the implied corner share can be compared against the implied win share. Large divergences are a signal that one of the two markets is mispriced, and the corners board is usually the softer of the two. The margin guide explains how to work out which is which.
First Corner, Last Corner and the Short Markets
Three more markets that behave very differently from the totals, and two of them are traps.
First corner. Usually priced by a moneyline on which team takes it, sometimes with a tie or no-corner option. The first corner is heavily influenced by the opening tactic. A team that starts by pressing high and going forward concedes corners almost immediately; a side that starts by building from the back will rarely give one away in the first ten minutes. First corner prices are often derived from the total corners line rather than being independently modelled, which makes them soft in an exploitable way. The trap is assuming the first corner is random. It is not; it clusters on the team more likely to attack, which is the favourite roughly two thirds of the time.
Last corner. This is a much worse bet than it looks. Last corner is heavily determined by trailing teams chasing a goal in the final ten minutes, which means it correlates with the match result more strongly than most bettors expect. It also happens in stoppage time, where a single attacking substitution or a red card changes it entirely. Any edge in last corner is mostly variance.
Corner race handicaps, which is a first-N-corners or next-corners-to-go market, are the most interesting of the three for in-play use. They have low margins in some books and they respond sharply to the game state, which makes them a genuine trading surface rather than a coin toss. The sports trading guide covers how to approach them with a price rather than a hunch.
Card Markets: Yellows, Reds and the Referee Effect
Cards split into several distinct products and they are priced very differently.
Total cards over and under. The most popular card market and the most efficient of them, because the over and under pair gets traded. A line of 5.5 is standard on a Premier League fixture. Cards are counted from the moment they are shown, so a second yellow that becomes a red is two cards plus the red booking on some sportsbooks and three in total, and the treatment is not universal. The Premier League data convention counts a dismissal as a red and not as two yellows plus a red, so a total cards line built on PL data will run about 0.3 lower per match than one built on continental data. That mismatch alone creates arbitrage between bookmakers on some fixtures, and it is worth checking before you assume a price is wrong.
Team cards. Which team collects more. A card handicap of -1 at 1.85 is a standard product. Team card markets are sharper than card totals for the same relative reason as corner handicaps: they are a question about one side's behaviour rather than a sum. They are also more sensitive to the referee, which cuts both ways.
Referee card markets. Some bookmakers offer a specific referee's card total or a specific referee to be carded. The prices are computed by taking an average referee card rate and adjusting for the teams involved, which is crude but occasionally exploitable. The arithmetic that matters: the average Premier League referee produces about 4.1 cards per match with a standard deviation of roughly 1.6, so a referee priced at a 6.5 card rate is being asked to be well outside his own typical distribution. Most of those prices are wrong by a small amount in a consistent direction.
Why Corners Track Defensive Shape, Not Attacking Quality
Corners are a defensive statistic that gets recorded at the attacking end. That sentence explains almost everything about how to price them.
A corner is produced when a defender puts the ball out of play. The two biggest causes are blocks and attempted clearances. Both are functions of how the defending team plays, not how the attacking team passes. So a team that sits in a deep back five, concedes possession, and clears everything that comes into the box will generate a great many corners for their opponent while creating very few themselves. The same team playing a possession side that builds through the middle will concede almost none.
The out-of-possession shape is the single most predictive variable, and it is stable across a season and mostly predictable from a club's recent tactical pattern. Teams that have used a back five in at least four of their last six matches concede a materially lower share of the corners. The market tends to price this roughly one game late, which is a real and repeatable edge on the corner handicap market rather than the total.
Second-order effects matter too. A team that loses the ball in high numbers in their own third will concede both corners and shots on target from the same source, so the corner share and the expected goals against are correlated. That correlation is why the corner handicap on a match where one side has a much lower expected goals against figure is usually about right, while the corner total in the same match is usually a coin toss. The expected goals guide covers how to build that defensive side properly.
Booking Points and the Smoother Alternative
Booking points price yellows as one point and reds as two, which turns a lumpy distribution into a smoother one and is the most sensible product in the card family.
The reason is variance. A total cards line of 5.5 is decided by a single red card in a 61st minute, and red cards occur at roughly 0.13 per team per game. Booking points at 34.5 are also affected by that red, but only by two points rather than a full card, so the tail is much thinner. If you are modelling rather than gambling, booking points is a better target because the distribution is closer to normal and the standard deviation is more stable.
Booking points also cross over nicely with cricket. The format is borrowed directly from cricket betting, where total match points work the same way, and the same arithmetic applies. The cricket betting strategy page covers that version, and the bet builder guide covers how a booking points leg behaves inside a same game multiple.
The weakness is a shared blind spot: booking points markets and card totals markets on the same fixture are both derived from the same card count, so a cross-book discrepancy between them is almost always a settlement or counting convention difference rather than free money. Check the red card treatment before you act on a gap.
Where the Margin Is Widest on the Board
If you take one thing from this page, take this ranking. It is ordered by how much the market overcharges you.
Widest, 9 to 12 per cent. Half-time corners, cards in a named half, corner race markets, and anything on a lower-division or women’s fixture. These are close to unpriceable and rarely worth the effort.
Wide, 7 to 9 per cent. Total cards and booking points, half-time cards, and the popular corner totals on non-mainstream leagues. Workable if you have a model, poor value if you are guessing.
Moderate, 6 to 8 per cent. Corner totals on mainstream league fixtures. The over 9.5 market is better priced than most bettors assume, but you still need roughly a three-point edge to break even.
Sharpest, 5 to 6 per cent. Corner handicaps and team card handicaps on major league fixtures. Thin margin, thin liquidity, and the best available price on this board if you do the work. The bookmaker rating methodology explains how we test which books carry which lines, and the arbitrage guide covers what to do when the gap between books is larger than the margin.
Seven Mistakes on Corners and Cards
These recur constantly, and several of them are about the shape of the data rather than the price.
- Treating corners as an attacking statistic. They are produced by defending. Model the defending shape, not the attacking talent.
- Using a league average from a different competition. A team that concedes six corners in Serie A does not concede six in the Eredivisie. Normalise to the fixture's own league.
- Trading card totals against booking points without checking the red card rule. The counting conventions differ between data feeds and the apparent edge is a data artefact.
- Betting the last corner. It is a heavily result-dependent, stoppage-time-corrupted coin toss with no genuine edge.
- Assuming the first corner is random. It clusters on the attacking side about two thirds of the time, and the price often ignores that.
- Chasing short prices because the market looks safe. A 1.50 on cards over 3.5 at 10 per cent margin needs a 63 per cent hit rate, and nobody has that.
- Ignoring the referee on team card markets. Two neutral referees on the same weekend can differ by nearly a card per team per game, which is larger than most form edges.
A Practical Routine for These Markets
How I would actually spend time on corners and cards without wasting it.
One. Ignore the totals. Start with the corner handicap and the team card handicap, where the margin is thinnest and the relative question is easier to answer.
Two. Build the defensive-shape input first. For each team, note the out-of-possession formation over the last six league matches and classify it as a back three, back four or back five. This takes about twenty minutes per team and it is the input that pays.
Three. Build a per-league baseline for corners conceded and cards collected over a full season, not a rolling window. Season data, twelve months minimum.
Four. Apply the referee, for card markets only, using the published average card rate for that referee and the two teams' historical card rates against that referee specifically if you can get them.
Five. Compare your estimate to the price after stripping the margin, and require a gap of at least four points. Anything less is noise at these margins.
Six. Log every one, priced and unpriced. The journal method page covers the format, and after a hundred corners bets you will know whether your edge is real or whether you are simply betting on variance, which is the only honest way to find out.
The Summary
Corners and cards are second-order markets that track the shape of a match rather than the quality of the teams, and they carry a 6 to 11 per cent margin against 4 per cent on the result. Corners are produced by defending, so the out-of-possession shape is the input worth modelling. The corner handicap and team card handicap are sharper than the totals and worth the effort. Booking points smooth the red card tail and behave better than a total cards line. Half-time corners and anything on a second-tier competition are the widest prices on the board and usually not worth a minute of your time.
