Death overs field setting: how the last four are built

The sixteenth over ends and the captain spends the next half a minute walking two fielders about fifteen metres each, one deeper and squarer, one round to the other side of the ground. Nothing else changes. No bowler has been swapped, no message has come from the dressing room, and the batter has not moved. That thirty seconds of walking is death overs field setting, and it fixes the range of things that can happen next more tightly than the deliveries themselves will.

Here is the whole problem in one sentence. From the seventh over onwards a fielding side may place five men outside the fielding circle, which leaves four inside it once the bowler and the wicketkeeper are accounted for, and those nine outfield bodies have to cover a circle whose boundary at international level can be anything from about 59 metres to about 82 metres from the middle of the pitch. Nine fielders cannot cover that. The captain is not choosing where to put fielders. He is choosing which shots to allow.

Everything below follows from accepting that framing rather than fighting it. The bowling half of the same problem, the deliveries themselves and the mental load of executing them, is worked through separately in the article on what makes the last four overs a distinct bowling discipline. This piece stays on the other side of the equation: the nine men who are not bowling.

• 5Fielders permitted outside the circle after the powerplay
• 2Fielders permitted behind square on the leg side
• 30Radius of each half of the fielding circle, in yards
• 4Fielders remaining inside the circle

Limits taken from the Laws of Cricket and the ICC men's T20I playing conditions rather than from match data. The inner figure follows arithmetically from the five-fielder limit once the bowler and wicketkeeper are removed from the eleven.

Five out, four in, and the sum that never balances

Take the eleven apart and the shape of the constraint becomes obvious. One man is bowling. One is keeping wicket. Of the nine left, five are permitted beyond the circle, so four remain within it. Those four are the entire infield: the men who stop the push into the covers, the men who cut off the nudge to midwicket, the men who might catch a leading edge.

Four fielders inside a circle marked by two semicircles of thirty yards from the middle stump at either end, joined by straight lines, is a thin covering. A batter looking at that ring can see three or four routes through it without moving his feet. Which is precisely the point of the restriction, and precisely the reason the death overs produce the scoring they do.

The captain therefore inherits a fixed budget and an unfixed problem. He has nine placements and something like eight distinct scoring directions to worry about, each of which can be attacked in two ways: along the ground through the ring, and over the ring into the deep. Nine cannot cover sixteen. So the question a good captain asks is not "how do I stop runs", it is "which two runs am I happy to concede in order to make the fourth and the sixth unlikely".

That reframing changes the answers. A field that tries to stop everything stops nothing well. It leaves each fielder in a compromise position, close enough to the rope to be useless for a single and far enough inside it to be jumped. The best death fields look lopsided on television, because they have decided something.

There is a second budget nobody counts, which is time. The stop clock now gives a fielding side sixty seconds between overs in international limited-overs cricket, which is the entire window in which a captain can rearrange nine men, speak to his bowler and change his mind. Fields at the death are therefore mostly pre-agreed in the innings break, adjusted by one or two placements between overs, and adjusted again by a single walked step during the over itself.

A catcher is a bet on a wicket, a saver is a bet on a single

The four inside-the-circle slots are contested by two very different kinds of fielder, and the choice between them is the first real decision in the sequence.

A catcher is placed to take the ball. He stands close, square or slightly in front of square, at a distance where a leading edge, a toe-ended slog or a checked drive will reach him before it bounces. He is trying to end the innings of the man on strike. A short cover, a short midwicket and a straightish mid-on are all catchers in this sense; they exist because a batter swinging hard at a slower ball produces catchable balls at a rate no other phase of the innings matches.

A saver is placed to stop the ball. He stands on the edge of the ring, usually square or behind square, and his job is to convert a well-struck push into one run rather than two. He will almost never take a catch there. He will touch the ball three or four times an over and each touch is worth a fraction of a run.

Those two are competing for the same four slots, and the arithmetic that decides between them is unforgiving. A catcher who takes one chance in an over has produced an enormous outcome. A catcher who takes none has, on most deliveries, been a hole in the ring through which a two became available. A saver produces a small, certain, repeated gain.

At the death the saver usually wins, and the reason is the one that governs the whole phase: a wicket in the eighteenth over is worth far less than a wicket in the eighth, because the batter it removes was going to face perhaps four more deliveries. Buying a small chance of a low-value wicket by conceding a certain flow of twos is a bad trade. It is a good trade in the powerplay, which is why the fields in the first six overs look nothing like the fields in the last four, and why the shift between them is one of the sharpest visible changes in the game. The way that first restriction shapes the opening phase is worked through in the piece on how the six-over restriction sets up the innings.

There are two situations where the catcher comes back. The first is a new batter, who has not seen the pace, has not set his base and is being asked to hit from ball one, and whose miscue rate for two or three deliveries is genuinely high. The second is a side defending a small total, where a wicket is worth a great deal because the batting order behind the man on strike is thin. Outside those two, a catcher at the death is a captain who has not done the sum.

Where a boundary rider actually stands, and why it is never on the rope

The five men outside the circle are usually called boundary riders, which is a slightly misleading term because a good one does not stand at the boundary.

A fielder pressed flat against the rope has surrendered half his range. He can move forwards and sideways, but any ball that lands short of him drops safely, and any ball that passes over his head is six whether he is there or not. His only useful function in that position is to prevent a ball that would have carried a metre or two past the rope, and he catches those far less often than the broadcast suggests.

A fielder standing a handful of metres inside the rope has bought back the whole of his forward range. He can sprint in for the mishit that falls between the ring and the deep, which is the most common way a death over concedes a boundary that nobody expected, and he can still turn and chase a genuine strike, because a well-hit ball travelling over him was clearing him from any position. The cost of being a few metres in is a small number of balls that go over his head and would have been caught. The gain is a much larger number of balls that drop in front of him and would have run away for four.

There are exceptions, and they are matchup-driven rather than general. A batter who has cleared the rope twice in an innings by a narrow margin is worth standing under, because the distribution of his hitting is stacked right on the boundary line rather than well beyond it. A very short straight boundary changes the sum in the same direction. And a fielder with an exceptional relay or a partner beside him on the rope can stand deeper than he otherwise would, because the catch has become a two-man operation.

The other thing a rider is doing, which is invisible unless you watch him rather than the ball, is angling. He is rarely square to the batter. He stands a fraction round towards the side the plan is protecting, so that the ball he most expects arrives on his stronger side and the ball he least expects is the one he has to reach for. Move the bowler's line and every rider shuffles two or three metres. That shuffle is the field re-expressing the plan.

63%38%

• Arcs with a fielder standing in them5
• Arcs left open3

A constructed illustration of the constraint rather than a measurement of any match. Divide the boundary into eight equal arcs and a captain with five outfielders can put a body in five of them, which leaves three arcs defended by nothing except the bowler's line and length.

Show the numbers

Worked example: the outfield cut into eight arcs at the death

Item	Value
Arcs with a fielder standing in them	5
Arcs left open	3

The square boundary decides the plan before the captain does

Grounds are not circles and the Laws do not require them to be. The ICC's playing conditions for men's T20 internationals set an outer limit of 90 yards and an inner limit of 65 yards measured from the centre of the pitch in use, which is a spread of nearly 23 metres between the longest legal boundary and the shortest one. Very few grounds sit at either extreme, but almost every ground is asymmetric, and many are dramatically so, particularly where a drop-in pitch has been placed off the centre of a square.

A short square boundary is the single most influential physical fact in a death over, and it is decided before either captain has thought about it.

The instinct is to defend the short side heavily. Put three riders over there, the reasoning goes, and the batter cannot get the ball past them. The reasoning is wrong, and the way it fails is instructive. Five riders defending a boundary is already thin. Committing three of them to one side leaves two to cover everything else, which means the long side of the ground, where a batter has to hit the ball twenty metres further, is now the easiest place on the field to find four runs. The captain has spent his scarcest resource protecting the shot that costs six and handed away the shot that costs four, and the batter takes the four every time because it requires no risk at all.

The workable answer is to price the short side rather than seal it. Two riders on the short square boundary, positioned to make the straight hit and the long-side hit the cheaper options, and a bowler instructed to bowl the line that makes those options natural. Against a right-hander with a short leg-side boundary, that usually means a wide line outside off, which asks him to hit against the shape of the ground. He can still go to the short side, but he has to fetch the ball from outside off stump to do it, which is a harder shot with a lower success rate.

The corollary matters as much. On a ground with a long square boundary and short straight ones, the whole plan inverts: the straight hit becomes the thing to price, the fine leg and third areas become live, and a bowler aiming wide of off stump is now sending the ball towards the part of the ground the batter most wants. The same bowler with the same skills bowls a different over at two different grounds, and a captain who has not walked the boundary before the toss is guessing. That is one of the small ways the ground itself, rather than the pitch, shapes a result, alongside the surface questions covered in the article on how different pitch types behave.

Two behind square on the leg side, and the shot that broke the field

There is one placement restriction that operates in every format and at every point of an innings, and it is a Law rather than a competition regulation: at the instant of delivery no more than two fielders, other than the wicketkeeper, may be behind the popping crease on the leg side. Exceed it and the umpire calls a no ball.

That limit was written to stop leg theory, a line of attack from a different era, and it now does something its authors never considered. The leg side behind square is exactly where the modern death batter has learned to score. The ramp, the scoop and the paddle all take a full ball and redirect it into the arc between fine leg and the keeper, and they work because the batter is using the bowler's pace rather than his own strength, which means the ball travels the same distance whether he is set or not.

A captain who wants to shut that arc down cannot. He can post a deep fine leg and a deep backward square leg and that is the end of his allowance. He has two fielders for an arc that a good batter can direct into anywhere across ninety degrees, and the shot itself is close to impossible to field because it goes over the ring rather than through it.

So the leg side behind square is the one part of the ground where the field is genuinely capped by regulation, and the tactical consequences run in every direction. It is a large part of why the wide yorker exists, because a ball outside off is much harder to scoop. It is why some captains push their two permitted men very fine and accept the square boundary going unguarded, on the reasoning that the scoop is the more likely shot. And it is why a keeper standing up to a seamer at the death, which looks eccentric, is occasionally sensible: a batter who cannot pre-move deep in his crease finds the scoop harder to reach.

Batters read the restriction as fluently as captains. A batter who sees a vacant fine leg and a deep square leg knows the scoop is on and the sweep across the line is not. The full grammar of where each position sits and what its name tells you is set out in the guide to reading the fielding map.

Shortest permitted boundary59.43m

Longest permitted boundary82.29m

Gap between the two limits22.86m

The outer and inner limits set out in the ICC men's T20I playing conditions, measured from the centre of the pitch in use. Domestic grounds and club grounds are not bound by them, so the real spread across franchise cricket is wider still.

Show the numbers

The permitted range of a boundary in men's T20 internationals

Item	Distance from the centre of the pitch
Shortest permitted boundary	59.43m
Longest permitted boundary	82.29m
Gap between the two limits	22.86m

Setting the field to the batter's base rather than to the bowler

A weak field is set to the bowler. A strong one is set to the specific person holding the bat, and the information the captain uses arrives in the two or three deliveries before he sets it.

The first thing to read is the base. Where does this batter stand at the moment of release, and where has he moved to by the time the ball arrives? A batter who backs towards leg stump is opening the off side and telling you he intends to hit through the covers or over extra cover, which makes deep cover and long off the two placements that matter and makes deep midwicket a partial waste. A batter who walks across his stumps towards off is aiming to work the ball behind square on the leg side, into the arc the Law has already capped, and the field cannot follow him there. Against him the answer is a line he cannot reach rather than a fielder he cannot beat.

The second thing to read is the shape of the arc. Some batters generate power almost entirely through a vertical swing, hitting straight down the ground and clearing long on and long off. Others are horizontal-bat hitters whose value comes square of the wicket. The two produce completely different fields from the same five outfielders, and the mistake most amateur captains make is to set a spread field that covers both badly rather than a lopsided one that covers the likelier of the two well.

The third thing is handedness combined with the bowler's angle, which is where the matchup actually bites. A left-arm seamer slanting the ball across a right-hander is pushing every scoring option towards the off side, so the off-side riders should be deeper and squarer and the leg-side ones can be pulled straighter. Reverse the handedness and the same bowler is angling into the batter's pads, the leg side opens up, and the field should rotate accordingly. A captain who does not move his fielders when the strike rotates between a left-hander and a right-hander has left half his field pointing at the wrong batter.

The fourth thing, and the one most often skipped, is what the batter has already shown. If the last two boundaries have both gone over the same arc, a field that has not changed is a field that has decided to keep paying for the same shot. Moving a rider to that arc costs a fielder somewhere else, which is real, but repeating a concession without reason is worse.

Plugging the single, and the one situation where it is right

The default instruction at the death is to concede the single and prevent the boundary. Six singles is a six-run over, which is a very good outcome, and one six with three singles is a nine-run over, which is a worse one. The arithmetic is straightforward and mostly correct.

It is also incomplete, because a single does more than score a run. It moves the strike.

Suppose one of the two batters at the crease is a genuine finisher and the other has arrived within the last few deliveries. A field that concedes singles freely is handing the finisher roughly half the remaining deliveries and the new man the other half, when the bowling side would much rather the new man faced most of them. In that specific situation the four inside fielders can be repositioned to make the single hard, usually by squeezing the straight areas where the easy push goes, and the bowler is asked to attack a line that makes the low-risk nudge unavailable.

The same logic applies with the last recognised batter at the crease and the tail behind him. Deny him the single off the first four balls of an over and he is forced to attempt a boundary from a position he did not choose, which raises his dismissal rate and, if he does get out, exposes a batter who cannot do the job.

Plugging the single is expensive and it should be rare. It costs the field its boundary protection precisely because the four inside fielders have been pulled straighter and squarer to close the gaps, and a batter who reads it correctly can hit over the top into space that is now genuinely empty. But it exists, it is a real tool, and a captain who never uses it has confused a default with a rule. The broader question of how these choices sit alongside bowler allocation and matchup selection is worked through in the piece on what a captain is actually deciding.

A right-hander and a left-hander at the crease breaks the field in half

A batting side that manufactures a left-hander and a right-hander at the crease for the final overs has won an argument before a ball is bowled, and the reason is entirely about fields rather than about batting.

Every death field is asymmetric, because it has decided which arcs to protect. That asymmetry is measured from one batter's stance. Rotate the strike to a batter of the opposite hand and every off-side placement is now a leg-side placement, the short square boundary is on the other side of the ground, and the arc the captain chose to leave open is the arc the new batter most wants.

The fielding side has three imperfect responses. It can reset the field every time the strike rotates, which is legal and correct but eats the sixty seconds it does not have, and which telegraphs the plan by showing the batter exactly what the captain fears. It can set a compromise field that works moderately for both, which is what most sides do and which means every delivery is defended slightly worse than it could have been. Or it can change the bowler's line rather than the fielders, using the crease and the angle to make the same field function differently against the two batters, which is the most economical answer and the hardest to execute.

This is why the last two overs of a tight chase contain so much apparently pointless singles-declining, and why a batting pair will sometimes turn down a comfortable run. Keeping the right batter on strike is worth a run. Forcing the fielding captain to choose between resetting nine positions and defending the wrong half of the ground is worth more.

The over-rate penalty removes a boundary rider at the worst moment

In men's T20 internationals a fielding side that is not in position to bowl the first delivery of the last over by the scheduled time loses a fielder from the deep for the remainder of the innings. Five outside the circle becomes four. Some domestic competitions apply their own version of the same idea at different trigger points.

The sanction is unusual in sport because it is not a fine or a suspension applied later. It changes the physical shape of the contest, immediately, in the phase where the change costs the most. A captain who has been running a plan built on five riders discovers with a few overs left that he now has four, which means one of the arcs he was covering is open and the batter can see it.

The tactical consequences are larger than they look. Losing a rider does not simply concede one more scoring option; it collapses the whole structure of the field, because the four remaining riders have to spread wider to cover the same arc, which pulls each of them further from the ball and makes every one of them worse at his job. A boundary that would have been cut off at two now runs. And the inside four cannot compensate, because pulling one of them out to the rope would leave three men covering the entire ring.

The practical upshot for a fielding side is that the over rate through the middle overs is a death-overs decision taken twelve overs early, which is a very difficult thing to make players feel. It is also a reason captains delegate the running of the clock, keep field changes between overs to one or two, and settle the death plan long before the death arrives.

Deep point or deep third: choosing which off-side gap to leave

Take one concrete choice and follow it all the way through, because the reasoning generalises.

A right-arm seamer is bowling wide yorkers outside off stump to a right-hander. The batter's scoring options against that line are, broadly, an inside-out hit over extra cover, a slice or steer behind square through the point and third region, and a paddle or ramp fine on the leg side. The captain has five riders and cannot cover all three of those plus the straight boundaries plus the leg-side arc.

Long off is effectively compulsory, because the inside-out hit over extra cover is the highest-value shot available and it goes for six. Deep midwicket or long on is close to compulsory as well, because leaving the whole leg side empty invites the batter to change his mind. The fine leg region takes one, given the ramp. That leaves one rider for both deep point and deep third, and they are twenty-five metres apart.

Deep point covers the slice off the toe of the bat that carries. Deep third covers the steer and the edge that runs away. The correct answer depends on this batter's technique against width, and a captain who knows it puts his man in the right place and accepts that the other option will occasionally cost four. A captain who does not know it splits the difference, puts the fielder halfway between the two, and has created a fielder who cannot reach either shot.

Splitting the difference is the most common error in death field setting and it is almost always wrong. Two half-covered gaps concede more than one covered gap and one open one, because a covered gap actually stops the ball and a half-covered gap does not. Decide, and be wrong occasionally, rather than hedge and be slightly wrong every ball.

The keeper, the follow-through and the two fielders nobody counts

Two members of the fielding side sit outside the placement arithmetic and both of them matter more at the death than at any other time.

The wicketkeeper is the only fielder whose position is unrestricted by the circle, and at the death his depth is a genuine decision. Standing back gives him range for the edge and the top edge that flies over the ring behind square, and it lets him take the ball cleanly against pace. Standing up, which is rare against seam, takes away the batter's ability to move deep in his crease or to advance without risk, which cuts off both the ramp from a deep base and the charge. It also makes a stumping live, which against a batter who is walking down the pitch to negate the yorker is not a fantasy. The trade is real: a keeper standing up will concede byes and will drop the occasional edge that a man standing back would have taken comfortably.

The bowler is the other. Once he has delivered the ball he is a fielder, and his follow-through leaves him in one of the two most productive fielding positions on the ground: straight, close, and in the path of the drive that has been hit hard along the ground. A bowler who finishes his action balanced and facing down the pitch is worth a fraction of a fielder. One who falls away to the leg side has taken himself out of the game and left a hole that the captain has to fill with one of his four.

Neither of these appears in a field diagram, which is part of why field diagrams flatten the subject. Nine positions plus a keeper's depth plus a bowler's landing position is eleven variables, and a good side is managing all of them.

Building a death field, decision by decision

1. 1Walk the ground before the tossMeasure the square boundaries with your eyes and find the short side. Every field you set for the rest of the day is a response to that asymmetry, and it is the one piece of information available before anything else has happened.
2. 2Agree the two plans in the innings breakThe bowler names a primary delivery and a fallback, and the captain builds a field that suits both imperfectly. Sixty seconds between overs is not enough time to design a field from nothing.
3. 3Fix the compulsory riders firstThe two straight boundaries and the highest-value arc for this batter take three of the five. What remains is the part of the field that carries any real choice.
4. 4Decide which single arc stays openFive riders cannot cover the ground, so name the gap deliberately and choose the one that requires the batter to play his least reliable shot to reach it.
5. 5Fill the ring with savers, not catchersFour inside fielders, placed square and behind square to turn twos into ones. Add a catcher only against a new batter or when defending a small total.
6. 6Read the batter's base at releaseBacking away opens the off side, walking across opens the leg side behind square. Adjust one fielder, not five, and adjust him before the batter looks up.
7. 7Re-angle rather than re-place when the line changesMoving the bowler's line two feet wider shifts every scoring arc, and the riders should shuffle a few metres each rather than swap positions. The batter sees a settled field and gets less information from it.
8. 8Reset on the strike rotation, or change the line insteadA left-hander replacing a right-hander inverts the whole field. Either move the men, which costs time and reveals the plan, or move the bowler's angle, which costs neither.
9. 9Absorb the over-rate penalty if it comesWith four riders instead of five, abandon the least valuable arc completely rather than spreading all four thinner. Four fielders covering their zones properly concede less than four covering five zones badly.
10. 10Between deliveries, walk one manA single fielder moved three metres during an over is the cheapest adjustment available and the hardest for the batter to read. Wholesale changes announce that the plan has failed.

The sequence a captain and bowler work through before and during an over at the death. The order is the reasoning rather than a rule, and the specific placements named are illustrative of the logic rather than prescriptive.

What to look at in the thirty seconds before the seventeenth over

Broadcast coverage almost never shows the field before the ball, which is the one image that would explain the over. Watch for these instead and the phase stops looking random.

Find the open arc. Count the riders, then find the section of boundary nobody is standing in. That gap is the captain's decision made visible, and it tells you what he thinks this batter cannot do. If the batter hits his next boundary there, the read was wrong.

Check whether the riders are on the rope or in front of it. Fielders standing several metres in are being asked to attack the mishit, which usually means the plan is a slower ball or a bouncer. Fielders flat on the rope suggest a side that has decided the batter is clearing them and wants bodies underneath.

Watch the ring, not the deep, for the catchers. A short cover or a short midwicket at the death is a statement. It means the captain thinks this batter is close to a mistake, or that the bowling side needs a wicket badly enough to buy twos.

Look at the leg side behind square. Two fielders is the legal maximum. If both are very fine, the side is defending the ramp. If one has been pushed squarer, the captain has decided the sweep or the pull is the bigger risk and has left the scoop open.

Watch what changes when the strike rotates. A captain who moves four men when a left-hander takes strike is doing the work properly. One who moves nobody has either an unusually good bowler or a plan he has not thought through.

Do that across a few matches and the last four overs resolve into something legible: a captain spending a fixed budget of nine bodies on an unfixed set of threats, making a small number of explicit concessions, and being judged on outcomes that the concessions only partly control. It is the most visible piece of tactical thinking in the sport and the least commentated. More of the format's structure is unpacked across the cricket section, and the rest of the long-form explainers live in the blog.