T20 fielding restrictions explained: one innings, two games

At the end of the sixth over of a Twenty20 innings, three fielders jog out to the rope. Nobody announces it, the commentary rarely stops for it, and it is the largest single change in scoring conditions that will happen all match. The bowling has not improved. The pitch has not changed. Three men moved, and the game the batters are playing has been replaced with a different one.

That moment is the whole subject. T20 fielding restrictions are usually explained as a rule about numbers, two fielders out and then five, which is accurate and almost useless. The rule matters because of what it does to a twenty-over innings, which is to cut it into two segments with completely different economics, joined by a step so abrupt that a batting pair can be correct at 7.35pm and wrong at 7.39pm without doing anything differently.

Every role in a modern T20 side exists because of that split. The powerplay opener who has no business batting at four. The middle-overs pair who look pedestrian until you watch what they are actually beating. The finisher whose strike rate would be indefensible in any other phase. None of those are personality types. They are jobs created by a painted line and a rule about how many people may stand outside it.

What the T20 fielding restriction actually says

The rule is short. During the first six overs of an innings, no more than two fielders may be outside the thirty-yard circle. From the seventh over to the twentieth, no more than five may be.

Four things about that sentence are worth pinning down before any tactics follow from it.

The count is nine, not eleven. The bowler is where his run-up puts him and the keeper is behind the stumps, so neither is placed by the captain in any meaningful sense. A captain has nine bodies. In the powerplay he must keep seven of them inside the circle. After it he must keep four. Every field setting in this article is somebody deciding how to spend nine.

The test applies at one instant. The field is judged as the ball leaves the bowler's hand. Where a fielder goes afterwards is his own business, and sides use this constantly, starting a man with his toe inside the line and letting him drift fifteen yards deeper as the batter shapes to play. A ring fielder in the last over is frequently a boundary rider by the time the ball arrives. The rule constrains the starting position, not the play, and reading a field from a wide television shot taken after the ball is struck will mislead you every time.

A breach is a no ball. Not a warning, not a quiet word with the captain. The umpires check the count before delivery and call it if the field is wrong, and most T20 competitions attach a free hit to every no ball, so a captain who miscounts has bought his opponent two deliveries and a licence to swing at one of them. This is why you occasionally see a fielder sprinting inwards while the bowler is still walking back, and why the twelfth man is sometimes the most useful person on the ground.

Two further limits apply throughout. No more than two fielders behind square on the leg side at the instant of delivery, which comes from the Laws rather than the playing conditions, and in international cricket no more than five on the leg side in total. Neither is a powerplay rule. Both bite hardest at the death, when a captain wants a leg-side field for a short-ball plan and finds the rule book has already spent his allocation.

• 6Overs under the tight restriction
• 36Deliveries in the restricted phase
• 2Fielders allowed outside, overs 1 to 6
• 5Fielders allowed outside, overs 7 to 20

Rule-defined values from the ICC Twenty20 playing conditions and the Laws of Cricket. The bowler and the wicketkeeper are excluded from the count of placeable fielders.

Thirty-six deliveries. That is the entire restricted phase, and it is worth holding on to, because almost everything sides argue about in T20 selection reduces to who should be facing those thirty-six balls and what they should be trying to do with them.

The parallel rules in the other limited-overs formats are set differently, with an intermediate band in a fifty-over innings and a phase counted in balls in the hundred-ball game. Those comparisons are laid out in the guide to how the restriction is written in each format, and the point that matters here is what twenty-over cricket leaves out. There is no middle setting. The restriction goes from its tightest available value to its loosest in the space of one ball, and no other format asks a batting side to absorb that in a single step.

Six overs and fourteen overs are two different games

Call them what they are. Segment one is thirty-six deliveries played against an outfield containing two people. Segment two is eighty-four deliveries played against an outfield containing five.

The difference is not a matter of degree. In segment one, a batter who beats the infield has, in almost every direction, nothing but grass in front of him. He does not need to find a gap, because the field is mostly gap. In segment two, a batter who beats the infield has beaten it into a sector somebody is probably standing in, and the shot that was worth four is now worth one and a scrambled two if he runs hard.

Here is the geometry, as a constructed example rather than a measurement. Take an idealised circular ground with the rope seventy yards from the middle of the pitch. Its circumference is a shade under 440 yards. Assume, conservatively, that a fielder on the rope reliably cuts off anything landing within fifteen yards either side of him, so each boundary rider defends about thirty yards of arc.

Overs 1 to 6, two fielders out14%

Overs 7 to 20, five fielders out34%

After an over-rate penalty, four out27%

Invented worked example on an idealised circular ground with a seventy-yard boundary, giving a rope of about 440 yards. Each outfielder is assumed to defend thirty yards of arc. Not a measurement of any real ground, where boundaries are oval and vary considerably.

Show the numbers

Share of the boundary a captain can defend, constructed example

Item	Boundary arc defended
Overs 1 to 6, two fielders out	14%
Overs 7 to 20, five fielders out	34%
After an over-rate penalty, four out	27%

Fourteen per cent against thirty-four. Under those invented assumptions a captain in the powerplay is protecting roughly one seventh of the rope and conceding the rest, and after the restriction lifts he is protecting a third. He still concedes two thirds, which is why T20 remains a scoring game rather than a strangling one, and the doubling is enough to change every decision a batter makes.

The same arithmetic runs on the inner circle. The thirty-yard circle is two semicircles of thirty-yard radius joined by two straight lines the length of the pitch, so its perimeter is roughly 232 yards. Seven fielders spread around 232 yards sit about thirty-three yards apart, which is a genuine cordon, close enough that a firmly struck ground shot along the turf is usually stopped. Four fielders around the same perimeter sit about fifty-eight yards apart, which is not a cordon at all. It is four men and a lot of daylight.

That is the trade the seventh over makes. The infield becomes porous at the exact moment the outfield becomes populated. Ground shots that were saved are now singles and twos, aerial shots that were fours are now caught or cut off, and the scoring geography inverts. A batting pair that does not invert with it is playing the previous game against the current field.

Why a batter should accept getting out in over three

The standard criticism of a powerplay dismissal is that the batter threw it away. Usually he did the right thing and it did not come off, and the arithmetic that makes it the right thing is not complicated once somebody writes it down.

Everything below is a constructed example with invented round numbers, chosen to be legible rather than realistic.

Take one particular shot, a lofted drive over extra cover, played by a batter good enough to middle it more often than not. Give him ten attempts in the third over of an innings, with two fielders out.

Constructed example, over three, ten attempts:

Outcome	Attempts	Runs
Clears the ring, reaches the rope	4	16
Clears the ring, cut off in the deep	2	3
Fails to clear cleanly, stopped in the ring	3	1
Caught	1	0
Total	10	20

Twenty runs from ten balls, one wicket. Two runs a ball, at a price of one dismissal per ten attempts.

Now the identical shot, identical contact, in the eighth over with five fielders out. Nothing about the batter has changed. Three men are standing in places that were empty.

Constructed example, over eight, ten attempts:

Outcome	Attempts	Runs
Clears the ring, reaches the rope	1	4
Clears the ring, cut off in the deep	5	7
Fails to clear cleanly, stopped in the ring	2	1
Caught in the deep	2	0
Total	10	12

Twelve runs from ten balls, two wickets. The gross return has fallen by two fifths and the dismissal rate has doubled, because the fielders who were absent from the outfield are now standing under the ball that used to bounce in front of the rope.

To finish the comparison, put a price on the wicket. Assume, again as an invented figure, that a set batter scores at 1.6 runs a ball and a new batter at 1.0 for his first ten deliveries while he gets his eye in. A dismissal therefore costs about six runs in lost scoring, and assume for now that wickets are not scarce, which in the first six overs of a T20 they are not.

• Over three, attacking. Two runs a ball gross, minus a dismissal risk of 0.10 a ball at six runs each, which is 0.60. Net 1.40 a ball.
• Over eight, attacking. 1.20 runs a ball gross, minus a dismissal risk of 0.20 a ball at six runs each, which is 1.20. Net nothing at all.
• The safe alternative in either phase. Call it 0.80 runs a ball with negligible risk, working the ball into gaps and running hard.

Attacking in over three returns 1.40 against the safe option's 0.80. Attacking in over eight returns zero. Same shot, same batter, same bowler. The rule changed and the answer flipped.

Two things fall out of that, and the second is the one sides get wrong.

The first is obvious enough. High-risk batting in the powerplay is not recklessness, it is the correct response to a field the rules have emptied, and a side that opens with two accumulators is declining the only cheap money in the innings.

The second is that a T20 innings has ten wickets and most innings do not use them. A side that finishes twenty overs four down has held four wickets it was never going to spend, and every one of those unspent wickets represents risk it could have taken and did not. Wickets are not a score to be protected. They are a budget with an expiry date, and the powerplay is where the exchange rate is best. Getting out in over three is not the failure. Scoring at seven an over through the powerplay and then getting out in over eleven is the failure, because that batter took his risk in the phase where it paid least.

There is a limit to this and it deserves stating plainly rather than leaving as an implication. The arithmetic assumes wickets are plentiful. Three down inside four overs and the price of the next one rises sharply, because the side is now at risk of running out of batters and forfeiting deliveries entirely, which is the one outcome no amount of hitting recovers from. Rational aggression in the first six overs is aggression by a side with a full budget, not by a side that has already spent most of it.

The same shot in over seven is a different decision

Watch the seventh over of any T20 and you will see at least one batter play the shot that worked in over five, find the man at deep cover who was not there four minutes earlier, and walk off looking puzzled.

The lag is real and it is structural rather than a lapse in concentration. A batting pair spends thirty-six deliveries building a model of where the ball can safely go, and the model is wiped in one ball. Nothing in the bowler's action signals it. The only cue is the field, and a batter at the non-striker's end has a far better view of that than the man on strike, which is one of the few genuinely useful things a non-striker does.

Good pairs treat the transition as a scheduled event. They talk in the gap between the sixth and seventh overs specifically about which three fielders went out and which sectors are now open, because the captain has just made his plan public and it will not change much for several overs. Sides that get this right usually have one batter whose explicit job at the change is to take a single, get to the other end, and look at the field from the position where it is easiest to read.

The shot selection changes shape as well as frequency. In the powerplay the target is the sector, because most sectors are empty. After the restriction lifts the target is the gap between two fielders, which is a much smaller object and demands a different sort of shot: along the ground, placed rather than launched, hit hard enough to beat the ring but not so hard that a sweeper cuts it off for one. Manipulating that field is an entirely different skill from clearing an empty outfield, and it is why the names and coordinates of the standard fielding positions are worth learning properly, because a batter who cannot name what he is looking at cannot plan against it.

1. 1Overs 1 to 6, two fielders outThirty-six deliveries against an outfield with two people in it. The ball is hard and swinging, so edges carry and the bowler has catchers he can afford. Both sides attack, because the rule has removed the defensive option from each of them.
2. 2The seventh over, three fielders walk outThe number permitted outside the circle goes from two to five in one ball. The infield thins from seven to four, the outfield fills, and the value of every aerial shot falls at the same instant the value of a well-run two rises. This is the sharpest change in the match.
3. 3Overs 7 to about 12, the ring goes huntingFive out, four in, and the four in the ring now stand roughly fifty-eight yards apart. Singles are available almost everywhere and boundaries have to be manufactured. Spin usually operates here, because five out is exactly enough protection to attack with.
4. 4Around over 13, the plan splits in twoA side with wickets in hand starts spending them; a side without starts rebuilding towards a smaller total. Both decisions are made against the same field, which is why the middle overs separate good T20 sides from ordinary ones more reliably than either of the loud phases do.
5. 5Overs 16 to 20, wickets stop being worth savingThe restriction has not moved since over seven, but the batting side's attitude to risk has inverted. A batter with three overs left and six wickets standing loses almost nothing by being dismissed, so the price of a wicket collapses towards zero and the attack rate rises to meet it.
6. 6The final over, and the over-rate clauseIn most competitions, a fielding side not in position to start the twentieth over by the published cut-off must bring a man back inside the circle for the rest of the innings. Five out becomes four, and a captain who was already short of cover loses a sector at the worst possible moment.

The over numbers and fielder counts are from the Twenty20 playing conditions. What each transition does to the batting and bowling plans is the tactical consequence, not a rule.

The bowler's problem is inverted at the same instant

Everything above is the batting side's version. The bowling side is solving the mirror image, and the mirror is not symmetrical.

In the powerplay, containment is not on the menu. A bowler with two fielders out has no sweeper, no long off and no long on. If he bowls a defensive length wide of off stump, he is inviting a batter to hit into an acre of nothing with no downside. The field cannot support a containing plan, so the only plan the field supports is the wicket-taking one: full, straight, at the stumps, buying a chance with the runs the rules have already promised the batter. Powerplay bowling looks aggressive because the alternative is unavailable, not because the bowler is brave.

This is why judging a bowler's powerplay figures by the same standard as his death figures is a category error. An economy rate that would end a career in the middle overs is unremarkable in the first six, and a wicket in over two is worth more than a wicket in over twelve because it puts a new batter in against a hard ball with catchers in place.

After the restriction lifts, containment becomes the primary weapon. With five out, a captain can close the two straight boundaries and both square ones, leave the least dangerous sector open, and bowl to make the batter hit there. The bowler stops trying to beat the bat and starts trying to make the batter play a shot into a covered area. That is a fundamentally different skill, and plenty of bowlers are good at exactly one of the two.

The captain's allocation problem follows directly. Four overs each, five bowlers, and the overs are not interchangeable because the phases are not. Spend your best bowler in the powerplay and you get him when a wicket is worth most, at the cost of somebody worse bowling the eighteenth. Hold him back and the powerplay is bowled by the man most likely to concede the runs the restriction has already made available. There is no right answer, only a live trade, and the error that recurs most often is using a frontline bowler in over eleven to break a partnership and finding a part-timer has to bowl over eighteen. The moment of the mistake and the moment of the consequence are half an hour apart, which is exactly why it keeps happening.

Why the new ball is an asset with a short shelf life

Two things are true at the start of a T20 innings and they are usually discussed separately. They belong together.

The ball is new, so it swings, seams, bounces truly and carries to the keeper. And the field is restricted, so a captain who wants a slip and a short cover can have both, because he has almost no boundary protection to sacrifice. He gives up nothing by posting catchers, since the rules have already taken the outfield away from him.

Put those together and the powerplay is the only period in a T20 where a bowling side can attack the way a Test side attacks, with a ball that is doing something and men standing where the mistake goes, against a batting side the rules are actively encouraging to play at it.

The asset decays inside its own phase. The lacquer goes, the seam flattens, the ball softens against the square boundary hoardings, and by the tenth over the bowler who was beating the edge is banging it into the pitch and hoping. That decay is why the argument about who bowls the first over is not really about the first over. It is about the second and the third, and a captain who uses his best new-ball bowler while the ball is genuinely new, then holds him back, has extracted more from the asset than one who bowls him straight through until it is soft.

The liability is the other half of the same fact. A new ball comes on to the bat, which is what a batter attacking an empty outfield wants most. Pace off a hard ball is the shot-maker's friend, and a fast bowler who misses his length in over two is offering exactly the delivery the field cannot protect. The reason powerplay bowling looks so binary, either wickets or carnage, is that both outcomes come from the same combination and the fielding side cannot have one without exposing itself to the other.

Spin inside the six overs stopped being a gamble

Bowling spin with two fielders out used to be treated as eccentric. It is now routine, and the reasons are structural rather than fashionable.

The obvious problem has not gone away. A spinner in the powerplay has no cover straight and none square, so a batter who gets to the pitch of the ball can hit through the line into the emptiest field he will see all night. The margin on a delivery that lands slightly short is close to nothing.

What changed is that sides worked out what the phase gives back. A hard new ball grips and bounces off the surface far more than a soft one, so a finger spinner extracts more in over four than in over fourteen. A batter who has set himself for a fast bowler has to reorganise his feet immediately, and a batter reorganising his feet is not clearing the ropes. Most usefully, the restriction that denies the spinner his boundary riders also hands him catchers: with seven men obliged to stand inside the circle, the mishit that would drop safely into an empty middle distance after the seventh over is a catch to short cover in the fourth.

That last point is the one the criticism usually misses. The powerplay is not a bad phase for spin. It is a bad phase for a particular kind of spin, the accurate holding bowler whose whole method is denying the batter room, because denial requires a protected boundary and he does not have one. The bowler who prospers is the one who turns it and bounces it enough to induce a false stroke, which is why wrist spin and the more aggressive finger spinners took the work while the containment specialists did not. Which variety a captain trusts also turns on the matchup with the batter's handedness and the direction of turn, a question the guide to the different families of spin bowling takes apart in detail.

The death overs are a third phase the rules never wrote

The playing conditions describe two phases. Every T20 side plans for three, and the third one is created entirely by the batting side's changing attitude to risk.

Nothing in the rule book distinguishes the seventeenth over from the ninth. Five out in both. What separates them is that the batting side's wicket budget has an expiry date and the expiry is approaching. A batter in over nine who gets out costs his side the difference between his scoring and a new batter's across eleven remaining overs. A batter in over eighteen who gets out costs very little, because there are barely enough deliveries left for the loss to register and there are wickets behind him that will otherwise go unused.

So the price of a wicket, six runs in the constructed example above, collapses towards zero for a side with wickets standing. When the price of a wicket falls, the optimal level of aggression rises. That, and not the field, is why the last four overs look the way they do.

The fielding side responds by giving up on wickets almost entirely. Five men are on the rope, four are in the ring, and a dismissal at the death is a pleasant accident rather than a plan. The objective becomes conceding the fewest boundaries across the remaining deliveries, which means the bowler is bowling to a field rather than to a batter. The wide yorker with a packed off side, and the short ball into the ribs with both permitted leg-side fielders back, are field settings disguised as deliveries: the ball is selected because of where the men are standing, and the mechanics of that trade are the subject of the piece on how death bowling is actually constructed.

The constructed geometry from earlier explains why boundary counting is the finisher's real skill. Five men defending about a third of the rope leaves two thirds open, and a good finisher has worked out which part of that two thirds is reachable off the shot he intends to play before the bowler starts his run. It is why the best of them look unhurried and the weakest look violent. One is aiming at a gap that is genuinely there. The other is hitting hard in a general direction and hoping the field has made a mistake.

Captains get this wrong in a predictable direction. They protect square before they protect straight, and straight is where the damage comes from, for two reasons that compound. A death bowler who misses his yorker length has usually missed it into the arc between long on and long off, and a batter who mistimes an intended six down the ground still clears the infield often enough to be paid for it. Long off is therefore two jobs in one body, the planned shot and the error. Deep point is one job, and only against a batter who plays that particular shot.

The over-rate clause bites here too, and batting sides know it. Where a competition penalises a slow over rate by forcing a fifth fielder back inside the circle for the remainder of the innings, the penalty lands precisely when the fielding captain needs his outfield most. Under the constructed geometry above, the defended share of the rope drops from about a third to about a quarter, which is a sector, which is a boundary. The exact trigger and sanction vary between the international game and each of the competitions on the franchise circuit, so the relevant playing conditions are the place to check rather than the version anyone happens to remember.

The restriction picks the side, not just the field

A format with two economic regimes cannot be played well by eleven generalists. That is the deepest consequence of the rule and the one that shows up in team sheets.

44%22%22%11%

• Ring, saving the single4
• Square sweepers, deep point and deep midwicket2
• Straight boundary, long off and long on2
• Deep behind square, third man or fine leg1

An illustrative arrangement of the nine outfielders immediately after the restriction lifts, with five permitted outside the circle. Constructed to show how the budget of nine is spent, not a transcription of any match. The bowler and wicketkeeper are not counted.

Show the numbers

A constructed field for the seventh over, by band

Item	Value
Ring, saving the single	4
Square sweepers, deep point and deep midwicket	2
Straight boundary, long off and long on	2
Deep behind square, third man or fine leg	1

Four in the ring is the number to notice. That is the entire infield for fourteen of the twenty overs, and it is why the powerplay batter and the middle-overs batter are different jobs. Against seven infielders the ground shot is stopped and the aerial shot is free. Against four the ground shot runs away for one and the aerial shot is caught. A player whose scoring depends on clearing the infield is enormously valuable for thirty-six deliveries and a passenger for the next fifty.

Four consequences follow, and sides now select for all of them explicitly.

The top order is picked for the first six overs, not for batting in general. An opener whose method is to take the aerial route from the first ball is a rational selection even if the same method would be indefensible at number five, because he is being bought for thirty-six deliveries in the only phase where his risk profile is correctly priced. Judging him by an average is judging a specialist with a general measure.

The middle order is picked for beating a spread field without hitting over it. Running hard, converting ones into twos into a thin outfield, and sweeping or reverse-sweeping a spinner to force a fielder to move are the skills that phase rewards. None of them look impressive and all of them are trained.

Outfielding became a selection criterion in its own right. With five men on the rope for fourteen overs, a fielder who covers ten more yards than his replacement turns fours into twos repeatedly across a phase, and across a season that is worth more than most bowling changes. Ring fielding is a separate specialism again, about anticipation and starting early rather than arm strength, and sides now treat the ring and the rope as two different jobs when they pick.

A sixth bowling option is insurance against the allocation problem. Five bowlers, four overs each, and one of them having a bad night leaves a captain bowling a part-timer at the death into a field designed for a specialist. The reason sides tolerate an all-rounder whose bowling is unremarkable is that the option has value even in matches where the overs are never used.

Where the impact player meets the phase split

In competitions that allow a tactical substitute, the phase structure is exactly what the substitute is used to exploit. A side can pick a powerplay specialist and a death specialist without paying for both in the same eleven, which loosens the constraint this article has been describing throughout.

The mechanics of that substitution, its deadline, its overseas limits and its effect on all-rounders are set out in the piece on the IPL's impact player provisions, and there is no point restating them here. The connection worth drawing is that the rule only has the value it does because the innings has phases. In a format with a single uniform set of fielding restrictions, a twelfth player would be a marginal upgrade. In a format that asks for three different batting skills and two different bowling ones inside two hours, an extra slot is a way of buying a second specialist for a phase, and that is why the rule changed team composition rather than merely deepening batting orders.

How a chase is planned around the two phases

Chasing is where the phase split stops being a batting theory and becomes arithmetic anybody in the ground can do.

Take a constructed target of 181 across 120 deliveries. That is a required rate of 9.05 an over from the first ball. A batting side does not chase at 9.05 an over, because the deliveries are not equally valuable. It chases in phases.

A constructed phase plan for a target of 181:

Phase	Overs	Balls	Runs allotted	Rate	Wickets budgeted
Powerplay	1 to 6	36	55	9.17	2
First middle block	7 to 12	36	40	6.67	1
Second middle block	13 to 16	24	30	7.50	1
Death	17 to 20	24	56	14.00	3

Invented figures, chosen so the arithmetic is easy to follow. The plan spends its wickets where they are cheapest and asks for the hardest scoring rate in the phase where the price of a dismissal is lowest, which is the whole logic of the split expressed as a table.

Now the decision that actually decides chases. Two sides reach the end of the sixth over of the same chase by different routes.

Route A, 55 for 2. They need 126 from 84 balls, a required rate of 9.00, with eight wickets standing.

Route B, 40 for 0. They need 141 from 84 balls, a required rate of 10.07, with ten wickets standing.

Route B looks safer and the numbers say otherwise. The difference in required rate is 1.07 an over across fourteen overs, which is exactly the fifteen runs they declined to score while the field was empty. To get them back, Route B has to score at better than ten an over against five fielders on the rope, which is precisely the phase where scoring at ten is hardest. The two extra wickets are worth those fifteen runs only if the side actually spends them, and sides that bat cautiously in the powerplay are usually the same sides that bat cautiously in over twelve. The wickets get carried to the end unused and the fifteen runs are gone.

That is the most common way a T20 chase is lost. Not a collapse and not a bad over. A side declines the cheap runs in the phase built to provide them, transfers the debt to the phase built to prevent them, and then blames the death bowling. How to phrase a chase as a set of phase targets rather than a single running rate, and when the calculation genuinely does favour preservation, is worked through in the guide to pacing a run chase.

One qualification, because the arithmetic is not universal. On a pitch where the ball is genuinely misbehaving, or against an attack whose best bowlers are all being held for the death, the wicket price stays high through the middle and preservation can be correct. The test is not which route feels safer. It is whether the side intends to spend the wickets it saved, and on which deliveries.

Watch the walk to the rope at the end of the sixth over

The most useful habit this article can leave you with costs nothing and works from any seat in the ground, or from a sofa with the sound off.

At the end of the sixth over, do not watch the bowler collecting his cap. Watch which three fielders walk out and where they stop. That walk is the captain's plan for the next several overs, stated in public, before a ball is bowled. Two men going straight, to long off and long on, says he expects the batters to hit down the ground and intends to bowl to make them try. Two going square says he has a batter he thinks will cut and pull, and a bowler who will feed it. Where the fifth one goes tells you which single sector he has decided he cannot afford to leave open, and that is usually the most informative fielder on the ground.

Then count, every over, at the moment the bowler releases. Two outside the circle means you are still inside the first six regardless of what the scoreboard graphic says. Five means the restriction has lifted. Four, late in an innings, means the fielding side has been penalised for its over rate and a sector has just opened up, which is worth knowing several seconds before the commentators mention it.

In the last three overs, stop watching the rope and watch the four men in the ring. Pushed in and saving the single, and the fielding side believes the match is close enough that ones matter. Dropped back towards the edge of the circle, and the captain has decided ones are irrelevant and is defending against the two and the mishit that clears the infield. That adjustment is the clearest available statement of what the fielding side genuinely thinks the equation is, and it is usually visible a couple of overs before anybody says it out loud.

Learn to read those nine bodies and you are watching the same match the captains are, which is the reward for taking seriously a rule most broadcasts reduce to a graphic. The rest of the tactical vocabulary that grows out of these restrictions, from field settings to bowling plans to how sides build an order, sits in the cricket section of this site.