Skip to content
Est. for the curious
The Cricket Daily
← Swing
SwingPart 3 of 12

Why the Ball Curves

A cricket ball can curve through the air with nothing touching it. The force that bends it is invisible — but it is not mysterious. It is the air itself, tripped uneven by a line of stitching.

By The Cricket Daily Desk · 7 min read

Somewhere in a Test match, a fast bowler runs in and lets go of a ball that leaves his hand aimed at the stumps. Halfway down the pitch it is still aimed at the stumps. Then, in the last few yards, it bends — sideways, through the air — and clips the edge of a bat that was never in the right place. No wind moved it. Nothing touched it. It simply curved.

That curve is swing, and in the first part of this series we drew a line around it: swing is the ball bending in the air, before it lands, and it is the only kind of movement the pitch has nothing to do with. This part answers the obvious next question. If the pitch isn't bending it, and no wind is pushing it, what is?

The answer is the air the ball is flying through — and the reason the ball's own surface is built, deliberately, with two different sides.

The thin sheet of air that does everything

A ball flying through still air isn't really moving through nothing. It drags a very thin layer of air around with it, clinging to its surface — thinner than a sheet of paper. Aerodynamicists call this the boundary layer, and almost everything about swing happens inside it.

The boundary layer comes in two kinds. It can be smooth and orderly — flowing in neat sheets, which is called laminar. Or it can be churning and chaotic, tumbling over itself, which is called turbulent. And here is the fact that runs against every instinct: the churning, turbulent layer clings to the ball longer than the smooth one does.

That sounds backwards. Rough, messy air should surely let go sooner. It doesn't — and you have already seen the proof. A golf ball is covered in dimples for exactly this reason. The dimples deliberately churn the air into turbulence so it grips the ball further round the back, which lets the ball fly further. Smooth would be worse. The same rule is the engine of swing.

TURBULENT AIR CLINGSSmooth ballair in clean sheets — laminarair →SMOOTHlets go earlywide wake — heavy dragRough ballair churned up — turbulentair →ROUGHclings — lets go latenarrow wake — less dragDimples make a golf ball fly far; roughness gives an old cricket ball its trick.
Turbulent air clings. Smooth ball: the air lets go early — wide wake, heavy drag. Rough ball: it clings past the widest point — narrow wake, less drag.

Where the air lets go decides everything

Picture the air flowing around the ball, front to back, splitting to pass over each side. At some point on each side, the boundary layer can't hold on any longer and peels away — it separates. Behind the ball it leaves a wake, the pocket of disturbed air trailing after it.

If the ball is symmetrical, the air lets go at the same point on both sides. The wake sits dead centre, and the ball flies straight. Swing is what happens when you make the air let go at a different point on each side. Do that, and the wake is pushed off to one side — and a wake pushed to one side means the ball is pushed to the other. That push is swing.

So the whole problem reduces to one thing: how do you get the air to cling further round one side of the ball than the other? This is where the ball's construction — the two sides we met in Part 2 — stops being trivia and becomes the point.

The seam is the switch

Down one side of every cricket ball runs the seam: the ridge of stitching holding it together. A bowler angles that seam, and holds the ball so one side stays smooth and polished while the other is left rough. When the ball flies with the seam angled off to one side, the seam does something small and decisive to the air passing over it.

For a long time the textbook explanation was that the seam “trips” the smooth, laminar air into turbulence. The modern wind-tunnel work refines this. High-speed infrared imaging by a Cambridge group in 2024 shows the seam behaving less like a single trip and more like a row of tiny stirrers, shedding little coherent swirls of air that turn the layer turbulent just behind it. The name they give it is worth keeping: the seam acts more like a row of vortex generators than a wire trip.

The effect, whatever we call it, is the one we need. On the seam side the air becomes turbulent, so it clings on further round the ball. On the smooth side it stays laminar and lets go early. The Cambridge measurements put actual numbers on it. Picture the angle measured around the ball's surface starting from its nose — the point that meets the oncoming air head-on — with the widest point of the ball at 90 degrees and the back at 180. On the smooth side the air peels away at about 80 degrees, just before that widest point; on the seam side it clings on to about 120 degrees, well past it and curling round toward the back.

THE NEW BALL: A TRIP-WIRE SEAMSMOOTHSMOOTHair parts round the ballnever meets the seam — stays laminarcrosses the seam — trips turbulentlets go earlyclings — lets go latethe wake skewstoward the clean sidepushed the way the seam pointsconventional swingNothing rough yet — the tilted seam alone trips one side's air, and the new ball bends.
The seam trips one side. Both sides polished; the tilted seam alone stirs the air on the side it faces — that side clings late, and the ball swings toward the seam.

Half a metre of curve

Forty degrees of difference is enough. The wake tips toward the smooth side, and the ball is pushed the other way — toward the side the seam is pointing. That is the direction of swing, and it is why a bowler aims the seam where he wants the ball to go.

HOW AIR BENDS THE BALLSMOOTHROUGHair parts round the balllaminar — smooth sheets of airturbulent — churned by seam and roughlets go earlyclings on — lets go latethe wake skewstoward the smooth sideair deflected upwardpushed toward the rough sidethe path bends — swingThe air lets go unevenly, the wake skews, and the ball is pushed sideways — swing.
The full picture: the laminar side lets go early, the turbulent seam side clings late, the wake skews, and the ball is pushed toward the seam side — the path bends.

The force is not large, and it does not need to be. The same 2024 study measured the sideways force in a wind tunnel and worked out what it does over a real delivery: a good swinging ball at about 85 miles an hour bends roughly half a metre sideways on its way to the batter. Half a metre is the difference between the middle of the bat and the outside edge. It is the whole contest.

How much a good ball swings~0.5 m sideways over the pitch, at ~85 mph

Why swing looks “late”

Commentators talk about “late swing” as if it were a special, more dangerous force that switches on near the batter. It isn't a separate force at all — it is a trick of the shape of the flight.

The sideways push is roughly steady the whole way down. But a steady sideways force doesn't move the ball in a straight diagonal; it bends the path into a curve — a parabola, the same shape a thrown stone follows as it falls. And a parabola does most of its sideways travel at the end. So even with a constant force acting evenly from the moment of release, the ball drifts only a little over the first half of its journey and swerves most in the last few yards — right in front of the batter. Every ball that swings, swings “late.” No extra force required; it is built into the curve.

WHY SWING LOOKS 'LATE'most of the bend happens here —the force never changedthe no-swing linereleasebatterhalfway down: only a quarter of the bendequal pushes, evenly spaced — the force is constant≈ half a metreedge vs middle of the batSame steady push the whole way — the curve just saves its movement for the end.
A constant sideways push, and the parabolic path it draws — little movement early, most of the bend in the final third. “Late” swing is the shape of the flight, not a new force.

The myth of the fast swinger

If the seam and the two sides are doing the work, where does raw pace come in? Less than you would think — at least for ordinary swing. The wind-tunnel data is blunt about it: below a certain speed, bowling faster barely changes how much the ball swings. The Cambridge group measured less than a five per cent change in the swing force as a new ball sped up from around 64 to 87 miles an hour. Within that band, a genuinely quick bowler and a brisk medium-pacer get almost the same curve. Pace buys other things — less time for the batter, more bounce, a price when you miss — but it does not, by itself, buy more swing. There is a second regime, at the top end of pace, where the rules flip and the ball can swing the other way. That is reverse swing, and it has its own part.

How often it really happens

One last, honest number. With all of this working — the two sides, the angled seam, a watchful bowler — how much does a new ball actually swing in a real match? Researchers tracked nearly seven thousand first-over deliveries across a decade of Test cricket. In 44 per cent of them, the ball barely swung at all. Only about 15 per cent swung by the big, edge-finding margins the science allows. Even at the highest level, on the days and grounds that suit it, a ball that genuinely talks in the air is the exception, not the rule. That rarity is exactly why a bowler who can do it on demand is worth so much.

So swing is not magic and not the weather. It is a thin layer of air, a line of stitching that stirs that layer on one side, and a ball built with a smooth face and a rough one so the two sides never behave alike. Point the seam, keep one side shining, and the air does the rest.

Which raises the question every bowler asks next — not whether to swing it, but which way. The same action, the same ball, can be made to curve in to the batter or away from him, and the good ones hide which until it's too late. That is where we go next.