Starc's 6/72 at Newlands: Nine Wickets for 126 Runs, and What the 20th Five-Wicket Haul Actually Measures

The board at Newlands on Tuesday read 203/1. Aiden Markram had 78, Marques Ackerman 95 off 94 balls, the opening stand was worth 140, and the series was locked at 1-1, which meant the first day of the deciding Test belonged to South Africa. The innings ended at 329. Nine wickets for 126 runs after that 203, and the last four of them inside six deliveries, including back-to-back golden ducks for Gerald Coetzee and Kagiso Rabada.

Underneath the whole ledger sits one name: Mitchell Starc, 6/72.

I do not measure a fast bowler's spell by its headline figure. 6/72 is an event. A twentieth Test five-wicket haul is a career entry. And when I opened that entry in a second ledger — innings per five-wicket haul — the number stayed identical while its meaning changed. I do not call a number a verdict until it survives the second ledger.

Start with the part the scorecard under-reports. South Africa's innings broke into four segments, and the four add up to one story. Segment one: 203/1, Markram's 78, Ackerman's 95 off 94 at a strike rate above a hundred, and David Bedingham's 50. Segment two: 203/1 to 230/6, five wickets for twenty-seven runs, with Bedingham and Wiaan Mulder (nought) falling to Starc. Segment three: Marco Jansen (52) and Ryan Rickelton (42) putting on 95 for the seventh wicket to drag the innings to 325/6. Segment four: 325/6 to 329 all out — four runs, four wickets, six balls.

The shape matters because 'Starc engineered the collapse' is an incomplete sentence. All six of his wickets were partnership-breakers — two at the top, one to end the 95-run stand, three at the tail. (Six wickets in a spell is rare; six wickets that each end a stand is rarer still.)

Now the record, without the theatre. 6/72 is Starc's best Test return against South Africa, improving on the 6/154 he took at Perth in 2026. Thirteen years apart, a five-wicket haul on both occasions. That is a ledger entry in itself. In Tests this is his twentieth five-for, with nine in ODIs — 29 international five-wicket hauls, level with Dale Steyn. Among fast bowlers all-time he sits sixth: Richard Hadlee (41), Glenn McGrath (36), Waqar Younis (35), James Anderson (34), Wasim Akram (31), then Starc and Steyn on 29. He is now clear at the top of the active fast-bowlers list in Tests, ahead of Kagiso Rabada (17) and Jasprit Bumrah (16).

Here is the part I opened the piece for. In a list, '20' looks like the biggest number until you add the innings column beside it.

Starc: 20 five-fors in 210 innings — one every 10.5 innings. Rabada: 17 in 136 — one every 8.0. Bumrah: 16 in 99 — one every 6.2. Josh Hazlewood: 15 in 151 — 10.1. Pat Cummins: 14 in 141 — 10.1. Kemar Roach: 11 in 156 — 14.2.

Twenty five-wicket hauls are a receipt for a long career, not for a high peak. Starc's figure is evidence of durability and volume. On innings-per-five-for, Bumrah is roughly 41 per cent ahead of him. That is not a slight; it is the correct column. If the question is who takes five wickets most often, the answer is Starc. If the question is whose innings is likeliest to produce five, the answer is Bumrah. Two questions, two answers — and most arguments collapse them into one.

The 29-29 with Steyn deserves the same treatment. Steyn's 29 split 26 Tests and three ODIs, about 90 per cent in the longer format. Starc's 29 split twenty and nine, about 69 per cent. Same total, entirely different construction. Five wickets in an ODI is harder than in a Test — ten-over cap, one new ball, a tail that must be cleared quickly. Starc's nine ODI five-fors are the most expensive portion inside that 29, and the table never says so.

The same distortion runs through the all-time list. Anderson's 34 include 32 in Tests; Waqar's 35 include 13 in ODIs; Starc's 29 include nine. One table is adding up three different kinds of career. A list that does not weight the formats can deliver a ranking; it cannot deliver a valuation.

Within the spell, the split is just as legible. Starc's first wicket was Bedingham, already past fifty. Mulder went for nought. So the first shove in the 203/1 to 230/6 segment was his, but the three wickets in between belong to someone else. I record this because reading 6/72 and assigning the entire collapse to one bowler is easy and wrong.

His real contribution was the second new ball. Removing Rickelton (42) to break the 95-run stand, bowling Jansen (52) when he was the most settled batter of the innings, then sending back Coetzee and Rabada on consecutive golden ducks. Four wickets in six balls. With the older ball in the first spell Starc stalled the innings; with the new ball in the second he closed it — two different skills compressed into a single '6'.

Now the brake. Watching 203/1 become 329 all out, it is easy to say the pitch came alive or Starc became unplayable. Neither happened. Jansen and Rickelton put on 95 for the seventh wicket — numbers seven and eight scoring at nearly a run a ball on the same surface against the same ball. A pitch where the seventh and eighth men add 95 does not support a 'spitting minefield' narrative.

The final six balls are a separate matter. Coetzee's and Rabada's consecutive golden ducks are a fact, but they are also a story about shot selection and tail-end technique, not only about bowling. This is where correlation detaches from cause. The outcome was four wickets in four balls credited to Starc; had Rabada left his, the figure would read 5/70 and the entire list would look different.

Based on my five decades of watching, one entry needs filing: the five-wicket haul is a very loud statistic built on a very small sample. Six wickets in an innings is sometimes proof that a bowler controlled an entire innings, and sometimes proof that he was handed the tail in the closing overs. Nothing in the headline figure separates the two.

Which is why '6/72 turned the match' is an unfinished sentence. Whether it turned the match will be settled by the second innings. What is settled is that 329 was made, and made at three different speeds — 203/1, then five wickets for 27, then a stand of 95, then four wickets for four.

The signal for the next round sits in three places. First: if 329 is below par at Newlands, Starc's spell is not the turning point but South Africa's batting error, and Australia's batters will answer that question themselves. Second: Rabada is bowling in this same match, on 17 five-fors from 136 innings — a better rate per innings than Starc — and one spell from him could invert this entire piece. Third: Starc needs two more international five-fors to level Wasim Akram on 31 and three to pass him, and nine of the 29 are already banked in ODIs.

Anyone who reads Starc only through the lens of 'twenty times' will miss the man who will stand at the top of his mark with a new ball the next morning. Nobody knows when the next ledger entry lands. What is certain is that it will be written by the next six balls, not by this 6/72.