What’s a wicket worth? Part 1: Baggy green bowlers

By All day Roseville all day / Roar Guru

“All wickets are equal, but some wickets are more equal than others,” As George Orwell would have put it had he spent more time watching cricket and less time penning Animal Farm.

Cricket careers are measured by aggregate, average and strike rate because those numbers are straightforward and comparable. Yet the wickets of some opposing batsmen clearly have far more value than those of others.

A bowler who dismissed Don Bradman and Stan McCabe cheaply and achieved a final analysis of 2-70 would have done far more to gain victory for his country than his teammate who cleaned up the Australian tail for final figures of 4-70. But unfortunately his bare figures wouldn’t acknowledge that matchwinning contribution.

Similarly, although Adam Voges’s batting average of 61.87 might place him second only to Don Bradman among Australian batsmen who have played more than 13 matches, it is generally acknowledged that weak opposition attacks and many not-out innings contributed greatly to that figure.

Statistical gurus such as Charles Davis and Anantha Narayanan have used complex algorithms to attempt to better compare players. These have included weighting top-order, middle-order and lower-order wickets differently, adding the batting averages of all batsmen dismissed and comparing individual bowlers and batsmen head to head. They have also standardised scores and scoring rates across eras.

(Mark Kolbe/Getty Images)

Such formulae have great merit even if subjective. Unfortunately I have neither the skill nor the time to apply them. Instead I’ll simply analyse which leading bowlers did and did not bolster their career records with lots of cheap wickets, and how. By doing so it is possible to identify those undervalued bowlers whom the record books do not fairly acknowledge.

While tailenders represent approximately 29 per cent of all wickets taken, many bowlers’ tallies are far different. Very few faster bowlers have been able to claim more than 30 per cent of their victims from the ranks of tailenders. Conversely, almost all every slow bowler can count at least 30 per cent of his wickets from those batting at Nos. 8, 9, 10 and 11.

Understandably, opening bowlers dominate the takers of top-four batsmen’s wickets. They enjoy the benefits of freshness, a new ball, a pitch with life and unknown characteristics, and batsmen yet to play themselves in.

Correspondingly, teams’ final four batsmen are dismissed more often by an individual slow bowler than paceman. Often by that stage of an innings the ball is old, the pitch has lost any life that it had and faster bowlers are resting from earlier efforts.

There are other factors that skew individual statistics. While a strong team will almost always take 20 wickets per match, each of its bowlers must share the spoils with three or more teammates. And even if a weak team struggles to regularly dismiss its opponent twice, it nevertheless may have one bowler who is great and as a result always gets the ball on those rarer occasions when lower-order batsmen do get to the crease.

(Wiki Creative Commons, CC BY-SA 3.0)

Whose career figures have been bolstered most?
Perhaps predictably, the seven most frequent takers of lower-order wickets are slow bowlers and five of them are leg spinners.

From the 1920s until the 1960s it seemed that every grade and Sheffield Shield team had a wrist spinner. The style was especially useful when Tests in Australia and South Africa were timeless, pitches were lifeless, run-ups were unprotected from rain, express bowlers were rare, new balls were few and far between, victory relied on taking ten wickets rather than bowling dot balls, first-class captains were more experienced in the use of slow bowlers and tailenders brandished toothpick-sized bats.

Stuart MacGill, for whom tailenders comprised 39 per cent of 208 wickets, was particularly effective, with the highest proportion of any Australian bowler and one exceeded by only a single overseas player. For the games that he played overseas that figure rose to an unprecedented 45 per cent. At the SCG in 1998-99 his matchwinning figures of 12-107 included the last four wickets of England’s first innings and the final five of its second.

His potency against lower orders has enabled him to become his country’s 17th-highest wicket-taker even though he is ranked only 21st in terms of top-seven batsmen dismissed.

Next comes Shane Warne (37 per cent of 708 wickets). To his credit not only could he bowl long threatening spells to recognised batsmen, but he had the stamina to then wrap up the same innings as well. Warne averaged 3.07 wickets of top-seven batsmen and 1.81 of tailenders per match. In addition, his career statistics overseas were superior to those at home.

Shane Warne (Photo by Hamish Blair/Getty Images)

Ashley Mallett (37 per cent of 132 wickets) follows. He was far more successful away than at home, where Johnny Gleeson, Kerry O’Keefe or Terry Jenner were often preferred. Not only did he take only 63 wickets in 21 appearances in Australia, but 43 per cent of them were of lower-order batsmen. Among specialist slow bowlers only Ian Johnson has taken fewer wickets per match of top-four batsmen, and only Johnny Gleeson took fewer per game of recognised ones.

The list is completed by Clarrie Grimmett (35 per cent of 216 wickets), Monty Noble (35 per cent of 121), Bill O’Reilly (34 per cent of 144) and Richie Benaud (33 per cent of 248). Grimmett and O’Reilly, like Warne, were also prolific wicket-takers against top-orders.

As Noble and Benaud were long-serving captains perhaps it’s understandable that they often found themselves with ball in hand when an innings was nearing its conclusion. And while Noble bowled himself relatively rarely on softer or matting pitches in England and South Africa, 40 per cent of those overseas victims were tailenders.

It is rare for a pace bowler to feature among the leading takers of tailenders’ wickets. However, two recent ones have made the top-ten list – 33 per cent of Merv Hughes’s 212 wickets were lower-order batsmen, while Mitchell Starc’s figure is currently 31 per cent of 255. Each has often been let loose successfully among opponents’ last few batsmen and added cheap wickets to their higher quality earlier ones.

Mitchell Starc (Photo by Cameron Spencer/Getty Images)

Whose career figures have been enhanced least?
By comparison with Hughes and Starc, the majority of faster bowlers’ statistics have suffered by claiming far fewer lower-order scalps. Once they had removed their opponents’ genuine batsmen, their job was often considered done and they were rested. Had they removed more tailenders, their career figures would be far more impressive.

James Pattinson (a record-low 16 per cent of 81 wickets to date), Stuart Clark (21 per cent of 94), Jeff Thomson (22 per cent of 200), Geoff Lawson (22 per cent of 180) and Keith Miller (22 per cent of 170) each fall into that category. It’s therefore understandable that current selectors hold Pattinson in such high regard and keep him in cotton wool to ensure that he is always fit when most needed in a match. As with Pat Cummins, hopefully it’s an investment that will bear fruit for some years to come.

Thomson is Australia’s 18th-highest wicket-taker overall but a far more impressive 13th in terms of top-seven batsmen removed. His final tally would have been in excess of 220 wickets with an improved average and strike rate had he taken lower-order wickets as regularly as Hughes and Starc.

Not far behind Thomson, Lawson and Miller were Jason Gillespie, Bill Johnston, Craig McDermott and Bruce Reid (each 24 per cent), and Cummins, Glenn McGrath and Graham McKenzie (each 25 per cent). In comparison, no leading slow bowler has dismissed tail-end batsmen less regularly than Arthur Mailey (26 per cent of 99 wickets).

James Pattinson (Photo by Ryan Pierse/Getty Images)

Which bowlers have been most successful against top-order batsmen?
Naturally, this list is dominated by opening bowlers.

The legendary Charles ‘The Terror’ Turner deserves a special mention. Being a medium-pacer in generally weak teams during the 1880s he could bowl all day, and he often did. He gained a record 5.94 wickets per match in total, including 2.94 wickets of top-four batsmen and 4.59 of top-seven ones. As a result tailenders represent a mere 23 per cent of his entire career tally of 101 wickets from just 17 matches.

Among modern pacemen two stand out clearly. Dennis Lillee averaged 3.69 top-seven batsmen per match, including 2.37 top-four ones. And Cummins is currently averaging 2.50 top-four batsmen per match, second only to Turner. The other faster bowlers with the leading records against recognised batsmen are Fred Spofforth, Tibby Cotter, McGrath and the previously acknowledged Pattinson.

Of course there are exceptions to every rule. Some great slow bowlers in weak attacks also dismissed high numbers of top-four batsmen, most notably Clarrie Grimmett (an average of 2.22 such batsmen per match) and Bill O’Reilly (an average of 2.00). Additionally the quickish left-arm finger spinners Jack Saunders and Bert Ironmonger regularly took the new ball on wet pitches and/or against weak opponents and achieved similarly high numbers against specialist batsmen, albeit over shorter careers.

Pat Cummins (Photo by Ryan Pierse/Getty Images)

Head-to-head comparisons can be revealing
A comparison of Warne (for whom tailenders represented 37 per cent of 708 victims) with McGrath (25 per cent) and Muttiah Muralitharan is particularly revealing. Although Warne took 145 more wickets than Glenn McGrath, 121 of them were lower-order batsmen. And Muralitharan dismissed fewer tailenders than Warne despite taking 92 more wickets in total.

While McKenzie took only two wickets fewer than Benaud’s final tally of 248, his leg-spinning teammate and captain dismissed a lower-order batsman 21 more times. McGrath has dismissed 164 more batsmen than Nathan Lyon, but tailenders contribute only 13 of that figure. Thomson dismissed 30 more top-seven batsmen than did MacGill but eight fewer in total than MacGill’s final tally of 208. And while Josh Hazlewood and Hughes have each dismissed 212 batsmen, Hazlewood has 13 more top-seven victims (and therefore 13 fewer tailenders) than Hughes.

Conclusions
It’s generally accepted that it’s not possible to directly compare players from different eras because conditions and therefore raw figures have varied so widely over time. Turner’s career bowling average of 16.53 will never be surpassed, but that figure is probably equivalent to one in the low 20s in current conditions.

However, this analysis of the value of wickets does suggest that sometimes it’s also difficult to compare bowlers from the very same eras and sometimes even from the same teams.

Many faster bowlers who have ended up with seemingly inferior figures by comparison with their peers deserve credit for their greater success against genuine batsmen. McKenzie, Thomson, Lawson, McDermott, Gillespie and Miller especially stand out in this regard, particularly in comparison with Hughes and Starc.

And while many slow bowlers and especially leg spinners significantly improve their statistics at the expense of tailenders, in doing so they still perform an essential service for their team. A MacGill, Warne, Grimmett or O’Reilly who can remove tailenders quickly while other bowlers rest is immensely valuable to his side. Metaphorically speaking a great wrist spinner is myxomatosis to rabbits, not to mention the ferrets that go in after them.

Such a bowler should be developed, selected and used at every opportunity. However, it is misleading to directly compare his figures with those of other bowlers. He may well have achieved them against far inferior batsmen overall. However, that may be fair compensation for having toiled in unfavourable conditions during the first three days of a match.

In the current era a strong case can be made to select Pattinson whenever fit ahead of Starc. If Pattinson dismissed lower-order batsmen as often as Starc does, then his tally from only 21 matches to date would be in excess of 100 wickets rather than its current 81 and with a significantly better average and strike rate as well.

Sports opinion delivered daily 

   

Footnote
Orwell actually did watch a lot of cricket and played for his schools, St Cyprians and Eton, as well. He later lamented the jingoism and win-at-all-costs attitude that characterised top-level sport following the golden age of cricket and WWI, including during the Ashes series of 1921 and especially the Bodyline series of 1932-33. In 1945 he famously described sport as “war minus the shooting”.

Part 2 of this series will analyse overseas Test players similarly. While the trends are similar, some names and statistics will surprise.

The Crowd Says:

2021-03-16T02:48:46+00:00

Peter Warrington

Guest


Hey Pope that's why I reckon Mikey Death's run from 81 is one of the great unheraldeds: 81-2 5 and 6 in the Melbourne loss 5 in the Sydney draw on a reasonable deck 5 in Adelaide and then he set off the tremors with the 6-21 in Perth in 84-5. world class quick bowling especially in the context of that brilliant attack! (after that first series out here, he averaged 17 against us! take, that, Philander! and yet most of us would have Roberts, Marshall, Ambrose ahead of him.... hmmmm???)

2021-03-13T05:18:20+00:00

Just Nuisance

Roar Rookie


Thanks All day… Had a brief look for now.. Immediately noticed the observation of the dramatic improvement in dropped catches by Pakistan which the article speculates may be the fields in the UAE.. But I maintain that its more to do with the appointment of Mickey Arthur as head coach.. He was known in South Africa for his passion and adherence to fielding skills and drills.. Only real dropped catch was Australia letting him go so soon into his tenure as head coach over a real petty issue.. :silly:

AUTHOR

2021-03-12T22:15:10+00:00

All day Roseville all day

Roar Guru


Thanks JN, It just needs someone with the passion and skill, and ideally also the discovery or reconstruction of missing ball-by-ball records of some matches. Charles Davis has done amazing work on this, which has enabled him to also calculate head-to-head records all the way back to Bradman v Verity, and Grimmett v Sutcliffe and Hammond- http://www.sportstats.com.au/bloghome.html

AUTHOR

2021-03-12T22:08:43+00:00

All day Roseville all day

Roar Guru


Thanks matth, Cricinfo's Anantha Narayanan recently wrote a piece on this, published on 14 Mar 2020- https://www.espncricinfo.com/story/anantha-narayanan-are-late-order-batsmen-contributing-to-team-scores-more-today-than-in-the-past-1219153 He found that across the entire history of Test cricket, partnerships for the last four wickets have comprised 26.4% of all runs scored, with almost no variation between eras. He did note that some earlier eras were relatively low-scoring, ie those runs would have been a similar percentage but of a lower total. He also noted that a lot of these runs would have been scored by a top-seven batsman in partnership with a tailender. Steve Waugh and Adam Gilchrist come to mind.

2021-03-12T08:53:39+00:00

Once Upon a Time on the Roar

Roar Guru


Yeah, same for their batsmen like Viv Richards and Riche Richardson.

2021-03-12T08:41:52+00:00

Pope Paul VII

Roar Rookie


True Bernie. Although they'd go large when the opp presented itself - Roberts 7/54, Holding 8/92 (+6) Croft 8/29 - I remember off the top of my head. Marshall's BB was a 7 for not many I think but I mostly remem his 5/29ish off about 300 overs at the SCG on a slow track. I can't remember Ambrose's BB (was it the Perth 7?) or Walsh's. B P Patterson was frightening but alas his career was short, 5/39ish pulled him up. Bishop may have extended the dominance, chipping in with 6 cheap wickets at Perth. Unfortunately injury stopped him. I can't believe I miss these blokes so much. At the time I didn't appreciate how great they were for the game.

2021-03-12T05:46:26+00:00

Just Nuisance

Roar Rookie


Immensely readable and thought provoking article.. Weighted stats just make so much sense.. It would be nice if we can when checking a players career stats to get both the sterile numbers and the weighted ones based on say 5 criteria deviants.. . World Cricket already uses a simple algorithm when determining World rankings with away games carrying more weight than home ones as does the Duckworth Lewis System. So not new.

2021-03-12T04:51:35+00:00

Chip

Roar Guru


:thumbup:

2021-03-12T03:41:32+00:00

matth

Roar Guru


Great article. I love it when someone comes up with a different way to discuss players’ careers. My only comment is that I believe the ability to wrap up the tail has become more important in recent years as tailenders in general have become more useful with the bat (not that I have any stats to back up that feeling). Starc’s ability to cleanse up with swinging yorkers is underrated. The problem at the moment is no one, including Starc, can predict whether he’s “on” at any given moment.

2021-03-12T03:27:26+00:00

Once Upon a Time on the Roar

Roar Guru


5 fors were rare in the attack of Roberts, Holding, Garner and Croft. It was a testament to Marshall's greatness that he did actually eke out 5 fors on a regular basis despite having to share the spoils with Holding, Garner and Walsh, and then later Patterson, Ambrose and Walsh.

2021-03-12T02:56:25+00:00

Pope Paul VII

Roar Rookie


That's interesting Bernie. Bob's 8/43 seemed like 20. I think Garner's BB innings was 6/56 or thereabouts, he did have some reasonable competition though.

2021-03-12T02:53:21+00:00

Pope Paul VII

Roar Rookie


And many with less Rosey.

AUTHOR

2021-03-12T02:44:02+00:00

All day Roseville all day

Roar Guru


Thanks TLN, It's nice to be appreciated. I don't dare add up the hours taken, but it never feels like a chore. matth has gone past 100 articles, so he's the benchmark. You're right about bowling partnerships and fielding pressure. When a captain a very long time ago, I always used to stress that it was 11 against 2. I also preferred winning a match when fielding second, with all 11 of us together rather than 9 watching from the sidelines.

2021-03-12T02:18:51+00:00

Once Upon a Time on the Roar

Roar Guru


Bob Willis (325) is the only 300+ test wickets bowler never to get 10 in a match. Next highest is Joel Garner with 259.

2021-03-12T02:07:01+00:00

The Late News

Roar Rookie


Another beautiful article Roseville! Gees you must have some spare time to do the homework. I think many bowlers operate in partnership with another. How often do you see pressure at one end and wickets at the other? Don't also underestimate the value of pressure from great fielding, and of course catching!

AUTHOR

2021-03-12T02:04:27+00:00

All day Roseville all day

Roar Guru


Every bowler with more than his 310 wickets got at least one 7-fer.

AUTHOR

2021-03-12T02:01:07+00:00

All day Roseville all day

Roar Guru


He took ten 5-fers, and all but 2 of them were completed by cleaning up the tail with 3-4 wickets at the end. A fit, intimidating pace bowler can be handy at such times. Both exceptions were against NZ. In Hamilton in 1999-00 he took 5 of the first 7, and in Adelaide in 2008-09 he took the first 4. In both cases, his team-mates got the cheap late wickets. And there's the SCG in 2003-04. He took 4-201, but India declared at 7-705 thereby depriving him of a 6-fer or even a 7-fer...

2021-03-12T01:37:18+00:00

Pope Paul VII

Roar Rookie


Thanks Rosey. Is he an enigma or merely an honest, very fast, toiler? He must be the only bloke to go plus 300 and never take a 6 for ? Many accomplished bowlers had big days out but somehow our Bretty never really annihilated ( like Thommo 6/46, 6/50 ) nor doggedly chipped away to a greater statistical reward eg Tangles 8/143. He did play on some drought stricken tracks which can't have helped.

AUTHOR

2021-03-12T00:09:19+00:00

All day Roseville all day

Roar Guru


Lee took 310 wickets in 76 matches. His 1.93 wickets of top-four batsmen per match, and 2.99 wickets of top-seven batsmen, were slightly above-average. Which for a bowler in a team with McGrath, Gillespie and Warne, is very good. But noting that his side almost always took 20 wickets in each match and never picked a fifth bowler, so barring run-outs and the odd draw there were plenty of tailend scalps to share around. 27% of Lee's wickets, and 1.09 per match, were tailenders. That is pretty standard for an Australian fast bowler. Clearly more than Gillespie (0.87 per match, equivalent to 17 extra wickets in a 76-match career), but similar to McGrath (1.15 per match).

2021-03-11T23:54:56+00:00

Once Upon a Time on the Roar

Roar Guru


I think from memory BLee's blowing away the tail was mainly in one dayers though I'll take correspondence on it.

More Comments on The Roar

Read more at The Roar