What is an all-rounder?

[War]

In light of what i would call some strange questioning of Ben Stokes credentials as potential all-rounder on this site & this stunning article of cricinfo - Blogs: Kartikeya Date: Can Kallis really be called an allrounder? | Cricket Blogs | ESPN Cricinfo, making an outlandish statistical criticism of Kallis credentials as an all-rounder - sometimes i wonder if cricket fans are guilty of over-exaggerating the main role of an all-rounder.

One poster said this in his criticism of Stokes made this assertion - Ten reasons England suck/it went wrong - PlanetCricket Forums

Stokes is a long way off being one of the best young batsmen in county cricket and he is a long way off being one of the best bowlers.

This perfect sense of a player who is good enough to make the team as a specialist batsman, or as a specialist bowler, and is used in both those roles, is realistically hardly ever been done by much all-rounders in test history on a consistent basis.

Some all-rounders may have done it i.e Sobers, Botham @ his 77-82 peak, Miller in the 50s. But most all-rounders were not that super skilled in cricket history. Majority basically had a core strength as a batsman/bowler - but their secondary skill was strong enough to elevate them into an all-rounder category. Some also had what i would call equal core strength with bat & ball.

And because of that in many cases this allowed teams to play 5 bowlers. This should not be under-rated. People tend to think 5-bowlers is a unnecessary luxury - but it really isn't, most of great teams in test history except Windies 76-91, WI 63-69 or AUS 95-2007 had 5 bowlers:

- Bodyline 1932
- Warwick Armstrong AUS of the 1920s
- Chappell's AUS of early/md 70s - Lillee/Thompson/Walker/Gilmour/Mallet
- Modern Smith S Africa, with Kallis
- ENG 1951-58 with Trevor Bailey the all-rounder supporting any 4 of Bedser/Trueman/Statham/Tyson/Laker/Lock/Wardle
- S Africa in 1970 before their ban
- S Africa in the 90s under Cronje - Donald/Pollock/Kallis/McMillian/Klusener

Then you had some very good/good teams who did it:

- Illingworth ENG of the late 60s/70s - Snow/Arnold/Greig/Underwood/Illingworth

- ENG under Vaughan circa 2003-2005 - Harmo/Hoggard/Jones/Flintoff/Giles or ENG under Hussain circa 2000-2003 - Gough/Caddick/Cork/White/Giles or Croft

- NZ under Fleming whenever Cairns was fit.

- AUS under Clarke currently

All time XI | Cricinfo Magazine | ESPN Cricinfo Most crucially when ESPN cricinfo & the panel of renowned journalist from all the nations sat down to pick the All-time XI for each team, they made a unanimous consensus to have 5 bowlers in each - even if all teams technically didn't have to go down this route.


So the aim of this thread is to try squash myths about all-rounders & by doing that i will first separate most in test history into the various categories. This IMO i think will help clear some cob-webs:

The complete all-rounders

These are the cricketers who i would say at the "peak" of their cricket careers could certainly be considered one of the top 6 batsmen in their teams & one of the top 4 bowlers:

- Gary Sobers, Keith Miller, Ian Botham

Batting all-rounders

These are the cricketers whose core strength was their batting - but the secondary ability as bowlers - made them 5th bowlers at best, but depending on conditions/team needs - their bowling could be elevating to a 4th bowling options.

- Jacques Kallis, Brian McMillan, Shane Watson, Mushtaq Mohammad, Trevor Goddard, Eddie Barlow, Tony Greig, Clive Rice, John Reid, Aubrey Faulkner, Frank Wooley,

Bowling all-rounders

These are the cricketers whose core strength was their bowling - but their secondary ability as batsmen - made them solid # 7 batsmen at best # 8 at worst. Depending on team balance or needs, some could bat as high as # 6.

- Imran Khan, Richard Hadlee, Shaun Pollock, Mike Procter, Kapil Dev, Jack Gregory, Riche Benaud, Davidson, Chris Cairns, Daniel Vettori, Intikhab Alam, Gary Gilmour, Learie Constantine, Bernard Julien, Dattu Phadkar, Heath Streak, Paul Strang

Basic all-rounders

These are the category of all-rounders who i would say whose abilities as batsmen or bowlers that were equal. This at times is even shown by the similarities in their batting & bowling averages. Due to this their roles in their respective teams, throughout their careers tended to fluctuate.

- Vinoo Mankad, Wilfred Rhodes, Trevor Bailey, Andrew Flintoff, Lance Klusener, Monty Noble, Jimmy Sinclair, Dwayne Bravo, Ravi Shastri, Craig White, Sakib Al-Hasan, Tiger Lance


Not all-rounders:

This is category that i think people get mixed up. These players usually bat @ # 8, some of of them even have test hundreds, while contributing using test runs - but they are not "bowling all-rounders" - just simply bowlers who are competent enough as batsmen to not be considered "tail-enders".

Or maybe they are batsmen who just bowl a bit & get a lot of wickets with their part-timer medium pace/spin - some of them even have 5 wicket hauls. But their bowling is strong enough to elevate them to the "batting all-rounder" status.

- Mitchell Johnson, Swann, Broad, Giles, Roger Harper, Ashwin, Brett Lee, Jason Gillespie, Harbhajan Singh, Malcolm Marshall, Warne, Vaas, Ray Lindwall, Wasim Akram, Craig Matthews, John Bracewell, Bruce Taylor, Pat Symcox, Nicky Boje, Bapu Nadkarni, Gubby Allen

Asif Iqbal, Doug Walters, Andrew Symonds, Waugh Brother, Bev Congdon, Ted Dexter, Walter Hammond, Hansie Cronje, Sanath Jayasuriya, Dilshan, Allan Border etc etc (this is a long category)

Using the above breakdown, some modern all-rounders in the game that are showing real talent that are likely to be big future players - Stokes, Corey Anderson, Ryan McClaren, James Faulkner, Andre Russell, Angelo Matthews - all of them are in the "basic all-rounder" category still. Only Matthews i would say is clearly a batting all-rounder.

However this is just my 2 cents on the matter...discuss

 

[CG123]

Vettori's an interesting one. Since 2003 (the last 10 years of his Test career) he has our 7th highest batting average, 3rd most runs and hundreds, and 4th highest strike rate (2nd highest excluding Southee and Sodhi, and his high strike rite shows that he doesn't play for his average when batting with the tail like Chanderpaul for example does).

Compare that to his bowling where his average is 8th and his strike rate is 13th (at least 20 wickets), and you could say that he's been more of a batting AR, i.e. in a given team over that period he's without doubt been one of six best batsmen but his bowling contribution has been debatable, especially on fast pitches where statically it might've been in our better interest to play four pace bowlers.

Batting - Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Bowling - Bowling records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

 

[War]

Well yes Vettori's batting in tests has certainly gotten better over the years, while his bowling has certainly either stalled/regressed.

But correct me if i'm wrong a lot of his innings & hundreds were done batting @ # 8. I always thought in recent years like Imran Khan did after his bowling fell of in 1988 - Vettori should have been batting @ # 7 or even # 6 on occasions, since NZ clearly needed an all-rounder to accommodate 5 bowlers.

Then again Vettori himself wasn't comfortable batting that high.

However now that Corey Anderson has emerged - as you said the need for him in tests in any role could be diminished already.

 

[StinkyBoHoon]

I think I said that original quote and I'll back up why I think it's a criticism. THIS IS LOOOOOONG POST so I've cut it up a bit which will probably make it even harder to understand but at least easier to swallow. I didn't intend to write in sections, maybe I should have. I have given them haughty sounding titles because I thought that would be funny.

the all-rounder today

first of all though, I think we need clear up that the game has changed a lot over the hundred years some old strategies which lent themselves to bits and pieces players have well and truely gone. the worlds smaller, cricket is professional. had england of the 50s been playing now laker would be on a permanent central contract and not dropped every now and again, india would not simply play a single seamer and bowl spinners almost exclusively through the innings.

indeed one of the most telling facets of how teams are built is that every modern team from about the 70s/80s on has an all-rounder. No one picks a wicket keeper that can't bat (like Evans for england or Tallon for australia back in the day)

so all-rounder should be looked at from a modern team selection point of view and not look to examples as far back as warwick armstrong or trevor baily or george giffen. even illingworth is pushing it.

Relativity and what is an all-rounder.

first there is relativity. shakib al hasan for example you've put in the basic all-rounder category. well, thinking about it, I think he'd probably make bangladesh on his bowling or batting. That may just be because he's from bangladesh but it's still true. I would not expect to see him dropped if he injured his back and couldn't bowl and if anything I'd lean towards his bowling as his stronger suit.

the issue of relativity is massively important when assessing an all rounder. it's possible stokes could play for india as a seamer, jadeja similarly could be a specialist spinner for south africa or england now swann has gone. these might prove a bit of stretch, I'm not quite sure stokes would muscle ahead of every indian seamer but his pace and the fact they seem to turn to ishant sharma a lot would certainly intrigue. Stokes however is so, so, so far behind the current batting depth of india he would never be in contention as a batsman for india, he may even bat behind ashwin.

so I think we have to accept that this notion of an all-rounder tends to come about because of the relative strength of the team mates around them.


Theory of Selection: Why I don't think all-rounders have a role.

the second aspect is I don't think in the case of those great teams you cited so much depended on the theory of having an all-rounder more than they were simply fortunate to have one.

this clash of theory and practicality is also important to think about. imo the best selection theory is you have 6 batsmen, 1 wicket keeper picked on ability to bat, 3 fast men, 1 spinner. (there are some little tweaks to this regarding having one or two attacking batsmen in there and a bowler who isn't bad with the bat but that's the gist) now, quite often we see teams drop superior players in order to stick to this theory. ishant sharma is not among the eleven most talented cricketers in india, the problem for india is that the 11 most talented cricketers in india are all batsmen. cricket unfortunately for them necessitates wickets, and india do need to put some bowlers on the field. in south africa around the most talented 6 or 7 bowlers are all quicks, but they need a spinner in many circumstances so find themselves benching superiorly talented cricketers in order in accomodate theories of how best to play conditions.

kallis, pollock, botham, sobers, khan, hadlee (hadlee, with an average of 27 is a very good example of someone that could only have been considered a great all-rounder due to the relative weakness of his fellow batsmen) provided the luxury of adhering to theories of team selection, but allowing them to bring something more. this particularly where I have issue with the stokes style all-rounder.

I contest he is neither relatively within the best 6 batsmen in england to play as a batsman, nor relatively within the best 3 bowlers to play as a bowler. so he violates the best way to set a cricket team out.

now, that's all well and good me saying that, but to have that opinion I must have something fundamentally against all-rounders for them not to come into my ideas of the best way to select a cricket team, so here that is.

the idea of contribution: why I'm not a fan of all-rounders


low batting averages do not suggest players that cannot reach 100 or make a telling 50. averages are not like that, it's a misleading word. averages are not really the average score, most decent batsmen don't tend to make scores of 45-55 way more than the rest of the time, they massively exceed them and then frequently fail to get near them. tendulkar, for whom there is more data than anyone else made plus or minus 5 runs of average 20 times in 329 innings. that's about 6%. They are more indicators of consistency of contribution and the significance of that contribution.

similarly all-rounders tend to have decent bowling averages but very low return on wickets per match. kallis' average of 32 is great but it doesn't tell the story that he took just over a wicket per innings. he made five 5-fers in his career, two against bangladesh (murali wishes his bowling would be afforded the lack of scrutiny kallis' is) i'm not disputing it's not useful, but out of four bowlers and 10 wickets an innings to take it's not a telling enough contribution to swing matches. they don't bowl through those long matches where someone is hammering them to all corners and they don't deal in the 4-fers and 5-fers regularly enough to provide telling contributions.

The temporal nature of the all-rounder

there is also the third problem that an all-rounder can sometimes be a fleeting thing depending on form. flintoff was a supreme all-rounder in 2004-2005 but that very much tapered off, he became a bit part strike bowler and englands form suffered massively for it. most all-rounders go through periods of being poor in one discipline, or even both, jadeja picked as an all-rounder against Oz was rubbish with the bat. his contribution with the ball was great though.

and this is the crux, they are gamble. you're not sure what sort of contribution you're going to get from them you're just hoping it's the contribution you need. add to this is that their stats show they don't really deal in telling contributions you're looking at low returns on this gamble. stokes 120 came in a match where australia had already set them a huge target and stokes took 2/82 in the 2nd innings. england could really have used a 3/40 or 4/70 or something like that. a meaningful contribution to the match with his bowling. as it happened his all-rounders contribution was a meaningless century.

when warne won the first IPL with rajasthan royals one of the things I likes was when he came in to coach them he gave all the players a card with what he expected their contribution to be. simple stuff, "restrict singles at point", "stop fours in the deep late one", "set an aggressive tone with the ball early on" for me with all-rounders you are not really sure what you want them to do and hence they're unable to deliver what you need them to do. they only do what they can.

that was a long post, that I made over a few trips to watch TV so is probably a bit all over the place. I don't doubt there'll be issues with stuff I've said because I've probably said things clumisly without maybe fully thinking them through but hopefully you the gist is there.

 

[cricket_icon]

War, you've made some interesting points and I agree with some of them but having Miller and Botham listed as complete all rounders is a bit far fetched. Botham at his peak possibly, but not Miller. And then to omit Imran Khan from that list, a man who missed so much cricket in his prime years due to injury but managed to average 50 with the bat for a decade and about 21 with the ball, is that not a complete all rounder? If t's not, then I don't know what is.

The problem with many all rounders since the end of the 80s is that none of them have managed to live up to the greatness of the 4 horsemen, Khan, Botham, Dev and Hadlee. Flintoff probably came closest, Razzak in ODIs fo a limited period of time and possibly Afridi on the off chance but no one else. Symonds had a spark but he wasted away.

The days of great tests all rounders who could average 20 odd with the ball and around 30 (career wise) with the bat seems to be long gone. Now if a test all rounder has an average of 25 with the bat and 30+ with the ball, that in itself is a luxury.

 

[War]

War, you've made some interesting points and I agree with some of them but having Miller and Botham listed as complete all rounders is a bit far fetched. Botham at his peak possibly, but not Miller. And then to omit Imran Khan from that list, a man who missed so much cricket in his prime years due to injury but managed to average 50 with the bat for a decade and about 21 with the ball, is that not a complete all rounder? If t's not, then I don't know what is.

The problem with many all rounders since the end of the 80s is that none of them have managed to live up to the greatness of the 4 horsemen, Khan, Botham, Dev and Hadlee. Flintoff probably came closest, Razzak in ODIs fo a limited period of time and possibly Afridi on the off chance but no one else. Symonds had a spark but he wasted away.

The days of great tests all rounders who could average 20 odd with the ball and around 30 (career wise) with the bat seems to be long gone. Now if a test all rounder has an average of 25 with the bat and 30+ with the ball, that in itself is a luxury.

Miller was certainly a complete all-rounder at his peak. He was genuinely a top 6 batsman & was part of the most lethal new-ball bowling pair after WW2 - in Lindwall/Miller.

If it is one is looking at his record & his criticizing him one has to acknowledge that clearly don't reflect the full wonder of Miller as an all-rounder - although it is pretty special anyway.

I'm sure you would have heard he was a fighter pilot & his perspective on cricket changed, when cricket resumed after the war by his famous statement: ""pressure is a Messerschmitt up your arse, playing cricket is not"

On Imran yes it is common knowledge that he averaged 50 & 22 (not sure the exact numbers) with the bat/ball in the final 10 years of his career. But as i've highlighted to people over the years - those 10 years was not totally as a "complete all-rounder".

When we think of Imran as a complete all-rounder i believe it is period in which he best combined competent batting - with 85mph to 90 mph bowling.

Based on my research & understanding of this period of his career & speaking to people who actually saw him bowl, this "peak" was from Karachi 1980 when he scored his 1st test Hundred to the 1988 series in WI, when his bowling was still quick. Before that 1980 test hundred, Imran was just like Warne or Vaas - a bowler who could bat.

During this period his aggregate was: 44 tests, 1881 runs @ 40.02 & 216 wickets 17.51

Still superb - but the reason i would still place him in the elite "bowling-all rounder" category, instead of "complete all-rounder" is because the majority of his runs in this period came from # 7. All his 100s & 50s during this period except his hundred @ Oval 1987 where he batted @ 6 (seemed tactical, since he wasn't a better batsman than Ijaz Ahmed) were @ 7 too.

I cut off the 1988 WI series with his bowling, because after that series his bowling declined pace wise & in its use. After the 1987 world cup semi-final he had retired - but he was coaxed back out a to tour the windies. They are no clips of that 88 tour of the windies unfortunately, but if you look at some clips of him bowling in ENG 87 or the WC - its clear he was still fast @ age 35, a few months before the Caribbean tour:

Plus i've spoken to journalist, former players (Roger Harper) & my Uncle in trinidad who saw him bowl in the 88 series & they say he was still sharp then.

However after 1988 until retirement his batting kept getting better to point where in the 1992 world cup final he was batting @ # 3, scoring his highest test score in AUS batting @ # 5 & a general top solid 6 batsman. But his bowling had clearly declined in pace & potency after 88.

This is why his batting averaged soared to 50 come the end of his test career, but clearly it didn't soar to 50 while he was still potent as a bowler.

 

[War]

@ SBH. Top post, respect your position as always, but of course there are some curious points your raised that i would certainly debate.

Yes no doubt with regards to keepers, they can't be as limited with the bat as Evans, Tallon, Wally Grout, Jackie Hendricks, Deryck Murray & still get picked. The game has certainly advanced in which keepers have to at least be averaging 30-35 consistently - especially with keepers batting as well as De Villiers, Haddin, Dhoni, Prior, De Kock in modern times.

This is why Windies have issues with Ramdin from time to time, Pakistan with Adnan Akmal & why Chris Read/James Foster couldn't have long ENG careers - since they are sub 30 averaging keepers.

However i'd argue in modern times if you can balance your team appropriately, you can still still pick a keeper like a Evans/Murray. He obviously can't bat @ # 7 - but i reckon you can bat him @ # 8 - if you have a few competent all-rounders in the XI.

A perfect example would be former S Africa keeper before Boucher, now ICC CEO - Dave Richardson. Look at this S Africa XI from his final test - 3rd Test: Australia v South Africa at Adelaide, Jan 30-Feb 3, 1998 | Cricket Scorecard | ESPN Cricinfo

You had Kallis, Klusener, McMillan, Pollock in this match & he would bat @ 8 - this team was not unbalanced due to his batting deficiencies. Nor was S Africa ever unbalanced in the early/mid 90s.

I've always believed because of that example, James Foster could have played more test for ENG if Flintoff body allowed him to play more tests.

India or Asian teams if they get the proper home conditions raging turner, would still use one quick & bowl spinners all day. India tried it vs ENG recently in Mumbai 2012 2nd Test: India v England at Mumbai, Nov 23-26, 2012 | Cricket Scorecard | ESPN Cricinfo.

It doesn't happen often as in the 60s & 70s - but IND/Sub-continent sides would try it fairly often still vs a non-asian side.

I didn't include George Giffen in all-rounders section, since along with guys like George Hirst i'm aware that they are all-rounders that wouldn't suite modern times. But i don't see why Warwick Armstong, Illingworth or Bailey type all-rounders would look out of place under modern strategies.

Fair enough if you view Sakib bowling in his strongest suite - but i think its about equal to his batting. His style is like a modern Vinoo Mankad or Wilfred Rhodes. Just like how ENG/IND used those to bat anywhere in the top 6 due to team needs - so too does Sakib bat anywhere for BANG. Unlike them though - he hasn't been tried as an opener.

Stokes could easily on current ability play for IND at all quick bowling all-rounder overseas. They tried a far less talented Irfan Pathan in that role in the past & he flopped. Plus in the S Africa series thread i've seen the regular IND posters like Aditya123, Aalay & Sairsiri regularly suggest possible seam bowling all-rounders in domestic cricket, that they may have chosen to tour S Africa.

The general consensus is indeed to balance a cricket team in modern times (especially in tests) with 6 batsman, 1 keeper, 4 bowlers (usually 3 quicks + 1 spinner or 4 quicks).

However since 20 wickets wins you test matches, this strategy only works well if your 4 bowlers are potent. Not every team has 4 bowlers of the quality of McGrath/Lee/Warne/Gillespie - Marshall/Holding/Robers/Garner - Steyn/Morkel/Philander/Kallis - Donald/Pollock/De Villiers/McMillan - Anderson/Broad/Tremlett/Swannn - Hall/Griffith/Sobers/Gibbs for eg.

This is where teams need 5 bowlers to help bridge the potency gap to help you get 20 wickets & this can generally only be done by having a all-rounder.

Modern New Zealand need 5 bowlers to take 20 wickets, because their seamers to date only look potent on green tops. Thus Corey Anderson's emergence is important for them.

Modern post Murali SRI most certainly need 5 bowlers (might argue 6) in order to take 10 wickets much less 20 these days. Only Herath is potent bowlers & all their seamers are crap. Thus they need captain Mathews to keep getting better with the ball.

West Indies could do with Dwayne Bravo become a solid test all-rounder again

IND will always need 5 bowlers bowlers overseas because their seamers are not consistent enough. They certainly wish they could put Kapil Dev in a time machine.

ENG under Duncan Fletcher & the captaincy of Hussain/Vaughan needed the 5 bowlers which all-rounders White/Flintoff helped to get 20 wickets. ENG would not have won the Ashes in 05 or in PAK/SRI 2000/01 without the all-rounder contributions of White/Flintoff.

And now that Swann is gone - ENG will need Stokes just as much as they needed White/Flintoff all-rounder contributions.

So clearly the role of an all-rounder has more use in fundamentals of picking a cricket team than you may be willing to acknowledge.

I wouldn't even bring Jadeja as an all-rounder discussion, since he hasn't performed creditably to be considered in any category. He is is still in the bits and pieces cricketer category.

On Kallis yes it common knowledge that his bowling for reasons you stated - flatter him a bit. This why for example when some people like to incorrectly compare or ludicrously say he is better than Sobers based on stats, they get found out.

I recall in the late 90s/early 2000s when Donald/Pollock lead the proteas attack under Cronje - Kallis was essentially a 4th/5th bowler. Essentially a luxury option.

Then in the circa 2001-2006 (before Steyn emerged & Ntini peaked as a bowler) when Donald retired /Pollock declined, when S Africa were in a bad period - Kallis's bowling didn't ever take up extra responsibility. Kallis technically did not take a 5 wicket haul vs a strong opponent since vs England @ Leeds 2003.

You compare that to Sobers who was generally a main bowler in a 4-man attack of Hall/Griffith/Gibbs.

However just because he didn't bowl a lot as other all-rounders or other bowlers with similar records - doesn't mean he cannot be called an all-rounder or should be criticized for it. He could still make most teams on the strength of his bowling & was just simply the most the elite "batting all-rounder" in cricket history.

As i mentioned before if all-rounders were not an essentially commodity to most sides in test history the ESPN cricinfo & the panel of renowned journalist from all the nations, who sat down to pick the All-time XI for each team, would not have made a unanimous consensus to have all-rounders in each - even if all teams technically didn't have to go down this route - All time XI | Cricinfo Magazine | ESPN Cricinfo

 

[sifter132]

I read that article by Mr Date, and replied to it I think it's a joke. Kallis is certainly an all-rounder. He may not be a 'pure' all-rounder or a 'balanced' all-rounder, or the ultimate all-rounder, but the man contributed significantly with bat AND ball. That's my definition of an all-rounder right there, can he contribute with both bat and ball?

The other thing with Kallis...if he'd been born a bit further north in Zimbabwe he would have bowled a lot more, taken more wickets per Test in a weaker attack, and would have been seen as a more balanced all-rounder. Circumstances shouldn't really be counted against him, he got 4th use of the new ball behind great bowlers like Donald, Pollock, Ntini, Steyn etc. of course he won't have taken a lot of wickets, relatively speaking.


I turned to statguru to tell me who/what an all-rounder might be.

Firstly, in my mind if you can make a century and take 5fers then you can contribute significantly with both bat and ball. 118 players in Test history have at least 1 century and at least 1 5fer. 45 players have at least 2 of each, and suddenly your list of all-rounders looks pretty good. If it's just 1 of each, then flukes like Jason Gillespie and Saeed Ajmal scoring centuries get in there.

Then I looked at it a second way:
As of a couple of days ago, statguru told me there has been 1,957,230 runs scored in Test history over 2,112 Tests. That's 926.7 runs per Test, meaning the average player would contribute 42.12 runs per Tests (926.7 divided between 22 players).
There has been 62,485 wickets taken in Test history too. Dividing that figure by 2,112 Tests and 20 players (wicketkeepers rarely ever bowl), and you get 1.479 wickets per Test for the average fielder.

Therefore if you are averaging more than 42 runs and 1.48 wickets per Test you are pulling more than your share of the weight for your team. If you are above average in both disciplines, then surely you are an all-rounder. It turns out quite a few players meet this definition, including Jacques Kallis who takes 1.76 wickets per match.

The final measure is if your batting average is more than your bowling average. After going through those 3 steps we were down to 11 players, an exclusive club of the best 11 all-rounders in history, by stats at least:
GS Sobers (WI)
Imran Khan (Pak)
GA Faulkner (SA)
KR Miller (Aus)
JH Kallis (ICC/SA)
Shakib Al Hasan (Ban)
IT Botham (Eng)
JM Gregory (Aus)
CL Cairns (NZ)
AW Greig (Eng)
WW Armstrong (Aus)

R Ashwin also qualified, but I wasn't comfortable listing him, he hasn't done anything outside India as yet.

 

[Owzat]

I would say a genuine all-rounder is someone who is as likely to perform with bat as they are with ball. Obviously all XI are required to bat, but sometimes people label someone who bats and bowls as an "all-rounder" when they offer way less with ball than bat.

Kallis is/was a very good bowler, but was clearly a far better batsman than bowler. In his last 20 Tests he took just 22 wickets, failing to take a wicket on 16 occaisions. By contrast he scored 1340 runs @ 43.23, it isn't necessarily that his bowling average was bad, but in terms of volume of runs vs wickets he was averaging barely a wicket a Test when most bowlers might expect to take 3-4, 67 runs per Test is not outstanding but closer to what you might expect.

And the classifications of batting all-rounder, all-rounder and bowling all-rounder cover most of the players who bowl and (can) bat. What Stokes might become remains to be seen, but for me Kallis (and Sobers) were batting all-rounders, Botham, Dev and Imran all-rounders although Imran Khan tailed off his bowling a bit towards the end, while the likes of Hadlee and Wasim Akram were bowling all-rounders who could make hundreds, but you'd see taking wickets far more often than perform exploits with the bat and win matches.

Averages can be a bit misleading for all-rounders, not outs for decent lower order batsmen who were frontline bowlers can lead to artificially high averages.

 

[Owzat]

Selected "all-rounders" (50+ Tests)

Ian Botham (102 Tests) : 5200 runs @ 33.55 (RPI 32.30) & 383 wkts @ 28.40 (WPM 3.75)

Keith Miller (55 Tests) : 2958 runs @ 36.98 (RPI 34.00) & 170 wkts @ 22.98 (WPM 3.09)

Imran Khan (88 Tests) : 3807 runs @ 37.69 (RPI 30.21) & 362 wkts @ 22.81 (WPM 4.11)

Chris Cairns (62 Tests) : 3320 runs @ 33.54 (RPI 31.92) & 218 wkts @ 29.40 (WPM 3.52)

Richard Hadlee (86 Tests) : 3124 runs @ 27.17 (RPI 23.31) & 431 wkts @ 22.30 (WPM 5.01)

Kapil Dev (131 Tests) : 5248 runs @ 31.05 (RPI 28.52) & 434 wkts @ 29.65 (WPM 3.31)

Jaques Kallis (166 Tests) : 13289 runs @ 55.37 (RPI 47.46) & 292 wkts @ 32.65 (WPM 1.76)

Gary Sobers (93 Tests) : 8032 runs @ 57.78 (RPI 50.20) & 235 wkts @ 34.04 (WPM 2.53)

Wasim Akram (104 Tests) : 2898 runs @ 22.64 (RPI 19.71) & 414 wkts @ 23.62 (WPM 3.98)



I could continue, but the point is that you can look at both Hadlee and Akram and see they were bowlers who could bat well on occasion, if I had time to add 5wis and 100s they both tip heavily towards 5wis (Hadlee 36-2, Akram 25-3)

On the flip side Sobers and Kallis are batsmen who were very good bowlers, but not used heavily enough bowling and not taking nearly as many wickets per match as their supposed peers. On the 100s to 5wis front it was heavily towards 100s as I'm sure you can all guess (Sobers 26-6, Kallis 45-5)

I could compare 100s to 10wms, but an all-rounder is not likely to be taking that many 10 wicket hauls, and my list includes those who have 1000 runs, 100 wkts and have scored at least one hundred and taken one 5wi so Gillespie makes the list laughably as do the likes of Kumble. If they haven't, like Marshall, Davidson and others who scored 1000 runs and took 100 wickets, then they didn't play or do enough.

Even Hadlee only took 9 10 wicket hauls, average on my shortlist is 1.7 and a few didn't take any - Kallis and Sobers the only ones I've listed above, but even Gillespie didn't manage it.

 

[War]

Hadlee fluctuated between batting @ # 7 & 8 during his career, due to circumstances & the batting strength of NZ. But the times he batted @ 7, he still averaged 32, which shows he was more a "bowling all-rounder" - rather than just a bowler who could bat.

Wasim Akram certainly had the batting talent to be a consistent bowling all-rounder, but he underachieved with the bat. He averaged sub-25 batting @ 7 or 8. So like Broad, Vaas, Warne, Johnson - his generally career efforts showed he was just a bowler that could bat.

 

[StinkyBoHoon]

The general consensus is indeed to balance a cricket team in modern times (especially in tests) with 6 batsman, 1 keeper, 4 bowlers (usually 3 quicks + 1 spinner or 4 quicks).

However since 20 wickets wins you test matches, this strategy only works well if your 4 bowlers are potent. Not every team has 4 bowlers of the quality of McGrath/Lee/Warne/Gillespie - Marshall/Holding/Robers/Garner - Steyn/Morkel/Philander/Kallis - Donald/Pollock/De Villiers/McMillan - Anderson/Broad/Tremlett/Swannn - Hall/Griffith/Sobers/Gibbs for eg.

This is where teams need 5 bowlers to help bridge the potency gap to help you get 20 wickets & this can generally only be done by having a all-rounder.

Modern New Zealand need 5 bowlers to take 20 wickets, because their seamers to date only look potent on green tops. Thus Corey Anderson's emergence is important for them.

Modern post Murali SRI most certainly need 5 bowlers (might argue 6) in order to take 10 wickets much less 20 these days. Only Herath is potent bowlers & all their seamers are crap. Thus they need captain Mathews to keep getting better with the ball.

I've snipped this for ease of quoting a bit but I'm trying to take into account everything you said. See I'm not entirely sure that having a 5th bowler always works. this is because of two reasons. first, you're right that sometimes if you have less than brilliant bowlers 4 of them working through a line up can prove a problem but I don't necessarily agree that throwing in a 5th is always going to make that easier.

for one, this isn't the 5th best bowler in the country. As I said before, averages and figures can be a mite misleading, they're indicators of how often someone makes a contribution and how telling that contribution is. Now if you had the best 5 bowlers in the country then maybe a 5th bowler would be good. it would be a nice luxury to pack an extra spinner on pitches that make break up outside the SC, as you can still have your 3 seamers but in the 2nd innings bowl spinners in tandem. It would also be nice to have extra fast bowler on low bounce pitches so they can run in harder. most of all bowlers tend to be creatures of rythm, so having an extra guy gives you an extra chance at finding someone having that rythm.

But I still think all-rounders don't provide enough of a return on that gamble because they're not the 5th best option most of the time. India had hey had yadav might have had a better chance of bowling south africa out for cheaper in the 1st test just a week or so ago, but they wouldn't have played yadav in that situation because he's not an all-rounder. rahane who was their 6th choice at batting proved to be invaluable in the next test too.

you could say "but if they'd dropped rohit..." as he was the only middle order batsman that failed that tour, but it's all hindsight, you don't play 6 batsman hoping they all score their average of 50 to give you 300. you play them because out of the 6 a couple of them are likely to get in and make a big score. it's more about playing percentages of success than combined averages.

I did a funny thing where I looked up kallis' contributions to wins and losses and found he took roughly 180@25 in wins, and 110@42, now the problem there is it looks like he was a help to south africa in wins, but he was still on the pitch in those losses/draws, and proved unable to make a telling contribution. I decided after I'd done it was a meaningless exercise and proved nothing haha.

an interesting thought experiment would be though if you had two teams, each that were roughly equal, both had the option of playing an average all-rounder (no geniuses like miller, kallis or sobers) at 6 or a decent 6th best batsman in the country with an average a good 15 runs or so higher. then maybe if both played the all-rounder the all-rounder would in each team would become redundant because with a weaker no.6 they wouldn't need to bowl through 6 good batsmen, they'd only need to bowl through 4-5 and get 5-6 of the weaker wickets so it might be an advantage to play the 6th batsmen if the other played their all-rounder. in this case no one wants to be out-manouvered so both teams would thus play the 6th batsmen to ensure some form of safety.

possibly somewhat dubious reasoning and by no means flawless but just chuking it out there.

so yeah, I still think there is a place for the exceptional all-rounder, the botham or the miller or sobers, but I think an average batsman has certain advantages an average all-rounder can't provide. this is all predicated of course on what I said earlier about the relative strength of the all-rounder to the rest to the team and of course, making a judgement of a players quality is never going to be as simple as this in the reall world. it's always going to be contentious.

As i mentioned before if all-rounders were not an essentially commodity to most sides in test history the ESPN cricinfo & the panel of renowned journalist from all the nations, who sat down to pick the All-time XI for each team, would not have made a unanimous consensus to have all-rounders in each - even if all teams technically didn't have to go down this route - All time XI | Cricinfo Magazine | ESPN Cricinfo

yeah, that's a pretty fair point I guess. If I was to have a go at explaining it though it would be because across the history of cricket you are going to get that exceptional all-rounder at some point or perhaps such an exceptional wicket-keeper to facilitate the luxury of an all-rounder. (I do think all-rounders have a place when you have a luxury player like de villiers is now for SAF, which is why I reckon it will be Mclaren coming in at 7 for them now)

for example in that australian all-time XI, if miller didn't exist, you have the option of either bringing ponting out of the cold or promoting gilchrist to 6 and bringing in Benaud. playing Miller or performing a re-shuffle for Benaud are only possible because we're picking players from the pantheon of australian cricketers.

same with sri lanka, only possible to play Vaas at 7 because Sangakkara was such an awesome wicketkeeping-batting selection. Vaas rarely played at 7 and when sanga gave up the gloves, was a permanent no.8.

no sobers then would we really pass over the three Ws to play Constantine or Hooper. No imran and so we'd pick Mushtaq Mohammad or Intikhab Alam over mohammad yousuf, I wouldn't anyway. (actually Mushtaq Mohammad doesn't look bad on paper, I only really heard of him through that cricinfo all time thing when I was looking there)

That's just be nitpicking though, I suppose though what the all-time XI shows is that if you have a very good, or great, all-rounder, you should look to incorporate him, but I think good or great all-rounders are in short supply and I still lean slightly towards the average batsman over the average all-rounder. perhaps though if it was the option between a great batsman and a great all-rounder the latter has the upper hand.

one thing mind, stokes was certainly worth his place against australia despite my earlier quote. he looked the best batsman on the tour, let alone some choice for balance.

 

[puddleduck]

Rob Key.

 

[cricket_icon]

Miller was certainly a complete all-rounder at his peak. He was genuinely a top 6 batsman & was part of the most lethal new-ball bowling pair after WW2 - in Lindwall/Miller.

If it is one is looking at his record & his criticizing him one has to acknowledge that clearly don't reflect the full wonder of Miller as an all-rounder - although it is pretty special anyway.

I'm sure you would have heard he was a fighter pilot & his perspective on cricket changed, when cricket resumed after the war by his famous statement: ""pressure is a Messerschmitt up your arse, playing cricket is not"

On Imran yes it is common knowledge that he averaged 50 & 22 (not sure the exact numbers) with the bat/ball in the final 10 years of his career. But as i've highlighted to people over the years - those 10 years was not totally as a "complete all-rounder".

When we think of Imran as a complete all-rounder i believe it is period in which he best combined competent batting - with 85mph to 90 mph bowling.

Based on my research & understanding of this period of his career & speaking to people who actually saw him bowl, this "peak" was from Karachi 1980 when he scored his 1st test Hundred to the 1988 series in WI, when his bowling was still quick. Before that 1980 test hundred, Imran was just like Warne or Vaas - a bowler who could bat.

During this period his aggregate was: 44 tests, 1881 runs @ 40.02 & 216 wickets 17.51

Still superb - but the reason i would still place him in the elite "bowling-all rounder" category, instead of "complete all-rounder" is because the majority of his runs in this period came from # 7. All his 100s & 50s during this period except his hundred @ Oval 1987 where he batted @ 6 (seemed tactical, since he wasn't a better batsman than Ijaz Ahmed) were @ 7 too.

I cut off the 1988 WI series with his bowling, because after that series his bowling declined pace wise & in its use. After the 1987 world cup semi-final he had retired - but he was coaxed back out a to tour the windies. They are no clips of that 88 tour of the windies unfortunately, but if you look at some clips of him bowling in ENG 87 or the WC - its clear he was still fast @ age 35, a few months before the Caribbean tour:

Plus i've spoken to journalist, former players (Roger Harper) & my Uncle in trinidad who saw him bowl in the 88 series & they say he was still sharp then.

However after 1988 until retirement his batting kept getting better to point where in the 1992 world cup final he was batting @ # 3, scoring his highest test score in AUS batting @ # 5 & a general top solid 6 batsman. But his bowling had clearly declined in pace & potency after 88.

This is why his batting averaged soared to 50 come the end of his test career, but clearly it didn't soar to 50 while he was still potent as a bowler.

The fact that you say Miller/Lindwall was the best fast bowling pair post WW2 shows clear bias. And, you've given a period of about 8 years for Imran at his peak, in which he quite honestly bettered anything Miller had done in his whole career. He was nowhere near the pace or potency of Imran Khan as a fast bowler, how many batsman of Khan's quality can say at their best, they were also one of the world's leading fast bowlers, in average, sr and pace. Khan in his generation has Lillee, Holding, Marshall, Garner and co and yet, in many cases, especially on docile subcontinental wickets, he was the best of the lot. He also has 6 test centuries to his name.

Come on, there isn't a comparison, at the top of the allrounder tree there is Gary Sobers, Imran Khan and Jacques Kallis.....followed, quite a few branches down, by Botham, a little further down by Hadlee and Dev and then Miller. Which isn't bad at all and I'm not discounting Miller but you can't mention him as a complete allrounder and then omit someone like Khan.

 

[War]

The fact that you say Miller/Lindwall was the best fast bowling pair post WW2 shows clear bias. And, you've given a period of about 8 years for Imran at his peak, in which he quite honestly bettered anything Miller had done in his whole career. He was nowhere near the pace or potency of Imran Khan as a fast bowler, how many batsman of Khan's quality can say at their best, they were also one of the world's leading fast bowlers, in average, sr and pace. Khan in his generation has Lillee, Holding, Marshall, Garner and co and yet, in many cases, especially on docile subcontinental wickets, he was the best of the lot. He also has 6 test centuries to his name.

Come on, there isn't a comparison, at the top of the allrounder tree there is Gary Sobers, Imran Khan and Jacques Kallis.....followed, quite a few branches down, by Botham, a little further down by Hadlee and Dev and then Miller. Which isn't bad at all and I'm not discounting Miller but you can't mention him as a complete allrounder and then omit someone like Khan.

Just to clarify, when i said Lindwall/Miller were the best new-ball pair bowling pair post WW2 - you do realize i was refering the the late 1940s & a large portion of the 1950s right?

I was by no means suggesting that they were the best new ball pair, since cricket resumed in 1946.

Yes Imran at his actually peak had better numbers than Miller - but as i said, you got to look beyond the numbers.

As i clearly showed at his 8 year peak, when he combined competent batting to his legendary fast bowling - Imran's batting was only good enough to bat @ # 7 - although he averaged 40. He never took up the responsibility or in some cases was simply not good enough during those 8 years to bat in the top 6. His batting got better after 1988 - when his bowling declined.

Miller was a genuine top order batsman - in fact in you check his career out of his 87 test innings - only 7 were @ #'s 6 or 7. And he combined this genuine top order batting with being a lethal new ball bowler.

So quite clearly Miller came into test cricket the full all-round package - Imran started as a bowler, evolved into a all-rounder, then finished as a batsman.

Its a matter of opinion if you think Imran was the best bowler of his time, since i'd give Marshall & Lillee at their best a slight edge over him.