THE CORDON HOME

BLOGS ARCHIVES
SELECT BLOG
January 23, 2009

Trivia - batting

A ranking system for Test openers

Anantha Narayanan
Herbert Sutcliffe
 © The Cricketer International
Enlarge

Mathew Hayden's retirement has drawn the curtains on the career of one of the greatest openers of all time. He, along with Sehwag and Greame Smith, re-defined the art of Test opening. Where do these wonderful opening batsmen stand vis-a-vis other greats like Hobbs, Gavaskar and Sutcliffe? I feel that this is the right time to do such a study.

The study of opening batsmen is a complicated task. Over the years the role of opening batsmen has changed. From defensive, stay-at-wicket-at-all-costs batsmen they have become match-winners who have been primarily responsible for the attacking attitudes which captains employ now. The study has to recognise this evolution and be fair to all types of opening batsmen.

The first task is to fix a minimum limit criteria. I have fixed this as 3000 runs, scored in the opening position (not complete career). This lets in most great openers. The only top-drawer opener left out is Hanif Mohammad (2638 runs). Unfortunately nothing can be done. I apologise to my Pakistani friends for this. I have also given at the end Hanif Mohammad's values. The other great opener left out, Victor Trumper, has scored only 1650 runs in the opening position. I wanted to avoid any longevity-based weighting and the only way is to keep a high entrance bar. The number of qualifying batsmen has also to be kept at a reasonable number, 35 in this case.

In order to cater to the different playing times, tactics, grounds et al, I have used the following 7 criteria. Each is explained in full later.

1. Home Batting Average.
2. Away batting Average.
3. Average Runs scored - weighted by the quality of bowling attack.
4. Scoring Rate.
5. Average opening partnerships participated in.
6. Quality of the top 3 pace bowlers faced.
7. Quality of batting support - Other opener and next 3 batsmen.

The principle I have followed is that the three direct measures, Home average, Away average and Average weighted runs, will carry a total weight of 50%. The other four secondary measures will have equal weight.

1. Home Batting Average (15 points).

This is the most basic of all measures. It is a straight forward computation of the home batting average. Since the minimum number of home runs scored by a batsman in the group of 35 is 1246 (by Michael Vaughan), any average figure will be valid.

The highest home average is that of Herbert Sutcliffe who has an outanding 64.60 average while playing as an opener in England. Mike Atherton of England is at the bottom with an average of 39.14.

2. Away batting Average (20 points).

This is the other basic measure. It is a straight forward computation of the away batting average. It carries a higher weighting than the home batting average for obvious reasons. Since the minimum number of away runs scored by a batsman in the group of 35 is 916 (by John Edrich), any average figure will be reasonably valid.

Away from home, the other great opener Hobbs averages 59.17. Mudassar Nazar travels very poorly with an average of 25.75.

3. Average Runs scored - weighted by the quality of bowling attack (15 points).

The first two were basic measures. However there is need to value the runs scored against better bowling attacks higher. Greame Smith should get much more credit for his knock of 154 against England as compared to his innings of 232 against Bangladesh even though both were match-winning innings and the second is 50% higher. This is done by weighting the runs scored by the bowling strength of the opposing team and averaging the same.

Hobbs' run tally comes down to 90% while Andrew Strauss' tally moves up to 109%.

4. Scoring Rate (12.5 points).

This is a new measure. The openers have changed the way the Tests are played now. First Hayden and then Greame Smith, Sehwag and Gayle et al have scored consistently at well above 3 runs per over and this has resulted in many more decisive games. This factor has to be recognized and has been.

We have accurate balls played information for the past 15 years and this can be used. For the early Tests I have assigned to the opening batsmen the team's strike rate for the innings. This might vary slightly from actual balls played information, which is, unfortunately, available nowhere. However this will even out over a career. It is also true that the olden day openers, barring a very few attacking players, played quite slowly and most of them would in reality be benefited by this methodology. For openers such as Jayasuriya, Greenidge, Haynes et al, wherever available, actual balls faced information is utilised.

The highest scoring rate for an opener has been achieved by Sehwag who has scored at an incredible 4.75 runs per over.

5. Average opening partnerships participated in (12.5 points).

This is a very good measure since it provides an indication of the effectiveness of the opener. Herbert Sutcliffe has averaged opening stands around 73 runs. The lowest figure is for Alec Stewart, around 36 runs.

6. Quality of the top 3 pace bowlers faced (12.5 points).

When the openers walk in at 0 for 0, they have a daunting task. If they reach lunch at xyz for 0, they would have done their job. Everything afterwards is a bonus. During these two hours or so, the opening batsmen are likely to face the three best pace bowlers of the other team. If these three happen to be Marshall, Holding and Garner as a few opening pairs faced during the 80s, as against the openers who faced Madan Lal, Amarnath and Solkar, they have to be given due credit.

The best three pace bowlers' averages are summed and averaged over the number of times the batsman opened.

Alec Stewart has faced the toughest pace bowlers with a low average of 27.75. A number of recent English opening batsmen have somewhat low figures since they have faced strong Australian attacks in frequent Ashes series. At the other end Hobbs, surprisingly, has had the easiest of opening stints at 37.09. Understandable since the non-English bowling between 1908 and 1930 was quite ordinary.

7. Quality of batting support - Other opener and next 3 batsmen (12.5 points).

Imagine Greenidge walking in with Haynes, with Richards, Kallicharan and Lloyd to follow. Or Langer walking in with Hayden with Ponting, Clarke and Hussey to follow. Contrast this with Gavaskar walking with the happy-go-lucky Srikkanth and P Sharma, Viswanath and BP Patil to follow. These are the extremes. This measure takes into account the supporting batsmen. The other opener gets highest weighting, followed by the no.3, no.4 and no.5 batsmen with progressive lower weightings. These proportionate averages are added and averaged. Higher credit is given for lower support averages.

It is clear that a strong bowler in a weak team has the benefit that he can take a greater share of wickets than a strong bowler in a strong team (Hadlee/Muralitharan against McGrath/Warne). Contrast this with batting where good support is always a boost to the batsmen.

As can be expected, Justin Langer has the best supporting batting with a figure of around 50. Don't forget that Langer had Mathew Hayden as the other opener. The one who had the least support is Chris Gayle with 33.63, despite the presence of Lara at no.4.

Table of top opening batsmen of all time

No Cty Batsman                 HmAvg  AwAvg AdjRpt ScRate OpPshp PaceBow BatSup

100.00 15.00 20.00 15.00 12.50 12.50 12.50 12.50

1.Eng Sutcliffe H 72.00 12.92 15.20 11.40 5.43 11.51 7.03 8.51 2.Ind Sehwag V 71.72 10.41 14.05 11.12 9.89 8.25 8.85 9.15 3.Aus Simpson R.B 70.71 10.51 15.60 11.36 5.56 10.71 8.05 8.93 4.Saf Smith G.C 69.46 9.13 15.31 10.13 7.69 10.03 7.60 9.57 5.Eng Hobbs J.B 68.70 10.46 15.78 10.08 5.98 10.05 6.45 9.91 6.Ind Gavaskar S.M 67.80 9.57 14.11 10.33 5.84 6.95 9.49 11.50 7.Eng Hutton L 67.69 11.60 14.54 10.56 5.01 8.52 7.34 10.11 8.Eng Amiss D.L 66.77 11.18 13.94 11.40 5.29 6.73 7.22 11.02 9.Aus Hayden M.L 66.26 11.58 11.38 10.33 7.51 8.91 8.53 8.01 10.Eng Boycott G 65.55 9.68 12.77 10.14 5.01 8.40 9.09 10.46 11.Eng Vaughan M.P 65.52 11.33 10.71 9.39 6.77 9.59 7.72 10.02 12.Win Greenidge C.G 65.15 9.84 11.34 9.18 6.78 9.30 9.31 9.40 13.Pak Saeed Anwar 64.61 9.27 12.72 9.44 6.97 5.80 9.14 11.28 14.Aus Langer J.L 64.42 10.15 11.98 9.97 7.24 9.27 8.50 7.32 15.Saf Gibbs H.H 64.40 9.22 12.92 9.66 6.54 8.61 8.82 8.63 16.Eng Trescothick M.E 64.22 10.21 9.63 9.40 6.81 9.52 8.48 10.18 17.Eng Stewart A.J 64.15 10.17 11.03 9.21 6.08 5.64 11.12 10.92 18.Win Haynes D.L 63.21 11.33 8.94 8.50 6.65 8.74 9.36 9.69 19.Aus Lawry W.M 63.21 11.27 10.56 9.67 5.38 9.15 7.92 9.24 20.Eng Gooch G.A 63.10 9.56 10.12 9.16 6.03 6.82 10.39 11.03 21.Win Fredericks R.C 63.06 9.22 10.68 9.18 6.24 9.69 8.58 9.47 22.Eng Edrich J.H 62.16 9.14 11.10 9.40 5.25 8.37 9.05 9.84 23.Slk Jayasuriya S.T 62.15 8.85 10.07 8.00 8.14 8.41 8.92 9.76 24.Eng Strauss A.J 61.92 8.15 11.97 9.19 6.11 7.97 9.03 9.52 25.Aus Slater M.J 61.77 10.53 9.41 8.43 6.66 8.55 9.02 9.18 26.Win Hunte C.C 61.64 10.70 10.32 9.42 5.78 8.12 7.18 10.12 27.Win Gayle C.H 61.60 7.66 11.28 8.36 7.17 7.00 8.54 11.59 28.Aus Taylor M.A 61.25 8.68 11.63 8.65 5.37 7.42 9.82 9.69 29.Slk Atapattu M.S 60.46 8.28 12.05 8.12 5.59 8.48 8.45 9.49 30.Aus Morris A.R 60.25 7.75 14.74 8.91 5.79 6.10 7.97 8.98 31.Saf Kirsten G 59.80 7.90 11.81 8.65 5.20 6.39 9.33 10.52 32.Eng Atherton M.A 59.27 8.28 9.63 8.35 4.86 6.40 10.98 10.76 33.Aus McDonald C.C 57.65 9.53 8.89 8.40 5.01 7.34 8.02 10.46 34.Nzl Wright J.G 57.11 8.40 9.00 7.48 5.08 6.03 8.98 12.14 35.Pak Mudassar Nazar 56.04 10.43 6.87 7.20 6.25 5.81 9.02 10.47

Herbert Sutcliffe's position at the top is a well-earned one. He leads in two of the key measures

	- Home average,
	- Average opening partnership and
	- Has a very good Away batting average of 57.00
He is only one of two batsmen, the other being Miandad, who has never fallen below 50 in their (reasonably long) career. He clocks in comfortably in the other measures. He however had good support (Hobbs/Hammond) at the other end. The bowling Sutcliffe faced was nothing great.

Sehwag's second position should not surprise any unbiased observer. His credentials are listed below.

	- 50+ averages both home and away,
	- Almost all his top scores have been against top class bowling,
	- He has an excellent strike rate of 4.75 rpo,
	- Has faced very good quality pace bowling almost always and
	- He has scored only around 200 runs in 5 Tests against Bangladesh/Zimbabwe.
In fact he would have been at the top if the Strike Rate measure was, say, 15.00 instead of 12.50. That would have been a worthy position for Sehwag. He has won many matches for India through his uncompromising attacking style.

Bobby Simpson is the surprise package. The main reason is that his overall batting average is only 46.82. However his opening average is 55.52, that too, 52.55 at home and 58.48 away. His opening partnerships, mostly with Lawry, averaged 68 and he faced good quality pace bowling almost always.

Then comes Graeme Smith, who is somewhat similar to Sehwag and Hayden. He has an away average of 57.43. He loses out slightly in view of the runs scored against weaker teams, and also the quality of pace attacks faced.

Then come three great openers of yesteryears. Hobbs, Gavaskar and Hutton. Each of them could have been at the top with no questions asked. All have very good averages. Gavaskar loses in the average opening partnership but gains on the pace bowling quality and a very average middle order.

Hayden has lost out a little because of the indifferent end to his career (His average dropped by 2.5 runs during the last 10 Tests). Otherwise he would have challenged for a place in the top 5.

Readers would note that the top 10 opening batsmen comprise of 3 attacking match-winning openers of today and 7 openers of the previous eras. It is clear that for any opener of today to break into the top-10 they have to be extraordinarily good, as these three have been. One does not necessarily have to score at around 4 rpo, in which case, they have to average well above 50, both home and away and do that consistently against the top sides, not just the minnows. Being part of a good opening pair and consistently putting up above average partnerships would help.

As I had indicated earlier, I have given below Hanif Mohammad's summary figures. What is very relevent is his away batting average, which, standing at 44.05, is 20% better than his Home average. Also the total lack of support batting.

Pak Hanif Mohammad   56.62   7.37  11.75   8.01   4.38   5.44   7.17  12.50

Finally a note to the readers. One factor I keep in my mind always is that each of the measures used in all my articles should be understood by all the readers, without exception. One of the reasons I try to stay away from complex statistical measures and methodologies.

Click here to view supporting information.

Anantha Narayanan has written for ESPNcricinfo and CastrolCricket and worked with a number of companies on their cricket performance ratings-related systems

RSS Feeds: Anantha Narayanan

Keywords: Trivia

© ESPN Sports Media Ltd.

Posted by Harsh Thakor on (January 21, 2010, 12:03 GMT)

Overall,Jack Hobbs was to me the best of all opening batsman.He mastered the wet pitches and even outscored the Don on those tracks.Above all he scored 12 centuries against Australia and featured in many of his team's famous victories.Hutton and Gavaskar are just below who fight for the second place in a photo-finish.Hutton's outstanding efforst on bad wickets may just give him the 2nd spot although Gavaskar faced the greatest pace bowling.At times these two batsman were over defensive and lacked match-winning instinct.Sadly,Barry Richards missed out in Test Cricket otherwise he may well have ben the best of them.No opener dominated bolwing to such a considerable extent and he proved his talent with brilliant knocks in World series Cricketin Australia in the late 1970's.Boycott although technically a master was too selfish a batsman.At his best Greenidge could have matched the very best with his form in the mid-1980's and so could Glenn Turner at his peak.

Posted by Rabin on (February 27, 2009, 13:19 GMT)

Dear Ananth,

I agree with most of the criteria used, however there should be additional criterias utilised such as batting to pass a follow on total in excess of 350 runs, batting in a fourth innings to win a match or to draw a game to save a series or to win the series. Furthermore the use of safety equipment is a significant factor. Think of Roberts/Holding and others at their fiery best at some of those venues in the West Indies in the past and one will see that Gavaskar would be rated in the top 2.

Presently there is a substantial amount of technology in use to study bowlers and batsmen to such a degree that a batsman can have the advantage to counter bowlers much more easily than in the past

Posted by Frank Matthews on (February 17, 2009, 0:54 GMT)

I can't understand people belittling Sehwag's epic 319 in Chennai against South Africa as scored on a 'flat track'. It was a 'flat track' for all the batsmen, wasn't it? How come Sehwag score a triple century when no one else scored a double century? What about quality of the SA bowling? How often has an attack containing quality bowlers like Steyn and Ntini been 'hammered'? Irrespective of the condition of the pitch and or opposition bowling line up, any triple century is a testimony to the batsman's skill, temperament, stamina and determination. Don't forget that Sehwag scored his triple at faster than a run a ball. Plus it came when India were under some pressure because of South Africa's huge first innings total. In contrast, Hayden's then world record 380 came against lowly Zimbabwe, surely the weakest team in the world then and now. Give Sehwag his due, he plays with a positive attitude, looking to score off every ball if possible.

Posted by graeme on (February 6, 2009, 11:29 GMT)

just further on qualification, I think a batsman has to play at least 3 tests in each of the 3 zones before he can qualify zone 1-sub-continent(Ind,Pak,SL,BGD), zone 2-southern hemisphere(Aus,NZ,SA) and Northern Hemisphere 3-(England,WI) that way you can see a batsman perform in a variety of conditions against different bowling attacks rather than just dominate one team as Sutcliffe did.

Posted by keyur on (February 2, 2009, 9:32 GMT)

i request you to do a post on oneday finishers(which can be defined as batsman batting in positions 5-8 since they would normally have to bat after the first 25 overs) you can keep a criteria of say minimum 50 innings or 1500 runs in those positions. while hussey,bevan,klusener should be a shoo-in at the top, i think dravid, yuvraj, dhoni, clarke, symonds too should be close...

Posted by keyur on (February 2, 2009, 9:24 GMT)

I appreciate the amount of hardwork you must have put in this article but this article has become a bit too complicated. i think the points about opening partnership or batting support are unnecessary. else every batsman opening with sehwag with dravid, sachin,laxman,ganguly to folllow should have succeeded.further opening partnership favors openers who were part of successful pairs and disfavors gavaskar,morris,etc. Finally about pace bowling, most people will agree that sehwags 300 in chennai against sa was not better than his 200 at galle although the pace bowling strength of sa was better than srilanka. i presume that the adjusted runs scored takes into account th scoring pattern of the match(high/low scoring) and the quality of opposition bowling attack-including spinners. hence i suggest that points should be calculated as follows: 20 strike rate 25 home avg. 30 away avg.(so total 55 points avg, 20 s/r) 25 adjusted run point as per opp bowling&pitch condition Total 100points

Posted by zakir on (February 2, 2009, 6:29 GMT)

i want to Bangladesh ODi batsman ranking total.

Posted by gmnorm on (January 31, 2009, 18:34 GMT)

Average opening partnership is an invalid criteria since it depends on the gent on the other end and not the person being evaluated. This is like giving extra marks to Damien Fleming because of McGrath at the other end

Posted by Aaron on (January 31, 2009, 9:36 GMT)

I feel that the way you have included strike rate bumps modern players far up the list. Langer, Gayle, Hayden, Smith, Sehwag, Jayasuriya all have ratings over 7 for this point. No one else in your table has a rating over 7 from any other era. If every great opening batsmen from this era has a good strike rate rating it suggests that a high strike rate is not unusually brilliant nowadays.

Also take a look at this example from your table: Michael Atherton: HAverage = 35 --> 8.28 points. Sutcliffe: HAverage = 64.6 --> 12.92 compared to Sutton Strike Rating ---> 5.01 Sehwag Strike Rating ---> 9.89

Your ratings system equates a difference in average of 30 to a difference in strike rate of 30. I know it's objective, but I think that is a bit farcical. The strike rate is given way too much credence in your table. [[ Aaron Your point has its intrinsic merits. However you should see that those who have been benefited are the ones who have consistently scored at above 3 runs per over. There are many modern openers who have scored well below 3 rpo have lost out. So this is not necessarily a measure which will benefit all the modern openers. As far as the olden openers are concerned they might have lost out a little in this measure but would gain on couple of other measures on Average and Opening Partnerships. Also they have been credited with the team scoring rate. In other words it is mostly likely that they would have benefited since the opening scoring would have been at a lower rate. Finally I strongly feel that the attacking openers have enriched the game to a great extent and have to be acknowledged. When Pietersen declared setting India 382 to win at just under 4 rpo, it was only Sehwag's positive thinking which won the test for India. Another modern opener or ,ost olden day openers would have aimed for a draw and a score of 40 for no loss at close play. This attitude has changed the face of test cricket and we cannot fail to acknowledge the same. If I had completely ignored the scoring rate and the lists was headed by a set of staid openers, that also would have been a farce. I certainly acknowledge the validity of your point. Ananth: ]]

Posted by Kartik on (January 31, 2009, 4:59 GMT)

Ananth,

When you did your 'Best ODI Batsman' analysis a couple of months ago, you used total career runs and career wins as factors in the player rankings.

Why not use those measures here? [[ Kartik Some one else, I think Andy, has raised this question and I have answered that comment. Ananth: ]]

Comments have now been closed for this article

ABOUT THE AUTHOR

Anantha Narayanan
Anantha spent the first half of his four-decade working career with corporates like IBM, Shaw Wallace, NCR, Sime Darby and the Spinneys group in IT-related positions. In the second half, he has worked on cricket simulation, ratings, data mining, analysis and writing, amongst other things. He was the creator of the Wisden 100 lists, released in 2001. He has written for ESPNcricinfo and CastrolCricket, and worked extensively with Maruti Motors, Idea Cellular and Castrol on their performance ratings-related systems. He is an armchair connoisseur of most sports. His other passion is tennis, and he thinks Roger Federer is the greatest sportsman to have walked on earth.

All articles by this writer