When Jack Hobbs and Herbert Sutcliffe batted at the top of the order for the first time for England in 1924, only one opening pair had managed to put together more than 1000 runs. Hobbs had been a part of that association as well, scoring 2146 partnership runs with Wilfred Rhodes at a fantastic average of 61.31. Thus, the bar had already been set high, but over the next six years Hobbs and Sutcliffe would raise it higher still.

In their first innings together, at Edgbaston, they put together 136 for the opening wicket in a match England won by an innings and 18 runs. In the next game, at Lord's, they added 268, which at the time was the second-highest opening stand in Test cricket.

Over the next six years the pair had phenomenal success. By the time they batted together for the last time at the top of the order, against Australia at The Oval in 1930, they had amassed 3249 runs - a record for opening pairs at the time - at the staggering average of 87.81, with 15 century partnerships in 38 innings, which works out to an average of one every two and a half innings.

Other pairs came along later and scored more runs, but none, till today, have scored them at such a prolific rate. Among opening partners who've scored more than 3000 partnership runs, the average achieved by Hobbs and Sutcliffe remains about 44% better than the next-best pair (Bill Lawry and Bob Simpson, average 60.94). Only one opening pair has more century partnerships - Gordon Greenidge and Desmond Haynes, with 16, but they batted together 148 times to achieve it.

How best to judge Hobbs-Sutcliffe among the other great opening pairs in Test cricket? There are eight others who've scored more than 3000 runs, with averages ranging from 61 to 40. Greenidge and Haynes till a week ago held the record for most Test runs by any pair of batsmen. Their average stand is relatively low, but is that partly because they faced better bowlers? Were the amounts of runs scored during any of those periods that these nine pairs played in much higher than in other periods?

The table below considers two things. First, the average partnership for all wickets in all Tests for each of the periods that these pairs played in, and then a comparison of each pair's average stand against that. So, for instance, during the period that Hobbs-Sutcliffe opened the batting, the average partnership across all wickets in all Test matches was 35.33, which means Hobbs-Sutcliffe were 2.49 times better than the average partnership. That's easily the best ratio among all nine pairs - they're about 35% better than the second-best, Bill Lawry-Bob Simpson. The Indian pair of Sunil Gavaskar and Chetan Chauhan climb up to third place in this ranking, ahead of Virender Sehwag and Gautam Gambhir, despite a lower average, because the average partnership during the Gavaskar-Chauhan era was about three runs lower than in the current age, where bat dominates ball. The overall partnership average was, in fact, highest during the Hobbs-Sutcliffe era, but the numbers in today's age aren't much lower.

This takes care of the overall batting numbers during each period, but it doesn't answer the argument that certain pairs from stronger teams were privileged because they racked up those numbers without having to face their own bowling attacks. Hence there's the column that looks at opening partnerships during the period in which the pair played, against all teams except their own. That works well for most pairs, but not for Hobbs-Sutcliffe, as England were involved in every Test match played during that period. During the period in which Hobbs and Sutcliffe were openers, England had 13 other pairs who opened the batting, and the overall average stand in all the Tests that England played during this period was 61.16.

That average is largely because of the outstanding numbers from Hobbs-Sutcliffe, which means their ratio diminishes because of their own brilliance, but it works in other cases. Take, for instance, Mark Taylor and Michael Slater: during the period in which they opened, the average opening stand against all teams other than Australia was 35.41, while Slater-Taylor averaged 51.14, a ratio of 1.44. Matthew Hayden-Justin Langer averaged marginally more, but in an era when openers generally did well against teams other than Australia. Their ratio thus falls, as do the ones for Sehwag-Gambhir and Alastair Cook-Andrew Strauss, because they've all prospered in an era that has been relatively good for opening batting.

Openers with 3000+ partnership runs

Pair	Innings	Runs	Average	100/ 50 stands	Overall ave*	Ratio	Opening stand ave^	Ratio
Hobbs-Sutcliffe	38	3249	87.81	15/ 10	35.33	2.49	61.16	1.44
Lawry-Simpson	62	3596	60.94	9/ 18	33.23	1.83	38.40	1.59
Chauhan-Gavaskar	59	3010	53.75	10/ 10	31.98	1.68	38.03	1.41
Gambhir-Sehwag	65	3562	57.45	10/ 19	34.96	1.64	40.56	1.42
Slater-Taylor	78	3887	51.14	10/ 16	31.59	1.62	35.41	1.44
Hayden-Langer	113	5655	51.88	14/ 24	34.66	1.50	42.89	1.21
Greenidge-Haynes	148	6482	47.31	16/ 26	32.59	1.45	38.11	1.24
Cook-Strauss	97	4147	43.65	11/ 16	35.17	1.24	41.28	1.06
Atapattu-Jayasuriya	118	4469	40.26	9/ 24	33.13	1.22	38.25	1.05

* The overall average partnerships for all wickets in all Tests during the period in which the pair played
^ Opening partnerships during the period in which the pair played, against all teams except their own

The Hobbs-Sutcliffe combination soars above the rest on just about any parameter. In only 38 innings they had 15 century stands, an average of 2.53 innings per century partnership. The next-best among these nine pairs is Gavaskar-Chauhan, with an average of 5.9 innings per hundred partnership. Gambhir-Sehwag and Lawry-Simpson are the only others who average less than seven, while Greenidge-Haynes averaged 9.25.

The failure rate for Hobbs-Sutcliffe was also remarkably small: in 38 innings, only six times was their partnership terminated for less than 20, which works out to less than 16% of the total innings when they batted together. Greenidge-Haynes, on the other hand, had 60 partnerships of less than 20 (excluding not-outs), out of their 148 stands, a percentage of 40.54, while for the Sri Lankan pair of Marvan Atapattu and Sanath Jayasuriya, it was more than 44.

The standard deviation for each of these pairs measures how close each partnership was to its average. Dividing the average by this standard deviation gives a measure of the consistency of a pair. Hobbs-Sutcliffe are again way out in front, with many of the other pairs bunched together.

The home-away stats for these openers bring out more contrasts than the standard deviation. Most pairs have done better at home, which is as you'd expect, and a couple - most notably Cook and Strauss - have done better overseas, but the biggest surprise is the difference in home and away partnership averages for Greenidge-Haynes. In 65 partnerships at home they averaged more than 65, with 10 century stands, which included three double-hundreds. However, overseas the average dropped to a below-par 35.51, with only six century stands in 85 innings. Breaking it up further, in Australia they managed only one three-figure partnership in 28 innings, while they fared slightly better in England, with two century stands in 19 tries. The subcontinent was the toughest of the lot for Greenidge and Haynes - in 26 innings they averaged 28.65. Pakistan's fast bowlers, especially, clearly had the better of them in home conditions, allowing them only 207 runs in 10 partnerships.

Greenidge and Haynes clearly preferred home conditions, but for Cook and Strauss the opposite seems true - their home average is seven less than their overseas one. For Gavaskar and Chauhan, it didn't seem to matter much if they played home or away - they're the only pair whose averages differed by less than one. They didn't enjoy Australian conditions much either, averaging 41.30 in 13 innings, but they more than made up for it in England, scoring 453 runs in seven innings.

The two post-1990 Australian pairs in this list, Slater-Taylor and Hayden-Langer, have similar home and away stats, and both were less than comfortable in the subcontinent: in 18 innings, Hayden-Langer averaged 39.33, while Slater-Taylor averaged 37.27 in as many innings.

The first instalment of this series featured the West Indian fast bowlers of the 1970s, '80s and early '90s. Opening the batting against them would have clearly been perhaps the toughest task in international cricket at the time. So to finish this piece, here's a look at the opening pairs who had the unenviable task of facing those bowlers most often, and how they fared against them.

Among the pairs who played the West Indian pace attack at least eight times in the period mentioned below, the one with the best average was Michael Atherton-Alec Stewart. In nine innings they averaged 46.11, despite three ducks. That's because they also had two century stands, including 171 in Barbados, which helped England win. Pakistan's Majid Khan and Sadiq Mohammad were superb too in 1977, averaging 46 against an attack that included Joel Garner, Andy Roberts and Colin Croft.

Geoff Boycott and Graham Gooch managed an even more creditable feat, facing the West Indians in two series and maintaining an average of more than 40. Gavaskar and Anshuman Gaekwad played them across three series, but their numbers were far less impressive: 608 runs at 27.63.

To check out the stats of opening pairs against Australia during their peak years, between November 1994 and May 2008, click here.)

Best opening pairs against West Indies in their heyday* (Qual: 8 innings)

Pair	Innings	Runs	Average stand	100/ 50 p'ships
Atherton-Stewart	9	415	46.11	2/ 0
Majid Khan-Sadiq Mohammad	10	368	46.00	1/ 2
Boon-Taylor	8	297	42.42	1/ 1
Boycott-Gooch	18	688	40.47	1/ 4
Broad-Fowler	8	244	30.50	1/ 1
Marsh-Taylor	13	361	27.76	1/ 1
Gaekwad-Gavaskar	22	608	27.63	2/ 3

* Between March 1, 1976 and February 28, 1995, and excluding the period between April 1978 and November 1979