Stats Analysis

Which Test records are the most unlikely to ever be broken?

From Jim Laker's match haul of 19 for 90 to Australia's 16-Test winning streak and Sri Lanka's total of 952 for 6, a look at various records and the probabilities of them getting bettered

Anantha Narayanan
Anantha Narayanan
06-Sep-2025 • 17 hrs ago
Jim Laker walks back after taking all ten wickets, England v Australia, 4th Test, Old Trafford, 5th day, July 31, 1956

Jim Laker's ten wickets in an innings was equalled by Anil Kumble and Ajaz Patel, but his 19-wicket match haul will likely remain unsurpassable  •  PA Photos

This is a quirky and offbeat article, one that has been a real fun piece for me and should be likewise for anyone who reads it.
I will look into some of the major Test cricket records and speculate, often in the fourth dimension, on the possibility of the record being equalled or broken. Let us consider the percentage chance of a record being broken. I will use the Test as the unit of consideration. A 1.0% chance converts to once in the next 100 Tests - that is, approximately during the next two years. A 0.1% chance converts to once in the next 1000 Tests - that is, approximately during the next 20 years. A 0.01% chance converts to once in the next 10,000 Tests - that is, approximately during the next 200 years. That's assuming an average of 50 Tests will be played per year, for many years to come. And assuming that Test cricket survives.
Another clarification. Beating the record, in most cases, is exactly what it means. A record has to be beaten, not just equalled. That means 401 runs have to be scored in an innings or ten wickets have to be captured for 52 runs or less. Let us accept the fact that Anil Kumble and Ajaz Patel have not beaten Jim Laker's record, they have only equalled Laker's feat of capturing ten wickets in an innings. The record is still with Laker. However, there are very tough records that could be equalled, at best.
I will not assign percentage chances to all the records, since that is basically shooting in the dark. I will do so only for the top few. I will look at each record to see what needs to happen for the same to be broken or equalled and come to a conclusion, mostly subjective in nature. Also, it is important to understand that we can never assign 0.0% probability to any event. That's the reason ESPNcricinfo's Forecaster assigns a 0.01% chance of a win for a team, even if, say, that team requires to score 100 runs in one ball. So, I will assign a 0.01% for that record which I think will never be broken, but present the others in order of the possibility of the record not being broken.
I am also going to group a set of records together in the order of their non-occurrence. The first group contains a single record only. Please remember that all the records within a single group carry equal weight in terms of their "beatability". Let me first start with the one record that I expect will never ever be broken. If, Test cricket survives, say until 2188, this record will still be there.

Group A

A. Wickets in a Test: Jim Laker's 19 for 90, 1956
Laker did not only get 19 wickets in a match. He laid down a marker so high that it is my firm belief that this is the one record that will never ever be beaten. To beat this record, a bowler has to take all 20 wickets, or capture 19 wickets for fewer than 90 runs. Consider the realities of this situation.
If the bowler bowls 50 overs in the match, his compatriots should not capture a single wicket in their 50 overs, there cannot be a run-out, the DRS has to work in that bowler's favour, and the opponents have to cooperate. One late-order batter could swing and get out, deliberately or otherwise. And to top it all, in 50 overs, the number of runs conceded would cross 90. If I reduce this to 30 overs, the pitch would have to be so diabolical (a wicket every nine balls for that bowler) that the other bowlers would have to be below club level to not capture even a single wicket. As a token number, I would assign 0.01% to the chance of this event happening.

Group B

I also don't expect these records to ever be broken.
B1. Stumpings in a Test innings: Kiran More's five stumpings vs West Indies, 1988
In nearly 150 years of Test cricket, only one wicketkeeper has stumped five batters in a Test innings. Two other keepers have stumped four batters. In the last 20 years, only two keepers have stumped three batters each. These three statements should highlight the enormity of More's achievement. Nowadays when stumpings are not that common, this feat is all the more a mirage and miles too far for today's keepers. In almost all cases, stumpings are effected off spinners and that breed is not flourishing. I think that even four stumpings is so far off that five stumpings, leave alone six needed to beat More's record, is at a 0.05% chance of being broken. It is worth mentioning that there has been just a single instance of six stumpings in a first-class innings.
B2. Low scoring rate in an innings: New Zealand's 69 of 6 in 90 overs (at 0.67 RpO) vs Pakistan, 1955
A minimum of 50 overs has been mandated for the scoring-rate records. New Zealand saved an amazing Test by batting out the better part of the fifth day while scoring a mere 69 runs. I can never imagine a situation in which this record will ever be broken. Let us speculate. A team is set 300 to win in 50 overs. They do not have the batters to go for the target. They tell themselves to be bloody-minded about the whole situation. They have the defensive batting skills. They play out the 50 overs and score 30 for x.
That is cuckoo land. I cannot ever imagine that happening. Equally unlikely is a last-day score of 65 for x in 90 overs. So, I should say that this record should be assigned the same 0.05% - in other words, nearly never ever. Let me take this opportunity also to tip my hat to the South Africans who faced 143.1 overs to score 143 runs, almost saving a Test against India in Delhi in 2015.
B3. Consecutive team wins: Australia, in 1999 and in 2005
First, under Steve Waugh, Australia won 16 Tests, home and away. Then a few years later, Ricky Ponting led the Australians to another sequence of 16 wins. The question here is whether any team, even a strong Australia, would win 16-17 Tests in a row again. With the way the WTC cycle is structured, the 16-17 Tests would form one complete WTC cycle. Imagine a team going through a whole championship or two halves of different WTC cycles with a 100% win record. I would say that there is almost zero chance of that happening.
B4. Consecutive team losses: Bangladesh
Bangladesh lost 21 Tests on the trot between 2001 and 2004. Nothing is gained by analysing that sorry sequence. I only need to speculate whether any team would lose 21-22 Tests in a row in the future. I cannot see the top teams incurring this long series of losses. And the weaker teams, out of the WTC, may need quite a few years to even play these many Tests. Also, the weaker teams would play among themselves often, so I would say it's extremely unlikely to happen. For both these records, equalling the current one is a monumental achievement since the period involved is long - three to four years.

Group C

Let me now move on to the category of records where there is very little chance these could be broken in the next decade or two.
C1. Team record score: 952 for 6 by Sri Lanka vs India, 1997
Sri Lanka, facing a total of 537, batted with one objective: to grind the Indian bowlers to dust and go past 903. No thoughts about a result, which they knew was impossible on a Colombo road. I am not faulting them. That was nearly 30 years back. The word Bazball did not exist.
How would this record be beaten? Will Premadasa be repeated? I doubt it. The teams would be crucified for batting that long. It is clear that the only way this could happen may be in a Multan 2024 type of situation, where England made 823 for 7 in response to Pakistan's 556 and won by an innings. The first batting team scores 200 or so and there is a Bazball explosion of 200 overs in which 952 is crossed. A day and a half is still left for a result. Maybe England, on a road to beat all roads, against an average bowling team. The chances of this scenario unfolding: very, very low.
C2. Highest partnership: 624 by Kumar Sangakkara and Mahela Jayawardene vs South Africa, 2006
In over 150 years, a single partnership has exceeded 600 runs, and another one touched the 500 mark. That indicates how tough it is to cross these high levels. It would require between 150 and 200 overs, which is very tough to fit within an innings. The only way for this to happen is a repetition of the scenario outlined for the team total situation - maybe if Joe Root and Harry Brook had got together in Multan, not at 249 for 3, but at 100 for 3. The chances, again, are virtually non-existent. The quantum of runs needed to break this record is really huge.
C3. Innings bowling performance: Jim Laker's 10 for 53 vs Australia, 1956
I have already referred to the problems in a bowler picking up all ten wickets. Everything has to be in place and work favourably. However, the fact that this feat has happened three times in the past 70 years makes me think that it could happen again. But for the record to happen, the runs conceded have to be below 53. However, the bottom line is that this is not an insurmountable record and could be beaten in the years to come. The fact that four nine-wicket hauls have happened with fewer than 53 runs conceded is also relevant.
C4. Catches in a Test innings: Five fielders effecting seven catches
This is somewhat similar to the stumping record. However, the record of seven catches in an innings has been reached by no fewer than five fielders, all them being wicketkeepers. As such, it is possible that this record will be reached by keepers in the near future, especially as the pace bowlers are dominating the bowling scene in Test cricket. However, let us not forget that the record is broken only when it is crossed, not just equalled. But equalling this record in itself is something monumental.
C5. Innings win margin: by innings and 579 runs by England vs Australia, in 1938
There are two scenarios in which a team can win by an innings and 580 runs. Team 1 - 750, Team 2 - 100 and 70 or Team 1 - 100 and 70, Team 2 - 750. Minor variations in scores are possible. What are the chances of these scorelines happening? Recently, Zimbabwe lost by an innings and 359 runs to New Zealand - on that occasion, the winning team scored in excess of 600. Let us not forget that the next-highest innings win is more than 200 runs below the record - Australia's innings-and-360-run win over South Africa in 2002. This indicates how difficult it is to get a really huge innings win. I would say, extremely unlikely.
C6. Runs in series: 974 runs by Don Bradman vs England, 1930
Since 1994, the highest series aggregate has been 798. That is nearly 20% off Bradman's record. This indicates how tough it is to score nearly 1000 runs in a Test series. Recently, Shubman Gill turned the cricket world upside down when he accumulated 585 runs in two Tests. Suddenly everyone was talking about Bradman's aggregate, over 1000 runs for the first time ever, and the like. Then the bottom fell off. Gill managed only 169 runs in the next three Tests and the aggregate was a very good 754, but way short of the Don's number. This example indicates that scoring 200 runs per Test, across a five-Test series, is not a cakewalk. Especially as only the Big Three are going to play five-Test series. So the record is likely to stand for a very long time.
C7. Wickets in series: 49 wickets by Sydney Barnes vs South Africa, 1913
In the last 20 years, the highest series haul is 40 wickets. That was Shane Warne in the iconic Ashes 2005 series. That is around 80% of Barnes' mark. Like Bradman, Barnes has also set the bar at a very high level. And recently, there have been no Gill-like bowling performances. So let me say that Barnes' mark is as safe as Bradman's. Two real giants of the game, indeed.

Group D

There is a fair chance that these records will be broken in the near future.
D1. Lowest team total: 26 all out by New Zealand vs England, 1955
A couple of months ago, West Indies were 26 for 9 against Australia in Kingston, and could easily have equalled New Zealand's record. In 2020, Australia dismissed India for 36 in Adelaide - a mere ten runs more than what New Zealand fell for. A year before this incredible event, Ireland collapsed for a total of 38. Not to forget that there were seven other scores between 40 and 50 after the millennium dawned. All these innings make me think that the 26 is not as safe as it looks. Maybe, in the next few years, if everything aligns - an overcast sky, a devastating collection of bowlers, some inept batting, and a couple of favourable DRS decisions - this record may go for a toss.
D2. Win margin by runs: by 675 runs by England vs Australia, 1928
There is only one scenario in which a team can win by, say, 700 runs. Leaving aside the first three innings, this requires a target of around 800 to win and the unfortunate team batting last folds for 100 or so. I looked at the targets set by teams. Only four targets in excess of 700 have been set until now, with 741 being the highest. That resulted in the 675-run win. It is necessary to mention that the next highest run-margin is over 100 runs away - a win by 562 runs. So, the target needs to be around the 800-mark. The chances of that happening are quite low, but possible.
D3. Seventh-wicket partnership: 347 by Clairmonte Depeiaza and Denis Atkinson for West Indies vs Australia, 1955
The next two are totally different records in that it is not the quantum of runs but the fact that one of the batters is a late-order one that makes these records tough to beat. Maybe the one reason this record has found its place here is because this is among the longest standing records currently - it has survived for 70 years. A score of 150 for 6 warrants this type of partnership. These situations will occur often. However, it is clear that one of the batters is the designated No. 8. That makes this record very tough to beat. The longevity of the record makes it a candidate for placement here. The nearest this partnership has come to being overhauled was when Waqar Hasan and Imtiaz Ahmed added 308 against New Zealand in Lahore a few months after it was set by the West Indian pair.
D4. Eighth-wicket partnership: 332 by Jonathan Trott and Stuart Broad for England vs Pakistan, 2010
The situation is quite simple. A score of 150 for 7 warrants this type of partnership. These situations will occur often. However, what is the possibility of a five- to six-hour stand by two batters, possibly Nos. 8 and 9? One possibility is a nightwatcher. That makes the junior partner the No. 8 batter. I would say that this record is very difficult to break, but no more than that. Stranger things have happened. Let us not forget that two partnerships above 250 have come in the 2000s. Also, Wasim Akram and Saqlain Mushtaq added 313 runs against Zimbabwe in 1996.
D5. Tenth-wicket partnership: 198 by Joe Root and James Anderson for England vs India, in 2014
I would say that this is a really tough record to beat. It is clear that one of the batters has to be a top-order one. However, the other batter must be a designated No. 11, a la Anderson. What are the chances that these two last over four hours and around 60 overs? However, I am not putting this in a more unlikely category simply because there was a 163-run partnership in 2013 and a 143-run partnership in 2012. So let me say only say that the chances are quite low.

Group E

Finally for the group of records that are likely to be beaten in the near future.
E1. Individual score: Brian Lara's 400 vs England, 2004
Two months ago, Wiaan Mulder had Lara's record in his hands, however, out of some quirky, but admirable, sense of respect for the great batter, he gave up on breaking the record. It is very likely that the next batter in a similar situation may not think likewise. It is clear that this opportunity will be there in the future. Imagine a team being dismissed for a sub-200 score and a perfect batting track is supplemented by an inept bowling attack. About 150 overs and a score of 650-700 could very well give a top-order batter the chance to make 400-plus runs. I would say that this is the record waiting to be broken, especially during these days of four plus runs per over.
E2. Run aggregate in a Test: Graham Gooch's 456 vs India, 1990
Mulder could have gone on to get his 400 runs. And for that matter, even 500. He could well have reached this mark before the end of the second day's play, and declared. That would have meant that he could have broken both Lara's and Gooch's records within a few hours. Alternately, it could have been a Shubman Gill-type situation. At Edgbaston, Gill could easily have scored 200 in the second innings and Gooch would have moved down to second place in the match-aggregate stakes. So, I will venture to say that this record, like the 400, is waiting to be broken.
E3. High innings scoring rate: South Africa 340 for 3 dec in 50 overs (at 6.80 RpO) vs Zimbabwe, 2005
A minimum of 50 overs has been mandated for the scoring-rate records. Any lowering of this bar would dilute the achievement. South Africa's record is significant. Facing a total of 54 in Cape Town, they reached 340 for 3 in 50 overs and declared. Similar situations could develop in future. Any first innings, whether batting first or facing a low-to-middling score, could lead to a scoring rate of seven runs per over. What it requires is a set of attacking batters, a race track masquerading as a pitch, and some ordinary bowling. In Rawalpindi in 2022, England were at 576 for 6 in 83 overs at 6.94 RpO. They could well have declared at that score. Then they would have beaten South Africa's record. So, it is clear that such opportunities would occur in future.
E4. The Test tie - a very special event
There is one other Test event that is worth mentioning: the tied match. It is not a record but a rare happening. There have been only two tied matches in 150 years of Test cricket. What are the chances of a third? I think it is, surprisingly, not that low. Last month, the chances of a tie at The Oval between England and India were quite high. An aggressive shot by Gus Atkinson for four off Mohammed Siraj and a single could very well have resulted in a tie. I would say that there is a reasonable chance that in the next 20 years, the tenth wicket in the fourth innings falls with a run still needed to win.

Career landmarks

Until now, I have covered only the Test-level individual/team records and not the career-level achievements, which are not really records but landmarks that are crossed as players play the required number of Tests. I am just going to touch upon the important career landmarks at their surface levels, without getting into any details.
  • The numbers 99.94 (Don Bradman's batting average), 10.75, (George Lohmann's bowling average), and 1.64, (Trevor Goddard's economy rate) are those 0.01% probability records. The year 2188 will still see these records right at the top.
  • Would any current (or future) player reach 200 Tests, the mark set by Sachin Tendulkar? Maybe the only current player with any realistic chance is Root, who has 42 more Tests to go - that is over four years away. Possible, but quite unlikely.
  • If Root does reach the mark of 200 Tests, he would have crossed 17,000 Test runs, which is way above Tendulkar's aggregate. The interesting fact is that Root could very well go past Tendulkar's mark even if he plays only another 25-30 Tests. So that means the Little Master's record is in some danger of being overtaken. Possibly also his record of 51 Test hundreds.
  • Muthiah Muralidaran has set the bar at the top of Mount Everest. Does anyone have any chance of passing 800 Test wickets? Nathan Lyon would have to play another 60 Tests to reach that mark. That is around six more years, assuming Australia select Lyon in every Test they play, home or away. So, let me say this: Murali's landmark is as safe as if it was housed in Fort Knox. There is very little chance that it will be ever breached. And certainly his 67 five-wicket hauls is housed in a six-inch steel vault inside a strong room in Fort Knox.

The less significant records/landmarks

I have not gone into detail on some of the less-significant records. I expect that most of these will never be overhauled.
Talking Cricket Group
Any reader who wishes to join my general-purpose cricket-ideas-exchange group of this name can email me a request for inclusion, providing their name, place of residence, and what they do.
Email me your comments and I will respond. This email id is to be used only for sending in comments. Please note that readers whose emails are derogatory to the author or any player will be permanently blocked from sending in any feedback in future.

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

Terms of Use  •  Privacy Policy  •  Your US State Privacy Rights  •  Children's Online Privacy Policy  •  Interest - Based Ads  •  Do Not Sell or Share My Personal Information  •  Feedback