Archive | statistics RSS for this section

Do breaks in play take wickets?

Lunch break at Lord's

Lunch break at Lord’s

It seems uncontroversial to state that batsmen are more likely to be dismissed immediately after an interval, than when they have settled back into the new session. But similarly well-worn aphorisms – the nervous nineties and batsmen tending to fall quickly after sharing a sizeable partnership – have shown not to stand up to statistical scrutiny. This post, therefore, attempts to apply numerical analysis to the received wisdom that in Test matches batsmen are more vulnerable immediately after resuming play.

Before introducing the numbers, it’s worth reflecting why this common understanding is so readily accepted by cricket followers. I think there are two mutually reinforcing factors at play, each of which could be supported by associated statistical evidence.

The first factor is that batsmen are at their most vulnerable early in an innings. Owen Benton, in his post ‘When is a batsman ‘in’?‘ demonstrated that the likelihood of a Test opening or middle order batsman falling before scoring another five runs is at its highest when on a score below five. It can be argued that this early-innings fallibility revisits the batsman in the analogical position of re-starting an innings after a break in play.

The second factor is that it is accepted good tactical practice for the fielding side to start the session with its most potent bowlers. While there are no statistics to hand to demonstrate that this tactic is actually applied regularly, nor that those bowlers are more threatening immediately after a break, it would be straightforward to compare the career strike rates of the bowlers opening after the resumption against other bowlers used in that innings.

To test the proposition that wickets fall more frequently after a break in play, I selected a random sample of Test matches played since May 2006 (the date from which scorecards recorded the score at every break in play). Details on the sampling method are provided at the foot of this post.

From the sample of 20 Tests, I noted the incidence of wickets falling in the three overs following (and prior to) 436 breaks in innings, including lunch, tea, drinks breaks, close of play and weather interruptions. Excluded from this figure are any breaks which coincided with the start of a team’s innings.

All results are strike rates expressed as wickets per over. In the period 2006-2015, wickets fell on average in Test cricket at 0.08 per over. As the chart below depicts, there was a 50% increase in the strike rate in the first over after a break in play (0.125). This effect wore off rapidly, so that the second over after the resumption saw a strike rate (0.090) that was barely above the period average and equivalent to the sample average (0.091).

wkts per over-page-001

The result for the 1st over after a break in play is statistically significant. The sample size doesn’t enable the analysis by type of break in play to be anything other than indicative, but is presented below for interest – based upon the first three overs of the restart.

interval types-page-001

Weather breaks appear to be the most damaging to a batsman’s prospects, but the 20 Test sample only featured 11 weather breaks. There does not appear to be any relationship to the duration of the break. For example, the overnight break was associated with a lower strike rate than the brief evening drinks break.

The sample results do seem to bear out the received wisdom that batsmen are vulnerable immediately following a break in play. However, the brevity of the impact – a single over – doesn’t strongly support the two explanations offered above.

If batsmen find a new session is like starting a new innings, then the effect would be visible in the second over, as six deliveries is unlikely to be sufficient for both batsmen to pass this phase.

If the phenomenon is caused by the more potent (and refreshed) bowlers, it too would be discernible in the second over (bowled by the other fresh strike bowler) and third overs of the new session.

There remains an explanation and it’s a prosaic one, which will often be used by commentators seeing a batsman fall soon after a break. It may simply be that the batsman’s concentration has been interrupted and not sufficiently refocused for that first over of the restart. There’s a message here for players – prepare psychologically for the new session – and spectators – don’t dither, get back to your seat for the restart.


Sampling method:

383 Test matches were played in the period from May 2006. Based on an estimated 9,000 breaks in play with an expected strike rate per 18 deliveries of 0.3, 478 breaks in play were needed to give a result with confidence interval of 0.04 at a 95% confidence level. Excel’s random integer function was used to pick numbers between the Test match references of the first (1802) and last (2181) in the sample period. It is worth noting that the random sample was based on Test matches, not breaks in play.

Using the number of relevant breaks in play from the 20 Test sample, a lower total number of breaks of play in the population was calculated for the population of Tests: 8,600. The adjusted sample size was 417, which is lower than the sample on which data was collected.


A select XI: cricket blog posts of 2015


This is the fourth annual Declaration Game selection of cricket blog posts. For the second year, to the basic qualification of being unpaid on-line writing, I have added the criterion that the writer should not have featured in one of the previous annual selections.

Blogging is a disposable activity: posts disappear in the vastness of the web, days or even hours after surfacing. It’s easy to miss what’s new and interesting, so the purpose of this post, as in previous years, is to breathe a little life again into some of the articles I have most enjoyed reading.

I have also applied the criterion of including only a single piece for any one blogger. Given that most of the individuals have produced highly readable, independently minded material all year, my other hope is that I can tempt readers to peruse their wider body of work.

In rough calendar order, here are the Declaration Game Select XI blog posts for 2015.

Published in the first week of the year, Srinath wrote about visiting the Bombay Gymkhana. It is shortly after Philip Hughes’ death and this famous field, hosting multiple cricket matches, makes Srinath anxious:

Mid off from one game would stand next to third man from another, square leg umpires would stand with their unprotected necks, yes, the Hughes region, facing the lashing pull shot that could come their way any time in the day.

Bat on, regardless is a reminder that cricket in many places did carry on as before, even though we might prefer and caution that it should not.

Cricketmanwales was a new and very welcome figure in 2015. Practising a cascading, informal, Gonzo-journalism style of writing, he enthused equally about thrilling top-level cricket and bringing the enjoyment of the game to youngsters in his day job as a cricket development officer. In Which cricket? (April) he celebrated the vitality of the recently completed World Cup:

Though we knew it was coming, this was the moment the dirt was wistfully then swiftly dribbled in over the coffin of yaknow… Richard Hadlee; Ian Botham; the Chappells – cricketing icons that played a patently different game. The gaudy, incremental hikes through T20 Blasts and IPL Extravagorgies seem done; now the World Cup is carnage of a uniquely modern or post-modern sort. It’s official; things have changed.

He finds links between that tournament’s success and his task of energising “the Youff of Today [who] are turned off by stillness and quiet seduction,” before questioning whether the official goal to grow cricket shouldn’t wait until we know ‘which cricket?’

June was the most fertile month. Dr Ayelet Lushkov, classical scholar, was first to show with the most memorable and extended metaphor of the year: an appreciation of Stuart Broad and streaky bacon.

It’s greasy, and crisps up in the pan, and it’s more than a bit American, which is fine, and, more importantly, wholly addictive. Once streaky bacon gets going, there’s no having just one strip, or even one pound. No, streaky bacon takes 7-44 on an afternoon, or 6-24, or scores a 165 at Lord’s. And struts around while doing it.

The month also featured a taxi ride with a West Indian fast bowler – the driver – taken by the Wandering Cricketer. The passenger, an American of South African background, has a project: to record cricket fans from across the world in their colour and their own words. The outcome is a highly distinctive blog, with luscious visuals and floating half-volleys of questions that bring out the best in the people he meets.

In this conversation, the driver says that the US cricket team asked him to play, but he refused because of the money. How much were you offered, asks the Wandering Cricketer:

“You don’t even want to know man…these guys sent me a letter. I still have it at home. They want to give me $400 a match. I make more driving! I like cricket, but I couldn’t go just because of the name, and what name do you make playing cricket in America?”

I hope one day soon that question will feel antiquated.

A more conventional, but no less compelling blog, was the third to yield a gem in June (as it did throughout the year). My Life in Cricket Scorecards is written by Peter Hoare, once of Kent and now living in New Zealand. It’s a receptacle for fond, well-articulated recollections of the incidents and the people Hoare has watched play cricket during his life. The relish with which he wrote about New Zealand’s World Cup campaign shows there’s much more than nostalgia here.

In June, Derek Underwood turned 70. My Life in Cricket Scorecards gave context to his appreciation of Underwood:

For years Playfair persisted in describing him as LM rather than SLA, which was true but missed the point, just as foie gras might be accurately described as meat paste. Underwood took the spinner’s role, to bowl long, constricting spells on good pitches and to attack when the ball turned. No commonplace spinner though, being quicker, Swiss-clock accurate and, at least early in his career, bowling cutters as much as conventional spin.

then adorned the description with a series of stories of Underwood’s specific achievements and their meaning for the author.

It was around that time in the summer of 2015 that Fantasy Bob published his 1,000th blog post. Admirable as that is, you don’t receive Select XI recognition for quantity. Fantasy Bob, Edinburgh club cricketer, sees cricket or its shadow everywhere: at the theatre, on holiday in Italy, in his record collection, in discussion with his wife. In August, the post titled, Grumpy, concerned a spousal conversation. His wife knew many of the cricket-related matters that could upset Fantasy Bob, but not the one that almost drew tears. Brief, profound and really funny.

Another post about club cricket was the next on this chronological list. Dennis Freedman is a digital phenomenon: unremitting; one minute he’s teller of truth to power; the next, utterer of puerile provocations; a moment later, creating an internet meme.

Dennis is normally puncturing the pomposity of international players, coaches, umpires or administrators. October 2015, however, saw him return to club cricket for the first time in 21 years. He was understandably eager to fit in with his new teammates:

The skipper wins the toss and bats. I can’t remember his name, but he is a friendly gentle soul who gives us a pep talk prior to the openers heading out.

I can’t recall what he said.  I was too busy concentrating on whether I should tuck or untuck my shirt. After a quick count, I decide to remain tucked. Eight of us dress properly. I am still cool.

And cool he was after a fine comeback, culminating in debut use of the team hashtag #bloodscricket.

The second outstanding new cricket blog of the year was statistics based and with its title – We need to look at the data – challenged those who, in impugning Peter Moores, dismissed a numerical approach to the sport as shallow. Blogger Owen Benton cut and sliced numbers to inform questions that are usually addressed only anecdotally. But in his October post, investigating how long it takes for a batsman’s average to fairly reflect his ability, he used a computer simulation, rather than actual match data. The method and findings, as in all his posts, are presented with clarity and are accessible to anyone searching for insight beyond the familiar cliches of the sport.

In November, I was drawn to a site with a name unlikely to feature cricket blog posts: Growth Mindset. Its author, Richard Jones is an educationalist and active junior coach. He has applied his professional expertise to an issue that usually goes unacknowledged in the grassroots game:

Junior cricket, particularly at club level is a constant mismatch between players who have clocked up hundreds if not thousands of hours of practice against those who have maybe clocked up less than 20 hours in some cases.

Jones exposes how, instead of acknowledging this truth, we talk about the juniors with the hours behind them as ‘naturals’ and allow those at the outset of their experience to think of themselves as ‘no good at the game’. In Cricket, Falling Junior Participation and the Fixed Mindset, Jones diagnoses the problem, its longer term significance and then ventures a solution, which he plans to implement in 2016 at his club. I am fascinated to hear how his bold plan fares.

Many English cricket bloggers spent 2014 at daggers with the cricket establishment. Changes to the ECB regime reduced, by a notch or two, the intensity of that antagonism in 2015, although it was the focus of a lot of strong writing. Perhaps the best of the lot came in December, from Tregaskis: Caesar’s Wife and Sports Journalism: When is Close too Close?

In a wide-ranging piece, showing a mastery of sources beyond the reach of most professional journalists, Tregaskis takes to task the ethics of cricket writers and produces a measured, subtle polemic.

The burden of proof is shifting. Journalists now need to earn trust through their record of impartiality. Drinks, dinner and games of golf with sources may be trifles but they do not shift the burden of proof. ‘We’re just doing our job’ is a mantra that offers old solutions to newer, more complex challenges. We need assurances that working relationships are just that. We need assurances that the journalist is not always the last one to see when the tipping point is reached.

The final place in this Select XI goes to the piece of writing that I have thought about most often this year. And in the fine tradition of cricket selection, it’s a ringer: read in 2015, but written in 2009.

Deep, backward, and square no longer publishes, but did leave us, almost accidentally, with Going downhill quickly. It’s a Test cricket stats piece about wagging tails that discovered something unexpected:

The lesson is clear: what happens in the first half of an innings tells us nothing about what we can expect in the second half. For example, on average throughout test history, whenever the fifth wicket has fallen at a score between 50 and 99, the remaining batsmen have added a further 95.7; whenever the first five partnerships have realised between 400 and 449, the last five wickets have typically amassed… 95.3.

It’s not just the fifth wicket: “At any stage of a test innings, what has happened up until the fall of a given wicket is a useless predictor of what’s going to happen afterwards.”

This is not merely a statistical quirk. It strikes at the way we talk about and analyse the game, which so often employs projection forward from what we’ve observed so far: the wicket’s (not) playing well; the bowlers/batsmen are on top, etc. But it’s not just us, on the sidelines, who think like this. The players do as well. All of us expecting to see more of the same, when, beneath our eyes, the numbers show a different reality.


I am grateful to all these bloggers, as well as the Blognoscenti listed to the right (or below, if reading on a mobile device) for the pleasure provided by their work this year. I would really welcome readers’ contributions by proposing your own favourite blog posts of 2015.

The Declaration Game selections for previous years can be found here: 2012, 2013, 2014.




The curse of Cook

Embed from Getty Images

Six opening partners tried and rejected in three years – a turnover of one opening batsman per Test match season. The inability to find a player to rise to the challenge of opening the innings alongside Alastair Cook is the most prominent of the selection problems besetting England. This post does not venture a solution (although I have provided a mathematical response), it looks at the impact on those six selected, then rejected batsmen.

The impact of playing with Cook and then being dropped is assessed in a narrow, statistical fashion. The first graph shows the ten innings batting average in first class matches before and after each player’s brief career as a Test opener. (NB Adam Lyth’s post-Cook average is based on the six first-class innings he has played to date)

Cook s curse 1-page-001

Across the group, there is a reduction in batting average of 40 runs per completed innings (56%). Joe Root has the sharpest reduction. He and Trott are the only members of the group who played Test cricket before opening with Cook; and Root is the only member of the group who played Test cricket after opening with Cook.

Three of the players (Compton, Root, Robson) may have harboured hopes that their Test opening careers would continue when they returned to first class cricket. Compton, for example, played four innings (including a century and a fifty) before his supplanting by Joe Root was made clear by the selection of an England side for a warm-up match. Selecting ten innings from his return to the Somerset side or from his official relegation from the England side makes little difference to this ten innings average (48.2 v 47.6).

We should not be surprised that players’ first class averages drop after a tough period as rookie Test match openers. They had been picked as form players – all six had short-term averages exceeding their career average when brought into the team – and their strong form had been interrupted by the stiffer challenge of Test cricket. In Trott’s case, his return to first-class cricket involved more than just re-finding form with the bat, but psychological health, too. The fall away in their performance, however, is noteworthy for its abruptness and consistency across the group.

To test whether it is a short-term effect, I have also compared their batting average for the last full season of first-class cricket before their selection as Cook’s partner and the first full season of first-class cricket following their demotion from the captain’s sidekick. In all cases except Root’s, the seasons assessed were England county seasons.

Cook s curse 2-page-001

In this analysis the average fall in batting performance is less severe and is less consistent across the group. Root, the only player to remain in the Test team, maintained his pre-selection season average and Carberry’s varied downwards by fewer than five runs per completed innings.

All of the six players struggled for most of the innings they opened alongside their captain in Tests matches. Once out of the team (or in Root’s case, batting lower in the order), they were unable to regain their earlier productivity.

Alex Hales is strongly favoured to be Cook’s next opening partner. His current 10 innings first-class batting average is 36.0 – lower than all of his predecessors (although Hales may have further innings in the County Championship and in the UAE to improve on this before the Tests against Pakistan). Hales will, of course, be aiming to repel the curse of Cook that leaves batsmen under-performing when dropping back into county cricket. The surest way of doing this is by scoring so many runs for England that he stays in the team, opening alongside the captain.

Quick single: Dr Seamus Hogan

Yesterday, I had intended to write up a piece about Steve Smith, Derek Randall and John Arlott. I was going to point out the similarities of Smith and Randall: fidgeting, stepping across the crease, playing shots without establishing the orthodox ‘solid base’; I would revisit Arlott’s famous article about the Nottinghamshire batsman and why his often replayed summation was unfortunate hyperbole. I was going to conclude that Smith didn’t remind me of my inevitable demise.

But yesterday morning I read the sad news of Seamus Hogan’s death. I knew Hogan in a very modern way: twitter exchanges, emails and comments on each others’ blog posts. Different hemispheres, shared interests, instant, brief communication.

Despite the limitations of that sort of acquaintance, he came across as very likeable, generous and clever. In fact, Hogan was a leading economist and academic in New Zealand. From time to time, he applied his academic rigour to cricket questions, generating counter-intuitive, but statistically solid insights.

Across the cricket world, he is probably best known for his work behind WASP, the result and score predictor used by, amongst other outlets, SkySports. Like any forecasting system, WASP was only really interesting for most viewers when it was wrong. On a couple of occasions, I followed Hogan as he patiently and politely responded to gloating and ill-informed criticism on twitter of the system he co-founded. Too patient and too polite, I thought. Those who really knew him may know if that was his general disposition – I suspect it was.

Hogan’s blogs – about cricket and other matters – can be found on the Offsetting Behaviour website, which he shared with a colleague, Peter Frampton, who wrote this touching, dignified notification of his death.

For all John Arlott’s erudition and eloquence, he was dreadfully wrong about Derek Randall and, indeed, any other batsman “who overplays his hand and falls into disaster”. That’s not what reminds you of your own mortality.

My thoughts are with Seamus Hogan’s colleagues, his students, his friends and, above all, his family.

Quick single: non-overlapping ODI twin hundreds

In their run-making at Edgbaston against New Zealand today (9 June 2015), Joe Root and Jos Buttler shared a unique feat that involves the two players not sharing the crease. Root came to the wicket in the first over with England yet to score. He was out in the 25th over with the score at 180. Buttler replaced him at the wicket and batted until the 48th over.

Root and Buttler are the only example in ODI history of twin century makers in the same innings who did not bat together.

A very similar feat was achieved earlier this year by Rilee Rossouw and AB de Villiers. The difference is that South Africa’s innings featured triplet, not twin hundreds. On 18 January 2015 at the Wanderers, Amla and Rossouw shared an opening partnership of 247 against the West Indies. After Rossouw fell, de Villiers joined Amla, putting on 192. Rossouw and de Villiers both scored hundreds, but didn’t bat together.

The list of twin century makers in ODI innings for England is reproduced below, with the partnership the batsmen shared.

Gooch (117*) and Gower (102) v Australia at Lord’s, 3 June 1985, 202 (2nd wicket)

Trescothick (109) and Hussain (115) v India at Lord’s, 13 July 2002, 185 (2nd wicket)

Trescothick (114*) and Solanki (106) v South Africa at the Oval, 28 June 2003, 200 (1st wicket)

Strauss (100) and Flintoff (123) v West Indies at Lord’s, 6 July 2004, 226 (4th wicket)

Strauss (152) and Collingwood (112*) v Bangladesh at Trent Bridge, 21 June 2005, 210 (4th wicket)

Cook (102) and Bell (126*) v India at Southampton, 21 August 2007, 178 (2nd wicket)

Strauss (154) and Trott (110) v Bangladesh at Edgbaston, 12 July 2010, 250 (2nd wicket)

Morgan (124*) and Bopara (101*) v Ireland at Malahide, 3 September 2013, 226* (5th wicket)

Quick single: 355 not out

Kevin Pietersen’s achievements and experiences of the last two days have, understandably, been the subject of a lot of passionate writing. I want to record, dispassionately, some facts about his unbeaten innings against Leicestershire at the Oval.

The research to identify comparable (possibly, superior) innings to Pietersen’s was carried out by Allan Draycott, who tweets about history, politics and cricket as @allanholloway.

Pietersen scored at almost twice the rate of the other Surrey batsmen: 355 runs from 396 deliveries faced. The ten other Brown Caps contributed 185 runs from 395 balls. He was responsible for 70% of the runs scored while he was at the crease.

Pietersen outscored the second highest score of the innings – 36 by Sangakkara – by 319 runs. The highest differential between top and second top score in an innings was when Hanif Mohammed set the record for first class individual score for Karachi v Bahawalpur in January 1959. Hanif’s 499 was 396 runs higher than the next highest score in that innings.

Other notable differentials between top and second top scores in an innings are:

  • Brian Lara’s innings that broke Hanif Mohammed’s record. Lara’s 501* for Warwickshire v Durham in June 1994 was 385 clear of the next highest score – the record for the County Championship.
  • Bert Sutcliffe scored 385 for Otago v Canterbury in December 1952, which was 356 runs more than the second top score in that innings
  • Graeme Hick’s 405 for Worcestershire v Somerset in May 1988 was 349 runs greater than second place.

Sutcliffe, like Pietersen, scored a triple-century in an innings when no other score above 50 was recorded. Allan Draycott identified the following other precedents:

And, of course, Pietersen batted with the prospect of attending a meeting at the end of the second day’s play that would determine his fate as an international cricketer. It’s not the sort of fact that gets recorded on a scorecard, but it will always be associated with 355 not out.

Quick single: the fourth innings 400 barrier

A lead of 400, entering the fourth innings of a match, is an impregnable position – virtually. A score that is routinely accrued in the early part of a match, becomes unattainable (almost) for the side batting last.

The trend of diminishing run-scoring returns as matches enter their final phase is one of the known, understood and acted upon patterns of cricket. There’s no mystery to it, not if you get close to the action and see the deterioration in the pitch that makes it less and less reliable for batting on as three or four days of play pass.

According to this list, Middlesex’s score of 405-5 against Somerset today is in the top 70 winning fourth innings totals in first class cricket history. This list spans 120 years, meaning that successful run chases of this size, or greater, occur slightly more than every other year on average. In fact, 13 came from the last decade, showing that their frequency is increasing, probably because the condition of pitches alters less as matches progress making run-scoring less hazardous deeper into a match and faster scoring rates mean that leads of 400 plus are achieved at an earlier point in games.

Nonetheless, they remain rare and noteworthy achievements (albeit Middlesex’s second in two years after scoring 472 against Yorkshire in 2014). And each time they happen, their effect is to make another occurrence both more and less likely. Teams facing a target of more than 400 have recent examples to emulate and may opt for the chase rather than survival. But teams setting targets, who can time when the fourth innings begins, will be a little more cautious, asking more in less time of their opponents. For a team on top after three innings, this maxim is particularly true: losing feels worse than winning feels good.

Using up your overs

50 overs-page-001Wickets tumble in the first innings of a one day international. Commentators will urge the batting team to do it. Failure to do it will be condemned as a cardinal sin. Using up your overs; batting the full 50 is the very least expected of a side batting first.

It’s a viewpoint that is well supported by the historical records of sides that have been bowled out without using their full allotment. In recent years, between one-quarter and one-third of all matches have featured the side batting first being dismissed. In the period shown in the chart below, only 20% of those sides batting first and losing all their wickets have won the game.

bowled out-page-001

There’s a strong association between the number of balls forfeited and chance of defeat. There’s no magic in that relationship. The shorter the innings, on average, the lower the score.

The importance of batting out the overs is a viewpoint with a critical consensus backed up by some solid data. It was almost shocking to hear a voice of dissent. It came on Test Match Special a few years ago. The batting team were struggling and the commentator made the usual injunction that the lower order see out the overs.

“Why?” asked Geoff Lawson, who went on to rationalise that if all the batting side attempted was to survive the 50 overs, they were very unlikely to set a winning total. Wouldn’t it be better, Lawson argued, to hit out with the aim of setting a challenging target, accepting the risk that they could be bowled out, than to crawl to an unsatisfactory total?

Lawson was positing the batting team, while in adversity, having a tactical choice to make. They had to decide how to balance risk and reward. It felt distinctively Australian to stress there was a route other than that of damage limitation; one that could very well leave the team scoring fewer runs than if they took the conventional approach of using up their overs.

It is also a notion that can be explored statistically. The chart below shows the win % of teams making below par scores (in 5 run intervals) in the first innings of ODIs since 1996 when the 50 over per side format became standard across the world. Matches where weather or playing conditions reduced the first innings to fewer than 50 overs have been excluded.

1st inns win-page-001


Between 145 and 170, there is a low but steadily increasing probability of winning. From 175 to 199, however, the win percentage levels out, before jumping from 20% to 35% when 200 is reached. Thereafter, it’s not until 230-234 that there’s another increase in win percentage.

The nudge & nurdle method can be measured against Lawson’s long-handle approach by selecting a scenario and applying some probabilities. I have chosen a team on 140-7 after 40 overs.

The approach of maximising the length of the innings can be expected to yield a total in the range of 180-184; although there is, say, a 20% possibility that they will be bowled out, despite their best efforts, for 160-164.

Their chance of winning (based on past results) would be:

(0.8 x 0.21) + (0.2 x 0.13) = 0.194

The long-handle method could take them to a healthier 210-214, accepting that about half the time they would swing and miss and fold for 150-154. Using past results, this produces a win likelihood of:

(0.5 x 0.07) + (0.5 x 0.37) = 0.22

In this particular case, with these probabilities, it’s better to attack, than see out the overs. Other situations, perhaps where, say 190, is the maximum that an aggressive approach could bring may repay the conventional conservation of wickets and seeing out the overs.

Individual match situations – the respective strengths of teams, expected change in playing conditions, the batsmen at the crease, perceptions of par scores for the venue, etc – will all trump the message of this historical analysis. What I hope it does show, however, is that a blinkered trudge towards the final ball of the fiftieth over will not always serve the batting team best.

Ian Bell: Breaking records by stealth

When Ian Bell became England’s all time leading run scorer in ODIs, while scoring a century against Australia in Hobart, I imagined cricket followers performing a sport-wide double-take. “He’s what? Bell? Is that right?” before making a mental note to themselves to check on statsguru when they got home and found a quiet moment with a computer.

The only Bell innings in an ODI that I have any sort of memory of (barring those in the current series) was at Southampton and it involved lots of lofted drives. I don’t follow ODI cricket with the forensic attention I pay to Test matches, but I would expect England’s leading scorer in the format to have made a stronger imprint on my memory. How then did Bell come to break this record and, other than the quantum of runs, how does his ODI record compare?

Bell’s first ODI was played in Harare in November 2004 – three months after his Test debut. He opened the batting with Vikram Solanki and scored 75 (115 balls) in a successful chase of 196. His most recent match, at Perth, was his 155th. Amongst batsmen, Bell is the third most capped English player – behind Paul Collingwood (197) and Alec Stewart (170). Quantity of cricket is clearly a large part of the answer to the question ‘how did Bell break this record?’ But my instinct is that Bell has not been an ODI regular over the last ten years.

Since his debut, Bell has played fewer than two-thirds of the 234 ODIs played by England. He has had two lengthy periods out of the team – Feb-Dec 2005 (16 matches) and Nov 2008 – July 2010 (33 matches) – as well as numerous ins and outs typical of a fringe player or of a Test certainty being rested between five day series. His longest continuous run of appearances is 35, from July 2007 to Nov 2008.

Bell’s runs have come more prolifically at home, when opening, and in the match of his recall (or call up) to the team. Bell circs-page-001


Bell’s first ODI century came in his 48th appearance. There have only been three more, but Bell has recorded the most scores of 50+ in ODIs for England: 36.

Two years ago I wrote about Bell, the Test batsman.

I have reconciled myself to Bell as a very good international batsman… he has reached the plateau of his level of accomplishment… I don’t expect him to  dominate a major series or change the flow of too many contests… Bell is really very good and that is good enough.

Since then Bell distinguished himself as the outstanding batsman of the 2013 Ashes, recording three tons. It was a peak above the plateau, to which he seems to have returned. And as a Test batsman, Bell has always had one or other of Strauss, Cook or Pietersen as his senior. He is now the senior ODI batsman, yet I stand by my appreciation of his contribution from that earlier piece. He has rarely been dominant in ODIs. Of the top 30 ODI series run aggregates by England batsmen, Bell appears once, in eighth place – scoring 422 runs in a seven match series against India in 2007.

It might be helpful to place Bell amongst his peers. Of the 22 England batsmen with over 2000 ODI runs, Bell has the seventh highest batting average and eighth swiftest scoring rate. The players whose record Bell’s most closely resembles are Allan Lamb and Paul Collingwood, two of England’s most respected short-form batsmen.

Bell and Eng bats-page-001

In the ten year span of Bell’s international ODI career, 28 other players have scored 4,000 runs or more. Bell is in 15th place and has the third highest aggregate of those averaging under 40. Only two batsmen in this group have scored fewer hundreds than Bell.

The scatter diagram of batting averages and scoring rates shows Bell is in the lower half of the range for both measures. Graeme Smith and Mahela Jayawardene are the batsmen closest to Bell on the chart.

Bell and all bats-page-001


A number of factors have helped propel Bell to this record. He has stayed fit, physically and mentally, over ten grueling years of international cricket. He has maintained good relations with the England team management and, as pointed out to me by @ballsrightareas, avoided the sometimes career-shortening office of captain.

The gap at the top of the England batting order created by Marcus Trescothick’s exit from international cricket in 2006 has given Bell more opportunities than he would have had. Kevin Pietersen’s and Jonathon Trott’s absences have also created space for Bell. The curtailing of Trescothick’s and Pietersen’s careers prevented those two players, more suited than Bell to short-form cricket, setting a more stretching total runs record for England.

In this period, England’s selection policies have not been consistent. Bell may have suffered some omissions because of this lack of clarity about what the best team is. I suspect that is balanced by some of his recalls being down to the same inconstancy of selection.

On reflection, I don’t feel ignorant to have been taken by surprise by Bell’s recent achievement. He’s played a lot of matches, but many fewer than a leading exponent of this form of cricket would have done. He has a good ODI record, certainly by England standards, but not a great one. I will be surprised again if Bell’s future performances force me to alter that view.

Survey results: the vital statistics of cricket blogging

blog types-page-001

This second post in the series on the blogger survey results, draws out some of the significant numbers that give shape to the pursuit of cricket blogging in 2014. It follows an earlier post about the bloggers and their motivations.

In wanting to give numerical substance to that shape – longevity, frequency of posting and readership – I realised that the survey was not providing me with an essential part of the picture. I needed to know the blogs’ subject-matter. Lumping together metrics for a blog dedicated to nineteenth century minor county cricket with a multi-handed sports news blog would generate little insight. So, to make good this omission, I have quite crudely attempted to categorise blogs by what they are about.

I applied the following categories:

  • topical – for blogs where the major stories of the day, globally or nationally determine what’s written
  • statistical
  • specialist – sites dedicated to a niche of the cricket world, historical, geographical, functional, etc.
  • visual – where pictures, not words are the key content

and finally, in a category that assembles blogs of many flavours

  • essays – blogs where posts range across subjects, perhaps related to topical issues, but driven by the writer’s own experience as player, spectator, viewer or reader.

Some blogs are truly versatile. In those cases, I have applied a judgement about which category best fits.

Subject matter

The incidence by subject matter of the 87 blogs that I was able to categorise is charted here. Topical blogs are the second most numerous but account for less than one-third of all sites. Essay blogs are the most common. Many bloggers have found a specialism, away from cricket’s mainstream. These two categories may bear out the finding of the previous post that many bloggers are motivated by a belief that they have something distinctive to write about.

blog subjects-page-001



Blog(ger) duration

98% of respondents to the survey have (or contribute to) live blogs. The range of time bloggers have been active, depicted below, is suggestive of a couple of trends:

1) one-third of the bloggers have been active for over five years, showing that there is a strong, durable core to the sector.

2) ten bloggers have published for less than one year, which appears small compared to the numbers that have existed for 1-4 years. If representative of this field, it could be an indication of a slowing to the rate of blogs being started, or a choice made by new writers to join existing on-line platforms.

Length of time-page-001


Blog type

The single-handed blog is the most common approach taken to publication, involving 83% of respondents. But another strong feature is light footedness, with over one-third of respondents having pieces published on two or more site types. The figures beside the arrows in the diagram below show the number of bloggers involved in both types of blogs. Included in those figures are eleven bloggers who are active in three of the blog types and one who publishes on the full set.

blog types-page-001

There does not appear to be any relationship between length of time blogging and type of site the blogger contributes to. For example, writing for professionally published websites is as much a feature of very new bloggers’ writing as it is more established bloggers. Essay and specialist blogs are very likely to be single-handed, with collaboration more often a feature of topical blogs.

Blogging frequency

The most common and the median frequency of publication is ‘more often than monthly but less than weekly’. Over one-quarter of blogs are ticking over, with new material published less often than monthly.

blog frequency-page-001

Looking at the features of those blogs publishing most often (more than weekly and daily), the following is apparent:

  • blogging duration: more likely to be blogging for 1-2 years; less likely to be in the first year of blogging. Otherwise, there is no strong association with blog age.
  • blog type: more likely to blog on multiple sites and to contribute to professionally run websites
  • blog content: surprisingly, topical blogs are not much more likely to be publishing at the more frequent end of the scale than the average for all blogs.

Those blogs ‘ticking over’ (i.e. publication less often than monthly) are associated with:

  • blogging duration: less likely to be those newest to blogging (two years or less)
  • blog type: less likely to contribute to professionally run websites
  • blog content: more likely to be essay style blogs.

Publicising blogs

Twitter is the dominant platform or method for publicising blog posts and is considered the most effective way of drawing an audience. Facebook is second for usage and effectiveness. The other digital media are all out-performed by direct contact with readers.

blog publicity-page-001

Looking at the combinations of publicity methods used by bloggers, almost publicity methods-page-001two-thirds use one or two (e.g. twitter and facebook). Nearly one in ten bloggers employ five or more methods to help get their writing known.

The survey also asked about the number of social media posts made per blog post. Three-quarters of respondents make four or fewer social media posts. Only 5% of respondents report plugging new material ten or more times on social media.


In our digital age, asking how many views your blog gets daily, is probably more intrusive than enquiring about the blogger’s age and not far from asking their salary. Nevertheless, 84 respondents supplied figures.

There is a broad base of low activity blogs (under 25 views per day), with an even distribution of sites in the low-middle readership range. Six sites sit clear of the pack, attracting over 1,000 views from readers per day.

blog views per day-page-001

The profile of blogs at the top end of readership (over 250 per day), compared to the rest, includes frequent publishing of new material (more than weekly); being longer established (five years or more); featuring topical material; and collaborative exercises (although single-handed sites remain the most numerous).

To gauge how on-line traffic is shared between blogs of different sizes, I have made some assumptions about actual daily views from the ranges offered in the survey. I have taken the mid-point of each range as the actual figure, except for the <25 (assumed to be 24) and >1,000 (assumed to be 1,000) ranges. Summing up the figures, there are about 15,000 views of these blogs daily. At the very least, 40% are visits to the six biggest blogs.

blog readership-page-001

If the assumption about the average actual readership of the six biggest blogs is raised to 2,000 per day, the % of views those blogs account for rises to nearly 60%; at 5,000 it exceeds 75%. No more than 5% are visits to the smallest 33 blogs.

By combining this data with my categorisation of blog types, I have been able to derive estimates for how many views each type of cricket blog gets daily. According to this calculation, topical blogs get just over one-half of all traffic, generated from one-third of blog sites. A similar number of essay style blogs attract one-quarter of daily views. The traffic to the other categories is in line with the number of blogs of that type.

Blog types & traffic-page-001

We know from the first post in this series, that many bloggers write for the pleasure of writing. Readership figures are not their priority. Nevertheless, I have looked at one other traffic statistic: views per post (derived from questions about daily views and frequency of posting new material). At the low activity end of a wide spectrum of readership, bloggers posting frequently (more than weekly) and getting fewer than 25 daily views are probably getting 50 views per post. To many familiar with the scale and reach of the Internet, that sounds paltry. I prefer a different perspective: prior to the Internet, how easy would it have been to find 50 people to read something on cricket that you, an unpublished writer, have penned? Pretty tough, I think.

This post has attempted to let the numbers tell the story of cricket blogging in 2014. In the next article, I will present bloggers’ own views of the state of cricket blogging, as well as their future aims.


Note: post updated to include analysis and chart of blogs and estimated daily views by blog type – 7 November 2014.