After a poor winter, Moeen Ali has been dropped from the England team. Ben Jones looks at the data to explain what went wrong for the Beard That’s Feared.
CricViz analysis of the second Test between India and England in Vizag.
Wide turn stymies spinners
This was a difficult pitch to read with ball-tracking data indicating that it took significant turn from the second day onwards and by the third day was taking more turn on average, 5.14 degrees, than the pitch in Dhaka did at the same stage, 5.09 degrees. The graph below shows the rate at which the pitch turned as the match progressed. However despite this significant deviation only 21 wickets fell in the first 380.1 overs of the match and only when eight fell in the last 69.2 overs did the scorecard begin to reflect the amount of turn on offer.
The critical difference between this pitch and the one in Dhaka was that the extra grass on this pitch held it together far better and for longer meaning in Rajkot the sharpest turning deliveries predictably pitched in worn areas of the pitch, whereas in Dhaka balls spun big unpredictably from previously compact areas of the pitch that were broken up by the impact of the ball.
The nature of this pitch meant that the sharpest turning deliveries pitched well outside the line of the stumps, nearer the bowler’s foot holes, as illustrated by the pitch map above. Naturally more of a threat is posed if balls turn big from within the line of the stumps. 41% of the 58 deliveries that turned more than eight degrees but did pitch within the line of the stumps were bowled in the fourth innings when the pitch was most worn.
England’s spinners improve; India’s get worse
Speaking after the match England’s coach Trevor Bayliss suggested that their spinners had improved their control of length. Ball-tracking data shows this not to be the case with England’s length percentages remaining almost exactly the same as in the Bangladesh series. What they did improve however was their line, illustrated by the pitch map below: they maintained tighter groupings and conceded runs at 3.36 runs per over compared to 3.63 against Bangladesh as a result.
India’s spinners meanwhile bowled with less control than against New Zealand, as illustrated by the pitch map below. This was the flattest of the four pitches India have played on this season and England’s first innings was the longest they have been in the field in a home Test since they played England in Kolkata in 2012. In these less helpful conditions Ravichandran Ashwin and Ravindra Jadeja struggled to maintain the exceptional groupings they managed against New Zealand.
England’s best spinner in this match was Adil Rashid who took match figures of 7 for 178. Rashid displayed significant improvement in his control of line and length from the Bangladesh series, illustrated by the pitch map below.
In this Test Rashid landed 60% of his deliveries in a two metre range between four and five metres from the batman’s stumps, in the Bangladesh series that figure was just 46%. The principal improvement came in bowling fuller: against Bangladesh he dropped 14% of deliveries shorter than six metres from the batsman’s stumps, in Rajkot that figure fell to just 9%.
England commit forward and back
It is perhaps too soon to pass judgement on England’s batsmen against spin given that this pitch did not break up and turn as both pitches in Bangladesh did and as they are expected to do more in the rest of this series. However, England’s four centurions, Joe Root, Moeen Ali and Ben Stokes and Alastair Cook, as well as debutant Haseeb Hameed showed really encouraging signs with their footwork against spin. None of Root, Moeen or Stokes played a single shot with footwork categorised as “no movement” in their hundreds suggesting that they committed clearly to going forward or back, which is critical to playing spin well, while Cook played just 28 out of 290 balls in the match as such and Hameed just 12 out of 259.
England escaped defeat in Chittagong thanks largely to a superb all-round performance from Ben Stokes and Bangladesh’s first innings batting collapse; they were not so lucky in Dhaka where their shortcomings playing and bowling spin were exposed again and they crashed to a heavy defeat. There is no shame in losing to this Bangladesh team but there is shame at the manner of the result, in which their batting, bowling and fielding were alarmingly substandard.
It is difficult to ascertain which area of their spin-game, playing it or bowling it is a greater problem and quite frankly it is facile to apportion blame to one or the other; both were poor and both must be improved dramatically if they are to avoid a thrashing at the hands of India.
As bad as England’s batting was though, chasing 273 in the fourth innings was always going to be a very difficult task on a turning pitch against an excellent spin attack, and the size of that run-chase can be traced back to bowling and fielding errors throughout the Test.
After the match Alastair Cook was forthright in admitting that “we didn’t bowl great. And yes, their spinners did out-bowl our spinners. We’re not hiding behind the fact that we haven’t got world-class spinners.”
If the spinners England do have cannot exploit helpful conditions at their disposal then they are always going to struggle to win matches on the subcontinent because they will more often than not be chasing too many runs.
In the first Test England’s spin problem was primarily their inaccuracy and in the second Test the same can be largely said again. England’s inability to land the ball in roughly the same area consistently contributed to Bangladesh’s spinners bowling 15 maidens in the match compared to England’s eight and 50 in the series to England’s 21.
This failure to build pressure prevents spinners finding a rhythm against batsman and lining them up. Balls that turn fractionally more or less than preceding deliveries are far more dangerous if they are bowled at the same batsman rather than a different batsman because the same batsman is more likely to be influenced by the ball before and play down the wrong line.
Cook said that he would be interested to see the stats comparing England’s spin lengths to Bangladesh’s, suggesting that their bowlers maintained a better length. In fact both sets of spinners bowled very similar lengths overall [see below] with Bangladesh over-pitching slightly less, but both teams landing about 60% of deliveries in a two metre range between 3 and 5 metres – categorised as a ‘good’ length for spin bowlers.
England’s lengths, were almost identical to the first Test [see below] while Bangladesh’s actually got slightly worse. The home team were able to maintain more control because they bowled far tighter lines [see above]. Bangladesh’s pitching line groupings are tighter and straighter than England’s which forces the batsman to play and increases the chances of getting a wicket bowled or lbw.
Moeen was the most accurate of England’s three spinners [see below] and improved the percentage of deliveries bowled in the 3-5 metre range from Chittagong from 58% to 61% by cutting down on over-pitched and short deliveries. He also improved his line considerably – recording a far higher percentage of deliveries that would have gone onto hit the stumps. Rashid’s lengths actually got worse from Chittagong, most notably pitching 13% of his deliveries six metres or shorter. Ansari, meanwhile, recorded better lengths than Gareth Batty did but struggled to maintain his line.
Speaking after the match Cook explained the problem inaccurate bowling poses for setting fields. “You always feel you are a fielder short,” he said. “If you are leaking four, five runs an over in a low scoring game you have to put your boundary-riders out. It would be great if you could attack but you have got to hold your line and length better.”
England struggled to play spin just as much as they struggled to bowl it. Their dramatic collapse from 100 for 0 to 164 all out in a single session was the vertex of a problem that had been made all too apparent in the three preceding innings in the series in which they were 106 for 5, 62 for 5 and 114 for 6.
As has already been mentioned, batting in these conditions is not easy and Bangladesh have a very good spin attack, however to lose ten wickets in a single session is indicative of a more deep-rooted problem.
In terms of personnel, five of England’s top seven: Cook, Joe Root, Moeen Ali, Ben Stokes and Jonny Bairstow, are proven performers at this level and are guaranteed selection, while Ben Duckett’s second innings fifty showed enormous promise. Only Gary Ballance, who could well be dropped for the first Test in India, can be said to be out of his depth. In this sense England’s problem is not who, it is how – it is technique and strategy.
What has characterised the success of Bangladesh’s spinners, like that of Ravi Ashwin, Ravi Jadeja and Rangana Herath too in recent years, has been the lack of ‘mystery’ in their bowling. Rather than doosras and carom balls posing the threat instead it has been orthodox spin bowling. Alongside their accuracy these bowlers’ chief weapon is very slight differences in deviation and the key variation is the one that goes straight on. What makes this particular variation so deadly is that more often than not it is a natural variation, meaning it cannot be consistently read from the hand.
Eight of the fourteen dismissals of England’s top seven batsmen were to deliveries that deviated less than the average for any of the frontline spin bowlers [see below]. Playing against natural variations such as these is understandably difficult because there are no visual cues on which to predicate decision making other than the trajectory of spin after pitching, by which point the reaction time is negligible unless the batsman has gone back deep in his crease and therefore has some time to adjust, or, the spin can be nullified if the batsman is well forward and has smothered the turn. This reemphasises the importance of clearly committing either forward or back.
Given the average height and stride length of a batsman, deciding whether to commit well forward or well back is particularly difficult to balls pitched between 3 and 5 metres from the batsman’s stumps, within which 60% of all Bangladesh’s spinners deliveries landed, and also, as the match progresses and bounce becomes less predictable, balls pitched between 5 and 6 metres from the stumps, within which a further 21% of the spinners’ deliveries landed.
All the deliveries to dismiss England’s top seven batsmen excluding Ballance’s second innings leading edge pitched within this three metre range. This illustrates how difficult it must have been for England to commit either forward or back to those balls that they got out to because they were landing within a length range in which making the appropriate footwork is extremely difficult and even then once having committed either back or forward they then have to play the ball successfully. The less committed they are to going forward or back the less spin they will have smothered or the less time they will have to adjust.
The foundational aspect to succeeding against spin bowling therefore is reading the length well, something that England are going to have to work on ahead of the India series where the opposition is even stiffer.
It’s hard to know right now if England are truly on the cusp of surfing a wave in one-day cricket, but there has certainly been something of a revival since their pitiful displays in the last World Cup 18 months ago.
In all, they have won 13 one-day internationals since then and lost nine. They have won series against New Zealand, Pakistan and Sri Lanka, and lost narrowly to Australia and South Africa.
The evidence suggests the batting is in excellent shape at the moment. In that time, England have posted grand totals of 408, 399, 365, 355 and 350 – five of their seven best totals ever. Even allowing for some flat wickets and the modern trend for a par score to be so much higher than in previous times, this is a remarkable feat. However the bowling has not always been so good.
Against Pakistan, in the five-match series that begins on Wednesday, England are likely to start as fairly strong favourites thanks to their batting might, but this would be a good time for them to have a strong series with the ball.
In South Africa in February, England did not generally excel in the bowling stakes, though they led 2-0 with three to play and at Centurion set a faltering side, who had just lost the Test series, 319 to win.
Let’s examine what happened from that point. David Willey and Reece Topley, both currently injured, failed to strike with the new ball, allowing Quinton de Kock and Hashim Amla to get away with a fast start. But it was what ensued subsequently that really hurt England. Ben Stokes, Chris Jordan and Moeen Ali – all in the squad for the upcoming series – failed to stem the flow or take wickets in the overs that followed. Only Adil Rashid, who returned an exemplary 1-45 from his 10 overs, emerged with credit.
The CricViz database records that Stokes, in returning 0-54 from 8.2 overs, attempted just three slower balls and found no discernible swing or seam. The lack of movement was not his fault – this was a flat wicket and he did not have use of the new ball – but should he have been trying more slower balls? Chris Jordan bowled seven slower balls in his 1-54 from seven overs. South Africa scored just two runs from Jordan’s slower deliveries suggesting it was a tactic worth pursuing, but Jordan bowled only one per over, Stokes less than one every two overs.
As for the spinners, whereas Adil Rashid created a degree of uncertainty by bowling seven googlies in his 10 overs, every single delivery sent down by Moeen Ali was an off-break. Little wonder, perhaps, that he went for 30 more runs than Rashid.
The stats show also that Moeen had a bad shift at the office with his lengths that evening: six short-pitched balls and three full tosses, one at beamer height. As for the seamers, Stokes only got one attempted yorker on the spot, three others ending up as half volleys, two as full tosses. Jordan got two in the spot, but bowled three full tosses and seven half-volleys in all. When Jordan dropped short he was punished, with 27 runs coming off 16 balls.
As for the bowlers’ lines, Jordan bowled the vast majority of his balls outside off or down the leg side. On the 11 occasions he bowled at the stumps, South Africa scored just four runs. Stokes bowled significantly straighter, while Rashid – as you would expect from a decent leg-spinner – bowled all but five balls on a line between middle stump and outside off. Moeen’s default line, outside off stump, was easily milked – 47 runs coming from the 32 balls he bowled there.
Let’s fast forward to the mid-summer home ODI series and the only game England game close to losing against Sri Lanka, which ended in a fascinating tie at Trent Bridge.
On this occasion, Sri Lanka batted first and hit 286-9. It should have been a winning score when England then collapsed to 82-6 – that was until Jos Buttler and Chris Woakes had other ideas. In the Sri Lanka innings, 40 of the 50 overs were sent down by bowlers picked for the series that starts on Wednesday. They produced some varied results. Rashid took no wickets but his 10 overs cost just 36 runs, against batsmen who play spin particularly well. Ali was again much more expensive, taking 1-69. Liam Plunkett returned 2-67 and Woakes 2-56.
Ali again produced barely any variety, with 58/60 off-breaks and 55/60 missing off-stump. His lengths were an improvement on the Centurion ODI – nothing short, no full tosses – but he sent down 11 half-volleys which cost 21 runs.
The most expensive seamer on the day, Plunkett, had a problem with his lengths – a beamer, four full tosses and eight half volleys among his bag. His line was OK, but movement was hard to obtain – the one leg cutter he produced was hit for four in any event.
Woakes produced a little more movement, with five off cutters, and his lengths were better than Plunkett’s. However his economy rate would have surely been even better had he not bowled 20 balls on leg stump or down leg which produced 24 runs in all.
Why was Rashid so hard to score off? His line was excellent; by now he had added the top spinner to his armoury (so was bowling three different deliveries) and 77% of his balls were at that ideal spinner’s length just short of a half-volley.
In terms of overall stats, Rashid now has the foundations of what could become be a very good ODI career. He has played in all 10 of England’s ODIs this calendar year, picking up 11 wickets from 10 matches at an economy of 5.03. Ali’s record over the same period is six wickets from eight matches at an economy rate of 5.75. You would have to worry that having been treated to some pretty rough treatment at the hands of Pakistan’s spinners in the Tests that England might be tempted to overlook Ali at the start of the ODIs.
If they want a second spinner, then Liam Dawson, who enjoyed a fine T20 international debut in July, might be a better option.
As for the seamers, they are also a unit in flux. Mark Wood has not played an ODI in almost a year but his raw speed, seen to fine effect on Twenty20 finals day last Saturday, will surely be called upon. England’s lack of variety elsewhere – made all the more acute by the absence of the left-armer Topley alongside Willey who can swing a white ball – is a potential weakness. It has been demonstrated in this piece that Stokes, Woakes, Plunkett and Jordan – as right-arm seamers who don’t move the ball a huge amount and who all bowl speeds in the mid 80s – can struggle to be an effective unit. And that problem only exacerbates itself when they miss their yorkers, bowl too short or slide the ball down the leg side.
It sure has been fun to watch England bat of late in white ball cricket – when Jos Buttler and Jason Roy are in full cry few teams can live with them – but sometimes the pyrotechnics of the batsmen have masked the imperfections of the bowlers. Against a Pakistan side invigorated by squaring the Test series in the final match of that rubber, England might need all 11 players to pull their weight.
One of the reasons Pakistan were so comprehensively beaten by England in the second Test at Old Trafford lies in the temporary fall from grace of Yasir Shah.
The tourists’ most potent bowling weapon, fresh from collecting an exceptional haul of 10-141 in Pakistan’s Lord’s triumph, came to Old Trafford hell-bent on further destruction. By reputation and statistics, Manchester’s international ground tends to be a welcoming one for spinners, who take their wickets at the same average as seamers (the English norm is of course that seamers are more handsomely rewarded).
And yet what befell Yasir Shah? A painful match bag of 1-266, that’s what. And how on earth did that come about? Plenty of credence was put into the theory that this was the first opportunity for England to bat first against a Pakistan team featuring Shah, Alastair Cook having lost the toss in all three UAE Tests, plus the one at Lord’s. Thus, instead of facing familiar “scoreboard pressure” batting second, England could create a platform against the seamers on the first morning, and continue the momentum against the spinner.
But one issue with Shah that simply doesn’t get enough of an airing is his reluctance to produce much variety with his bowling, and in particular his lack of googlies. A leg-spinner cannot be a great leg-spinner without being able to confidently land and (to at least some extent) disguise a googly in whatever conditions he is presented with. Yet in this match Shah sent down four googlies. From 378 balls. Of the remaining 374 balls, one was classified as a quicker ball. The remaining 373 out of 378 were leg-breaks. So there came a point when the batsmen could sense there was a very, very high chance (98.94% as it happens) that the ball, on pitching, would move away from them if they were right-handed and into them if they were left-handed.
I plucked a random one-day international in which England’s leg-spinner Adil Rashid (who some felt was unlucky not to feature at Old Trafford), and spotted he bowled seven googlies and two top-spinners in his 10 overs. That means on average there’s one ball an over which will be different to his others.
Now I am not writing this to denigrate the usual excellence of Shah, whose international pedigree far exceeds Rashid’s. Shah’s 87 wickets in 14 Tests overall is a very healthy return indeed. But I am fairly convinced from watching great leg-spinners of the past – Abdul Qadir, Shane Warne and Anil Kumble – that variety is the spice of life, or at least the ingredient that contributes to a long and healthy career. Too often recently, exciting spinners from the sub-continent have endured frustratingly truncated careers and it would be nice to see Shah continue to spearhead Pakistan’s slow-bowling attack for a good while longer. To do that, he will need to make the batsman guess what might be coming out of his hand. If the man with the willow can put his mortgage on it being a leg-break, then the mental battle is less easy for Shah to win.
Another area in which Shah could improve is adapting his speed. You’ll often hear commentators mention that different wickets have different optimum speeds for spinners. Whereas he had done well at Lord’s on a wicket that was pretty slow and low and got lower as the match wore on, Old Trafford offered a faster surface where the bounce, fairly unusually, increased through the course of a match. England’s spinner Moeen Ali dropped his average pace quite considerably from the first innings to the second (and he worked quite heavily in practice with Saqlain Mushtaq to get more loop on the ball too). Moeen was successful in both innings – he was given more work to do in the second innings after Ben Stokes picked up an injury – and picked up highly satisfactory figures of 2-43 and 3-88. He bowled at an average of 54.87mph in the second innings and 53.05mph. By contrast, Shah actually bowled slightly faster in the second innings than he did in the first, just as he had done at Lord’s. In London, where the wicket got increasingly slower and lower, it was the right tactic. At Old Trafford, it wasn’t. Interestingly, the one wicket Shah did take at Old Trafford, Chris Woakes stabbing back a catch to the bowler, came from a delivery with a pace of 50.9mph. Yet his average paces were 51.73mph for the first innings and 52.18mph.
Much of the talk in the aftermath of England’s defeat to Pakistan in the first Test at Lord’s was on the contrast between Moeen Ali and his Pakistani counterpart Yasir Shah.
As Yasir sprung to the top of the ICC Test rankings for bowlers, Moeen now has his place under threat. While Yasir got more turn and had a had a higher chance of taking wickets according to nearest neighbour analysis, the difference in terms of pure data between the two bowlers was a slim one.
|England v Pakistan 1st Test, Lord's||Moeen Ali||Yasir Shah|
|Average degrees of turn||3.22||3.86|
|Average length (metres from stumps)||4.6m||4.64m|
|Average weighted runs per ball||0.43||0.43|
|Average wicket probability per ball (%)||1.2||1.34|
Moeen didn’t bowl badly. He performs at pretty close to his potential ability as possible, maintaining a consistent line and length and getting the ball to turn on a regular basis.
While he had the odd horror ball, like the ball in each innings that slipped out of his hand and got to the batsmen at head height, this was a steady performance by England’s spinner. He was just played very well by the Pakistan players, especially Misbah-ul-Haq in the first innings, who took him for 32 runs from just 23 balls, including seven fours. Misbah targeted him successfully in the first innings, but less so second time around.
This is not to say Moeen is as good a bowler as Yasir, who is clearly a world class performer streets ahead of the England off spinner, but the difference between the two is not as simple as saying Moeen bowled badly and Yasir bowled well.
The massive difference between these two bowlers is that Yasir has the ability to turn the ball both ways. Every single delivery that Moeen pitched in this Test turned from off to leg, so into the right hander and away from the left hander. There was a wide range of the degrees of turn, from 0.69 degrees to 7.9 degrees, but all of them went the same way. The batsmen knew what to expect and could adjust.
Yasir on the other hand got the ball to turn in both directions. His biggest leg break turned 11.16 degrees, his biggest turning googly moved 2.81 degrees the other way. A difference of 13.97 degrees – more than enough to keep a batsman guessing.
Batsmen can play with more confidence against off-spin than they can against high quality leg-spin. This difference in approach is very apparent in the context of these teams’ respective batting line-ups. England as a batting unit lack a plan against Yasir that the tourists have for Moeen.
The point here is that Moeen is not objectively bowling badly, he is just struggling to trouble Test class batsmen on a consistent basis. The question isn’t one of whether he is bowling well, it is whether he is good enough at this level. If it was a question of form rather than ability the fix would be an easier one, but throughout his career Moeen has made the absolute most of what he has got.
Adil Rashid has been brought into the squad for the Old Trafford Test. He has the advantage of possessing a googly that can turn the other way from his stock ball, something that Moeen has never mastered despite occasional talk of him working on a Doosra. That makes him a different challenge for the Pakistan batsmen, but history tells us that Rashid is more likely to bowl boundary balls than Moeen, not less.
With few other notable spin options in county cricket the selectors have a straight choice between premier off-spinner and leg-spinner. However, with a glut of all-rounders at their disposal they might choose both for the second Test.
Yasir, offering control and wicket-taking threat, is two bowlers in one. It is perhaps logical that England need to select two spinners to instil those characteristics in their spin attack.
The United Arab Emirates is an appropriate place to seek the fresh laying of solid foundations. England have not settled on a Test opening partnership in recent years and Alastair Cook will have another new partner as his team seeks to construct some high-rise totals in keeping with the Emirati skyline.
Six players have tried to fill the role Andrew Strauss vacated in 2012. The lack of progress is shown by the fact that the man first given the chance was the most successful. Nick Compton averaged 57.9 in his 17 opening stands with Cook; none of the subsequent five candidates have averaged above 32.3 in unison with the skipper.
Compton was partly dropped for his slow scoring, a trait that has characterised all of these partnerships – the desire to pair Cook with a more fluent scorer led the selectors to Adam Lyth, whose average first wicket run rate of 2.83 with Cook was the highest of the six combinations.
|England opening partnerships since August 2012|
|Cook and..||Partnerships||Runs||Highest||Runs per over||100 stands||Average|
Current candidates Alex Hales and Moeen Ali offer various attributes, but both have the range of shot and intent that is seemingly required in the continuing search for top order stability.
After hitting 907 Test runs at an average of 50.4 this year, Cook’s patient approach of accumulation is in good order – will it be Moeen’s elegant left-handed aggression or the powerful belligerence of Hales that provides the impetus?
The Cook – Compton axis was a crucial part of England’s success in India in 2012/13. They piled up 493 runs in their eight opening stands, at an average of 70.4. Their steady scoring rate of 2.69 runs per over was not a problem in the context of such productivity – Cook in particular went on to score heavily against toiling spinners when well-set.
However, a solid base does not guarantee success in spin-friendly environments. David Warner and Chris Rogers largely did a good job at the top of Australia’s order in their humbling 2-0 defeat against Pakistan in the UAE in 2014/15. Australia were comprehensively out-batted overall.
They averaged 53 in their four partnerships, recording their team’s highest stand of a disastrous tour, 128 in the very first Australian stand of the series at Dubai. Pakistan’s average opening partnership was 35.8, but this was the only area that the tourists out-batted the series winners.
|Average partnerships 2014/15|
The first wicket was the only one in the top seven for which Australia had a higher average partnership than Pakistan. Solid starts were wasted by an under-performing engine room: Pakistan averaged 174 for the third wicket, Australia 16.3. The disparity was 170.5 runs for the fourth wicket.
Australia’s batsmen were blown away in the UAE in 2014/15. England will need to have more than a steady opening partnership if they are to prosper against Pakistan’s talented bowling unit.
LATEST FROM THE BLOG
- CricViz Analysis: Five Key Numbers from IPL Week 2April 20, 2018 - 10:57 am
- CricViz Analysis: Five Key Numbers from the IPL this WeekApril 13, 2018 - 12:00 pm
- CricViz Analysis: County Championship 2018 PreviewApril 12, 2018 - 3:21 pm
- CricViz Analysis: The Demise of Moeen AliApril 5, 2018 - 7:39 am
- Morne Morkel – Legend, or Just Great Fun?March 30, 2018 - 7:37 am