Here you will find detailed analysis highlighted from the CricViz app.

Measuring boundary attempts in the IPL

Boundary-hitting is arguably the most important skill in Twenty20 cricket, and is therefore a critical factor in assessing performance.  Metrics such as boundary runs and boundary-balls-percentage don’t give us the full picture.  A batsman who scores 65% of their runs in boundaries might be below average at rotating the strike.  A batsman who hits half his deliveries to the fence could be swinging every other ball.

Ideally, we need to know how often a batsman attempts to hit a boundary and how successful they are when doing so.  A boundaries attempted metric will be quite subjective.  When a batsman has a big swing, it’s obvious what his intentions are regardless of the outcome.  On the other hand, a simple leg-glance for a single might turn into a boundary if timed well enough.  It’s also important to consider how the field is set and the situation of the game to give us extra clues about the batsman’s plans.  However, without someone sitting down and explicitly recording boundary attempts in every match (as is done with ESPNcricinfo’s control metric) we will have to infer it from ball-by-ball data.

Defining a boundary attempt

Our data records 22 types of shots a batsman could play including all manners of drives, sweeps, pulls and cuts.  Using data from over 2,000 T20 matches, we can analyse which types of shots are most likely to result in a boundary.

shot typeboundary %
Slog-sweep40
Upper Cut34
Scoop34
Switch Hit31
Pull29
Slog27
Hook27
Reverse Sweep26
Drive23
Sweep22
Late Cut18
Flick18
Cut17
Glance14
Steer8
Fended5
Worked2
Pushed1
Dropped1
Backward Defensive1
No Shot1
Forward Defensive1

Slog-sweeps result in a boundary 40% of the time followed by upper cuts and scoops on 34%.  The top performing shots on the list down to and including reverse sweeps we may reasonably assume are played with the intention to hit a boundary.  So let’s draw the line here and use these top 8 shots as our proxy for boundary attempts.  Drives and conventional sweeps do result in boundaries but we’re not confident enough that they are always boundary attempts.

We also have data on what connection the batsman makes with each shot.

batting connectionboundary %
Middled83
Strong60
Well Timed40
Outside Edge27
Top Edge25
Thick Edge24
Gloved13
Inside Edge10
False Shot10
Bottom Edge8
Leading Edge3
Mis-timed3
Neutral2
None2
Padded1
Bat-Pad1
Hit Helmet1
No Shot0
Missed (Leg Side)0
Hit Body0
Hit Pad0
Missed0
Left0
Play and Miss0
Shoulders Arms0
Play and Miss (Leg Side)0

Middling the ball or getting a strong or well-timed connection results in quite high boundary percentages.  We will take these three shots to add to our definition of a boundary attempt.  Finally we will assume all free hits are boundary attempts.

Boundary attempts in the IPL

In this season’s IPL, there have been 2,484 boundary attempts from the 32 matches so far.  That’s nearly 2 per over.  Our definition covers 92% of all boundaries scored i.e. about 8% of boundaries are unintentional.  The average boundary-success rate across the tournament is 49%.

The graph above shows the boundary success rate broken down by team.  This correlates quite well with the current standings; Mumbai Indians have the highest success rate and are one of the form teams at the moment.  Contrast this with RCB who have a success rate of 10 percentage points fewer, near the bottom of the table.

Gujarat Lions attempt by far the most boundaries per 120 balls faced.  Their batting lineup, which includes Raina, McCullum and Finch, are making a concerted effort to hit as many balls to the fence as possible.  However, their below-average success-rate suggests they’re not executing their plans.  Interestingly, Mumbai Indians attempt the fewest boundaries of all the teams despite having the highest success rate.  The likes of Nitish Rana, Jos Buttler and Pollard are choosing their boundary options with more care and it’s been working effectively so far.

Boundary attempts by player

Let’s take a look at boundary attempts on an individual batsman level.  The table below shows the 47 players to have attempted at least 20 boundary hits.

batsman nameboundary attemptsintentional boundariesboundary success %balls facedballs per boundary attempt
Manan Vohra3625691243.44
Sunil Narine382668822.16
Ajinkya Rahane3624671704.72
David Warner6945652543.68
Shaun Marsh412663952.32
Sanju Samson5333621963.70
Jos Buttler5031621563.12
Moises Henriques3622611444.00
Hashim Amla6640612103.18
Sam Billings2716591053.89
Chris Gayle3420591303.82
Hardik Pandya291759792.72
Robin Uthappa7745582002.60
Kedar Jadhav5029581342.68
Kane Williamson331958932.82
Gautam Gambhir8247572873.50
Yuvraj Singh301757732.43
Kieron Pollard4123561583.85
Nitish Rana5128552084.08
Manoj Tiwary351954862.46
Brendon McCullum7238531882.61
Glenn Maxwell5730531111.95
Steven Smith5931532093.54
Suresh Raina7438512212.99
Jason Roy201050412.05
Krunal Pandya261350803.08
Rahul Tripathi5728491442.53
Chris Morris381847852.24
Parthiv Patel5124471402.75
Shreyas Iyer3215471083.38
Chris Lynn321547742.31
AB de Villiers3717461173.16
Aaron Finch6228451121.81
Rohit Sharma291345973.34
Manish Pandey5826451943.34
Ishan Kishan251144532.12
Virat Kohli4118441333.24
Dwayne Smith321444732.28
Rishabh Pant371643952.57
Yusuf Pathan281243812.89
MS Dhoni3816421303.42
Shikhar Dhawan9037412502.78
Mandeep Singh22941823.73
Axar Patel2711411063.93
Dinesh Karthik4819401443.00
Ben Stokes301137963.20
Karun Nair25936903.60

Sunil Narine, in his role at the top of the KKR batting order, has the second highest boundary success rate of 68%.  He also has the fifth-lowest balls-per-boundary attempt figure – every other ball.  At the other end of the scale, Pune’s Ben Stokes has been struggling, producing a 37% success rate.  Meanwhile, Stokes’ teammate Ajinkya Rahane attempts a boundary almost every 5 balls – the highest in the list.  However he does have the 3rd highest boundary success rate suggesting he is quite picky over which balls to target.  This approach is perhaps not serving him so well as he has the lowest strike rate out of the top 20 run-scorers of the season so far.

This boundary-attempts metric has been fairly crudely formulated in this article. But there is clearly potential to lend insight into how teams and batsmen approach a T20 innings, and contribute to a more comprehensive analysis of a side’s performance.

Imran Khan, @cricketsavant

The top 10 T20 death bowlers: IPL 2017

Now that cricketers have dispensed with the tedium of a winter rest (“Sleep’s for wimps”; Derek Trotter, 1985), instead plying their trade on the global cricketing treadmill, we are better equipped to look at recent form. There are almost no significant gaps in the T20 calendar, so a bowler who rocks up for one tournament in April probably had a stint in another tournament a few months prior. Their form guides are more accurate now than, say, in Test matches twenty years ago when – sweet mercy – cricketers had months to themselves, or spent it playing cards on the boat back home from somewhere hot. Or in the case of Andrew Caddick, learning to fly helicopters while painting Taunton’s pavilion – disappointingly not at the same time – as a memorable issue of Cover Point once regaled.

 

For us, it’s great news. In years gone by we’d look at a small sample of a player’s stats, based on last season, and make little of it. Now, however, we have an almost continuous sample of live data coming in which gives us a much broader view of the game’s critical areas, while ensuring it’s recent and relevant.

 

All of which leads to the following list of bowlers, and how they perform at the death (any bowler who has bowled more than six overs since the start of 2016, in overs 16-20).

IPL economy rates at the death

PlayerTeamRunsBallsMatchesWicketsEconTypeYear
Shane WatsonRoyal Challengers Bangalore17411016129.49pace2016
Joe LeachWorcestershire15410512108.80pace2016
Mitchell McClenaghanMumbai Indians1301011497.72pace2016
Harry GurneyNottinghamshire124921198.09pace2016
Graham NapierEssex129851299.11pace2016
Jamie OvertonSomerset78721096.50pace2016
Mustafizur RahmanSunrisers Hyderabad1821521687.18pace2016
Matthew TaylorGloucestershire1291041287.44pace2016
Chris MorrisDelhi Daredevils64681275.65pace2016
Graham WaggGlamorgan (Wales)100661279.09NULL2016
Rumman RaeesIslamabad United4652775.31pace2017
Dwayne BravoGujarat Lions1951211579.67pace2016
Benny HowellGloucestershire85611478.36pace2016
Sunil NarineMelbourne Renegades8860678.80spin2017
Tymal MillsSussex89701167.63pace2016
Lasith MalingaSri Lanka6543469.07pace2017
James FullerMiddlesex86731167.07pace2016
Mitchell ClaydonKent145981168.88pace2016
Matt QuinnEssex126871468.69pace2016
George EdwardsLancashire87551069.49pace2016
Wahab RiazPeshawar Zalmi6372965.25pace2017
Chris JordanSussex6167965.46pace2016
Mohit SharmaKings XI Punjab143931469.23pace2016
Andrew TyeGloucestershire128801369.60pace2016
Umesh YadavKolkata Knight Riders8252969.46pace2016
Chris RushworthDurham100661269.09pace2016
Ben LaughlinAdelaide Strikers3736566.17pace2017
Paul CoughlinDurham61531166.91pace2016
David GriffithsKent139941368.87pace2016
Jade DernbachSurrey (England)6967656.18pace2016
Ashok DindaRising Pune Supergiants7752958.88pace2016
Rory KleinveldtNorthamptonshire96711058.11pace2016
Mohammed ShamiDelhi Daredevils6246858.09pace2016
Yuzvendra ChahalRoyal Challengers Bangalore69431349.63spin2016
Anwar AliQuetta Gladiators8256948.79pace2017
Sandeep SharmaKings XI Punjab100661449.09pace2016
Jasprit BumrahMumbai Indians136991448.24pace2016
Andrew CarterDerbyshire8455849.16pace2016
Michael HoganGlamorgan (Wales)45481345.63pace2016
Kesrick WilliamsWest Indies4336447.17pace2017
Ben DwarshuisSydney Sixers6043448.37pace2016
Tymal MillsQuetta Gladiators3240544.80pace2017
Usman ArshadDurham137921448.93pace2016
David WilleyYorkshire5244847.09pace2016
Bhuvneshwar KumarSunrisers Hyderabad1661111748.97pace2016
Mark SteketeeBrisbane Heat4242646.00pace2017
Ravi AshwinRising Pune Supergiants64601446.40spin2016
Ben HilfenhausMelbourne Stars6052836.92pace2017
Keaton JenningsDurham64431538.93pace2016
Jasprit BumrahIndia4143535.72pace2016
Murugan AshwinRising Pune Supergiants74521038.54spin2016
Shane WatsonSydney Thunder3937536.32pace2017
Zaheer KhanDelhi Daredevils110671239.85pace2016
Lewis GregorySomerset67641136.28pace2016
Sunil NarineLahore Qalandars5537838.92spin2017
Tom CurranSurrey (England)83661337.55pace2016
Imran TahirNottinghamshire6644639.00spin2016
Wahab RiazPakistan60364310.00pace2017
Oliver Hannon-DalbyWarwickshire120801339.00pace2016
Chris JordanRoyal Challengers Bangalore7246939.39pace2016
Dhawal KulkarniGujarat Lions88611438.66pace2016
Tymal MillsEngland4538337.11pace2017
Hasan AliPeshawar Zalmi86641038.06pace2017
Jake BallNottinghamshire83571038.74pace2016
Mohammad SamiIslamabad United7057937.37pace2017
Barinder SranSunrisers Hyderabad60471437.66pace2016
Tim BresnanYorkshire93641438.72pace2016
Sohail KhanKarachi Kings7546929.78pace2017
Mohammad NawazQuetta Gladiators76491029.31spin2017
Shiv ThakorDerbyshire69531127.81pace2016
Nuwan KulasekaraSri Lanka4237726.81pace2017
Clinton McKayLeicestershire50561225.36pace2016
Piyush ChawlaKolkata Knight Riders47381127.42spin2016
Jordan ClarkLancashire74491029.06pace2016
Tino BestHampshire5537928.92pace2016
Dwayne BravoSurrey (England)5641628.20pace2016
Andrew TyePerth Scorchers5939729.08pace2017
Richard GleesonNorthamptonshire61531026.91pace2016
Sunil NarineKolkata Knight Riders63471128.04spin2016
Tim SoutheeMumbai Indians57361129.50pace2016
Johan BothaSydney Sixers4636627.67spin2017
Matt HenryWorcestershire65461018.48pace2016
Ravi BoparaEssex99741518.03pace2016
Mohammad AamerKarachi Kings75511018.82pace2017
Jeetan PatelWarwickshire49361218.17spin2016
AzharullahNorthamptonshire72571207.58pace2016
Wahab RiazEssex5036508.33pace2016

 

Shane Watson, the interim captain for RCB in IPL 2017, was their go-to death bowler last season; his strike-rate was good, but economy was higher than the average. Jamie Overton, who has been on the cusp of an England career for a while, took death wickets for Somerset last year and kept his economy under 7/over, which isn’t approaching deadly-assassin status, but something close to it.

 

Chris Jordan, however, struggled in his the 2016 IPL (economy approaching 10/over) before following it up with a tighter performance for his home team Sussex a few months later. The green, green grass of home.

 

Try sorting the table and looking at who might pop up when we review the IPL mid-way through the tournament.

Say Elo to the IPL

The IPL, about to enter its 10th season, pits teams that are somewhat evenly matched and compete in more or less homogenous conditions.  This coupled with player drafts every three years should result in a competitive and exciting tournament every season.  Six different winners in nine seasons suggests this has largely been the case.  Imran Khan looks deeper into how certain teams have performed throughout their history by considering their Elo rating – a system that evaluates teams purely on their results.

Elo ratings introduction

A team’s Elo rating indicates its relative strength compared to other teams.  When a team wins a game it gets transferred a certain number of points from the other team i.e. the total points for both teams stays the same.  This ensures the average across all the teams stays roughly constant.  Additionally, a stronger team will gain fewer points when beating a weaker team than the other way around.  At the very beginning, every team is assumed to be of equal strength so start off with the same Elo rating (in our case it will be 1500).

Let’s say we have a match between Team A and Team B.  Team A is stronger and has an Elo rating of 1600 compared to 1400 for Team B.  The probabilities of the teams winning can be calculated to be 76% and 24% respectively (assuming there are no ties).  If Team A does indeed win, their Elo rating will go up to 1604 and Team B down to 1396.  However, if Team B causes a small upset and wins its Elo rating will go up 11 points to 1411 while Team A goes down to 1589.  The intuition is that we would like to reward underdogs more for winning than we reward favourites for winning.  Note that the total number of points is the same before and after the game.

A brief history of the IPL

Of the 584 scheduled IPL matches to date, there have been 568 outright winners, 6 ties and 10 no results.  I have counted the winners of the super over after a tie as the overall winner and given half a win each to the teams involved in no results.  The plot below shows the Elo ratings for each of the 13 IPL teams to have existed over the course of every season.

Although this looks quite cluttered at first, we can distinguish some general trends.  The league was fairly closely contested in the first four seasons with 2010 the most tightly packed.  No team reached an Elo rating of beyond 50 points from the average of 1500 in that season.  In fact, 4 points separated 6 teams in the final standings with even bottom-placed Kings XI Punjab taking points off both eventual finalists.  From 2012, things started to spread out a bit more as Chennai Super Kings dominated and some teams, in particular Delhi Daredevils, started to fall away.

Chennai Super Kings

The Super Kings, the most successful IPL team so far, won titles in 2010 and 2011, have a win rate of 61% and have made the playoffs in every season in which they participated.

During CSK’s first title win in 2010, the Elo ratings suggest they were far from the best team for much of the season.  They scraped into a playoff spot on net run rate.  They also won as many games as they lost in the group stages beating teams that eventually finished in the top 4 on only two occasions.  In seasons 2013-2015, the Elo ratings suggest that there was a significant gulf between them and most of the other teams.  The finished top of the group in 2013 and 2015 but ultimately stumbled at the playoff stage.  What the graph above illustrates quite well is their performance peaking in the middle of each of the 2013-2015 seasons as opposed to at the end in 2010/2011.

CSK enjoyed the highest Elo rating of all time when they beat Kolkata Knight Riders by 2 runs in the middle of the 2015 season to take their rating up to 1609.  We can conceivably say that this CSK team was the best of all time during that period.  However, it was not to continue having staggered through the latter half of the 2015 season and losing in the final to the Mumbai Indians.

Daredevils and Warriors

In contrast, the Daredevils have been in a steady decline since 2010 after finishing 4th, 3rd and 5th in the first three seasons.  Apart from in 2012 where they exhibited a brief resurgence, Delhi have finished at or near the bottom in every season since.

Their lowest point occurred at the beginning of the 2015 season.  However, this was not the lowest of all time; the dubious honour being claimed by the now defunct Pune Warriors India.

The Warriors finished in the bottom two in each season of their fleeting existence winning only 27% of their games.  A run of 9 consecutive defeats culminated in an Elo rating trough of 1378 points when they lost to the Mumbai Indians near the end of the 2013 season.

Biggest upsets

We can use the Elo ratings to compare the relative strengths of teams before each match.  Of the 570 matches in which there was a result, 55.4% of favourites, according to Elo, won the game.  If a team with a lower rating beat a team with a higher rating, we can quantify how much of an upset this was.

The table above shows the ten biggest upsets defined by the difference between the ratings of the two teams.  The top match on the list was a surprise in more ways than one.  Chennai were coming off the back of their peak Elo rating and Delhi were near the bottom of theirs.  In that match, Delhi restricted CSK to just a run a ball and knocked them off with ease.  The Daredevils also feature in a further six of the matches in that list highlighting how, in the last few seasons, any victory was a seen as a shock.

Biggest contests

We can also consider which games were of the highest quality defined by the sum of the Elo ratings for both teams.

According to the system, the final of the 2015 season between Mumbai and Chennai was the highest rated match of all.  Both teams had quite similar ratings.  But as the graph shows below, Mumbai were the form team going into the final while Chennai were riding on their early-season performances.  It’s probably no surprise that Mumbai won by a pretty hefty margin.

The Mumbai/Chennai rivalry takes up eight of the ten slots in the list.  Mumbai follow a similar pattern to Chennai in terms of the Elo ratings although they are slightly out of phase.  They tend to get off the mark slowly at the start of the season then surge to peak near the end.  This is most noticeable in seasons 2008 and 2013-2015.

Improvements to Elo

The Elo ratings are based on a simple concept – a team is credited for winning and penalised for losing, while underdogs are credited more for winning etc.  The ratings can be refined by considering home advantage, major team changes after auctions for example and the margin of victory.  We can also dynamically change the importance of certain matches. For example, playoff matches may offer greater payoffs in Elo points while a team’s matches from several seasons ago can be discounted in value.

Imran Khan, @cricketsavant

Tigers find their bite in Colombo

You may be familiar with the CLR James epigram which features in the preface of Beyond A Boundary: “What do they know of cricket who only cricket know?”

I often find a total break from the game – in my case a week covering horse racing’s Cheltenham Festival – gives me renewed energy to enjoy cricket, and a hunger to delve into the unique CricViz stats that underpin a particular match.

In this case, I am drawn inexorably to Bangladesh’s maiden Test victory over Sri Lanka. Put in the shade as it invariably will be by the titanic continuing struggle between India and Australia (1-1 heading to a Dharamsala decider, by the way) I certainly feel it deserves some extra attention.

First of all, there is the sheer landmark nature of this result. In their 100th Test, this was only Bangladesh’s ninth win and their first away from home against a team other than Zimbabwe or West Indies. It also comes five months after their Dhaka win against England which followed a winless year in 2015.

Then there was the less than ideal background to the win: a heavy defeat in the first Test, serious pressure and rumours of an impending axing for the captain Mushfiqur Rahim, an injury to the wicketkeeper Liton Das and three other players dropped after the Galle setback.

WinViz suggests it was the Shakib innings that turned Bangladesh from underdogs to favourites

And finally there was the troubling scorecard late on day two in Colombo: Bangladesh up against it at 198-5 on day two, some 140 runs behind. WinViz had a Sri Lanka win at 62% with Bangladesh at 27%: not a hopeless position for the tourists but an unencouraging one.

It was at this stage that Shakib Al Hasan crafted one of the most important centuries of his career. A naturally exuberant player (his strike rate exceeds that of all top current batsmen other than David Warner) he elected to curb his instincts to some degree but still scored at a healthy rate.

He watched the ball onto the bat well: only playing and missing five times from 159 balls faced while producing only two outside edges. When attacking, he timed the ball well – indeed our analysis shows he mistimed just two shots, the second of which finally brought his dismissal on 116, an innings which turned a probable Bangladesh deficit on first innings into a very valuable lead of 129.

WinViz suggests it was the Shakib innings, alongside valuable contributions from Mushfiqur and Mossadek Hossain, which turned Bangladesh from underdogs to favourites. But sometimes the hardest thing in a Test match is to reinforce a dominant position, or to get the job done when you hold all the aces – particularly if you’re a team without much experience of winning.

The next part of the job was carried out by Bangladesh’s bowlers, led by the hugely exciting fast bowler Mustafizur Rahman, backed up admirably by Shakib’s resourceful slow left-arm stuff.

There have been plenty of false dawns for Bangladesh fans in the past, but the rare successes are worth cheering

 

The key period came just after lunch on day four when these two bowlers operated in tandem and the draw had moved in excess of 50% probability on WinViz. Sri Lanka were 137-1, nudging into an overall lead – but suddenly Bangladesh found their bite.

The first breach came when Mustafizur had Kusal Mendis caught behind with a delightful delivery. It was the last ball of the over, and at 79.8mph it was the fastest too. Pitching on a fairly full, almost half-volley length – 5.9m from the stumps – it induced the drive.

All six balls in the over offered to swing away from the right-hander. But unlike two previous balls, which had carried on with the angle after hitting the wicket, this one straightened just enough (moving 0.9° degrees away from Mendis) to take the outside edge. Mushfiqur, the stand-in keeper as well as captain, gleefully accepted the chance.

Five overs went without a wicket before Mustafizur struck again. Continuing with a full length, he had Dinesh Chandimal fishing well wide of off-stump and nicking off. This was not per se a brilliant delivery, but an intelligent one, the sort with which Ian Botham used to take countless wickets. A tempting outswinger sometimes looks like it’s there to be hit. But Chandimal had not been at the crease long enough to play a relatively risky cover-drive and paid the price.

It was Shakib’s turn to get involved next: Asela Gunaratne lbw padding up for just seven. A misjudgement for sure, but again the bowler’s skills played their part: this ball drifted a fair bit, 2.5° into the right-hander who felt that on his initial observation he could afford to let this one bounce and turn away from him. The thing is the extra drift meant the ball was arrowing into the stumps and relatively modest turn away (2.6°, around half of the previous ball’s turn) meant it was straight enough to be hitting.

PlayViz recorded a -38 fielding score for the hosts

Shakib’s next wicket soon followed, a dismissal that reduced Sri Lanka to 190-5 (effectively 61-5). Bowling with a lovely rhythm, Shakib was getting some deliveries to turn really quite sharply, others to skid on without any turn at all. At times like these, batsmen often believe their best bet is to premeditate, and invariably out comes the sweep shot.

Niroshan Dickwella played one such sweep to a ball that turned a lot (6.6° in fact, putting it in the top dozen of Shakib turners for the innings). Also, the length was short of ideal length for sweeping, so the ball had time to turn and bounce before Dickwella’s bat made contact with the ball. Mushfiqur, who had a fine game behind the stumps, anticipated everything smartly to move across to complete the catch.

Sri Lanka fought on. The ninth wicket put on 80 to leave Bangladesh some kind of challenge, namely a target of 191. However the Tigers would not be denied and man-of-the-match Tamim Iqbal hit 82 (Shakib arguably had stronger claims to that individual gong). A memorable victory was achieved with four wickets in hand.

A final footnote: Sri Lanka have been poor in the field for much of the past six months and PlayViz recorded a -38 score for the hosts against a +52 aggregate for Bangladesh. The differential in batting was even more stark at +141 for the winning side (who had the clear disadvantage of batting second) against Sri Lanka’s -21. These indicators are very welcome for Bangladesh going forward while Sri Lanka’s side, still in transition following the retirements of so many key players of late, could find more roadblocks in their path.

2017 Pakistan Super League Review II

This article is contributed by Imran Khan, who you can follow on Twitter @cricketsavant.

 

Daren Sammy led an inspired Peshawar Zalmi side to take the honours at the second edition of the Pakistan Super League. CricViz takes a statistical look at what took place during the tournament.
General Overview
Peshawar Zalmi’s opening batsman Kamran Akmal topped the run scorers list with 353 runs from 11 innings including two fifties and the only century of the tournament. The next Peshawar batman on the list was Shahid Afridi in 11th place with 177 runs and one fifty. The champions largely relied on one-off performances from their batsman, with Dawid Malan, Mohammad Hafeez, Eoin Morgan along with Afridi all registering one fifty each during the tournament. Peshawar’s relatively modest batting numbers
were generally due to chasing down or defending low scores.

Seam bowlers dominated the top wicket-takers list, with only Usama Mir, Sunil Narine and Mohammad Nawaz representing the spinners.  However, spinners did have a lower economy rate and strike rate than seamers over the whole tournament.  Wahab Riaz and Hasan Ali, who picked up 27 wickets between them, spearheaded the bowling attach for the champions Peshawar. We’ll investigate what made them successful later on.

Kamran Akmal

Tournament top run getter Kamran Akmal was most destructive against right-arm spinners and medium pace seam bowlers, against whom he scored 150 runs at a strike rate of 195.

The beehive plot below shows all of Akmal’s boundaries against these types of bowlers.  He rocked onto his back-foot and seized on anything marginally short from the slow bowlers or went down the track to hit over the top.

In fact, Akmal scored 55 runs from the 36 occasions he played a shot off his back foot against right-arm spinners and medium-pace seamers.  He also accrued 31 runs from the 10 balls he went down the track during the tournament.

Against fast seam bowling however, Akmal was restricted to a run a ball and dismissed three times.

The heat map above is generated from his dot balls against fast seam bowlers with his dismissals in red.  He was generally kept quiet by good to back-of-a-length deliveries just outside off stump.  In fact, from these types of deliveries Akmal scored 46 runs at a strike rate of 90 – well below his overall strike rate of 129.

Akmal was also relatively constrained by left-arm orthodox bowling scoring at a little over a run a ball and being dismissed four times.

The pitch map above shows a distinct cluster of dots and singles (red and blue) where the bowlers like Imad Wasim and Hasan Khan pitched it on a very good length. However, on the few occasions where they bowled full or a touch shorter, Akmal would take advantage.

Chris Gayle

The talk before the PSL was of Chris Gayle’s overwhelming superiority in T20 cricket having scored nearly 10,000 runs in the format.  However, he is still 63 runs short after the end of the tournament having scored just 160 runs at an average of less than 18 and a meagre strike rate of 115.

Gayle struggled against fast seam and off spin bowlers striking at less than a run a ball and being dismissed four times each.

The heat map above shows the distribution of dot balls faced by Gayle against fast bowlers with his wickets in red.  Back of a length and outside off stump is where most of the dots are concentrated.  This is a bowling plan that has historically kept him quiet.  In fact, he scored 25 runs from 24 balls including just 4 boundaries when facing deliveries in this region.

Gayle did not fare particularly well against right-arm off spin and left-arm orthodox bowling.  The heat map above shows that his dot balls generally came when they bowled very straight to him on a good length, again something that spinners have exploited previously.  Of the 21 balls he faced from off spin and orthodox bowling and which were on the stumps, he scored only 12 runs and timed exactly one ball (which was hit for six).

The best bowlers

Top wicket-taker, Karachi Kings’ Sohail Khan utilised his slower ball to great effect.  He sent down 44 throughout the tournament earning him six of his 16 wickets whilst only conceding a run a ball (compared to his overall economy rate of 7.61). Second highest wicket-taker Wahab Riaz fared just as well conceding just 28 runs from his 35 slower deliveries and picking up two wickets.

Sohail Khan’s slower ball was very consistent in terms of speed as the histogram shows above.  Wahab’s distribution is less pronounced indicating that his slower balls were a variety of speeds, with a slight bump at 130 kph.

21-year-old leg-spinner Usama Mir who took 12 wickets for Karachi Kings, the highest for any spinner, predominantly stuck to his stock delivery during the tournament.  However, he did send down variation in the form of googlies and quicker balls a total of eight times, from which he picked up three wickets and conceded only eight runs.

Mir bowled his quicker delivery more than the googly as the very slight bump in the positive spin region illustrates.  Compare this to Sunil Narine who more or less bowled his variations as often as his off-break delivery – something no doubt Mir will gain the confidence to do in his career.

Ground analysis

Every match of the tournament was played on two grounds – Dubai and Sharjah, with the exception of the final which was held in Lahore.  This allows us to reliably compare the two grounds to see how different conditions may have affected team strategies.

The table below summaries some statistics for seam bowlers for all matches in both grounds.  Seam bowlers enjoyed a lower economy rate in Dubai, although at the cost of a slightly higher average and strike rate.

Similarly for spinners, Dubai offered a lower economy rate by 0.5 runs an over, while their average and strike rate were higher.

We can ask whether these differences came about due to the different pitch conditions in the two grounds.  The table below summarises the average off the pitch and in the air movement for seamers and spinners in degrees.

Although Sharjah offered a little more swing for the seamers, they extracted a lot more seam movement in Dubai. Dubai also saw more drift and a lot more spin.

The histogram above shows the distribution of spin at both the grounds.  Both grounds are slightly skewed towards negative values of spin (indicating spin away from the right-hander and in to the left-hander).  Sharjah has a much higher and narrower peak demonstrating the relative lack of assistance from the pitch for spinners.

Importance of the toss

In every game bar one the winner of the toss chose to bowl first.  The table below shows the same attributes from before but separated by first and second innings.

It’s evident that the second innings offered a bit more for every factor but not significantly more – certainly not enough to convince captains to bat first.

As has become a wider trend in T20 cricket, knowing your target is advantageous enough to always bat second regardless of conditions.

The graph above shows the average worm for matches where the team batting second won the game in this year’s PSL.  The green line shows that teams generally start off quicker in the Powerplay before slowing down somewhat to keep wickets in hand until the tenth over.  After this they overtake the first innings worm to chase down their target in the latter overs.

Imran Khan, @cricketsavant

2017 Pakistan Super League Review

A statistical summary of the 2017 Pakistan Super League.  

Read more

Analysing Australia ahead of the 2017-18 Ashes

Australia pulled off an impressive victory within three days against India in Pune, their first win in India since 2004.  This comes off the back of a middling home season where they lost 2-1 to South Africa but recovered to beat Pakistan 3-0. CricViz looks ahead to what England might expect later this year in the 2017-18 Ashes

General overview

Australia lost the first two Tests of the season to South Africa with their worst performance occurring in Hobart, losing by an innings-and-80 runs.  They then won the day-night contest in Adelaide before sweeping aside Pakistan.  The chart below shows the scores and the number of wickets they lost in each of their 11 innings.

The blue bars (indicating victories) show that they generally racked up big scores in the first innings and either scored quick runs before declaring or chased down small targets in the second innings.

Driving these performances were Australia’s top order, three of whom scored at least 500 runs over the six matches.  The table below shows their top ten run-scorers.

Newcomers Peter Handscomb and Matt Renshaw have adapted to Test cricket quickly, scoring three hundreds and three fifties between them. This includes a Renshaw 184 against Pakistan in Sydney.

Australia’s attack was spearheaded by their opening bowling duo of Hazelwood and Starc, picking up 60 wickets between them, and adequately supported by Nathan Lyon’s offspin (17 wickets).  We will take a closer look at what made them so successful later in the piece.

The dependable Steve Smith

Firstly, let’s take a look at Australia’s captain and top run-scorer this season, Steve Smith, who scored two hundreds and three fifties from 11 innings.  The beehive plot below shows where he scored his 653 runs.

From first glance there isn’t much we can say.  Perhaps he puts away a lot of full or wide balls to the boundary and picks up singles from balls closer to the stumps.  We can filter this further to see how he performed against particular bowlers and types of bowlers.

Smith faced a lot of Yasir Shah so it’s not surprising that he scored the most runs and was dismissed by him most often.

The heat map above shows the distribution of deliveries faced by Smith from Yasir with his dismissals in red.  He favoured a good length just outside off-stump, shown by the dark green regions.  Smith pounced on anything marginally short or full.

The blue balls show Smith’s boundaries many of which are just above the dark green areas.  The pitch map below also illustrates how Smith punished Yasir for bowling too short or too full.

David the Destructive

Warner also enjoyed a prolific season scoring nearly 600 runs at a remarkable strike-rate of 93.

Warner targeted the spinners more than the seamers, scoring at a strike rate of 113.  The heat map below shows how he scored his boundaries against spinners.  Deliveries wide of off-stump and fractionally short were, more often than not, cut to the boundary.   

In contrast, the heat map below shows the distribution of dot balls for Warner.  Spinners who bowled closer to middle and leg stump with a more consistent length generally kept Warner quiet.

Against seamers it’s a similar story with Warner dispatching deliveries wide of off stump of any length, shown in the heat map below.

A seam bowler’s best bet to restricting Warner to dot balls is to bowl back of a length on off-stump as the heat map below shows.  There isn’t really an obvious pattern in his wickets (shown in red) which suggests his dismissals come about from a lack of concentration or simply one hit too many.

Starc and Hazlewood

Australia’s opening pair put in a big shift for their side, between them bowling more than half the total overs in their six home Tests.  The heat maps below show how they bowled to both right and left-handers with their wickets in red.

They both bowled quite consistently slightly back-of-a-length just outside off-stump.  They did however get many of their wickets from fuller and straighter deliveries indicating that they used movement in the air and off the pitch pretty effectively.

The histogram below illustrates how much Starc and Hazlewood swung the ball.  Negative values of swing, measured in degrees, indicate that the ball swung away from the right-hander and swung in to the left-hander and vice versa.  Starc mostly favoured outswing (to right-hand batsman), while Hazlewood employed inswing the majority of the time.  However, the distributions overlapped suggesting both bowlers had a number of deliveries that swung in both directions.  It’s notable how similar the distributions are in terms of height and width – both bowlers had similar plans in terms of how often they bowl their stock delivery compared to their variations.

We can take a look at how much Starc and Hazlewood swung the new ball and when it got older.

The graph above shows the absolute value of the swing (how much it swung regardless of direction) during a particular over.  A moving average of six overs is taken to dampen out the fluctuations.  Both bowlers swung the new ball, the magnitude of which steadily declined until the 10th over.  Hazelwood is generally a bigger swinger of the ball up until the 30th over, when Starc starts to visibly make use of reverse-swing between the 30th and 50th over.  Hazlewood swings the ball most prodigiously with a 70 to 80 overs old ball, although it should be noted that he only bowled two overs in this period.  When the second new ball is taken after 80 overs, a similar trend is seen as with the first new ball.

Additionally, Starc and Hazlewood extract pretty much the same assistance from movement off the pitch.  The histogram below illustrates this where, as before, negative values indicate movement away from the right-hander etc.  The opening pair marginally favour seam movement away from the right-hander but are more than capable of bringing it back in or away from left-handed batsmen.

Nathan Lyon

We can take a quick look at a similar plot that shows how much spin Nathan Lyon gets.

Lyon has quite a broad distribution indicating that he varies the amount of revolutions he imparts on the ball, as well as being a consequence of the different pitches he bowled on.  There is a slight bump at 0 degrees – his quicker and flatter delivery which he bowls about 5% of the time.

The beehive plot below shows Lyon’s release points when bowling.  Over the wicket, he is fairly consistent bowling quite wide of the crease.  When going around the wicket, he varies his release point a bit more, bowling from quite close to the wicket to very wide of the crease.

Ground analysis

Finally, we can also take a look at which grounds are most conducive to swing and spin bowling.

The table above shows the average swing, seam and spin in degrees at each of the grounds that hosted a Test this season.  England will be playing in all these grounds apart from Hobart.  The most swing-friendly ground was Perth although it was also the least seam-friendly and spin-friendly ground.  Brisbane offered the most spin of all the grounds.

The graph above shows distributions for Perth and Adelaide, the grounds with the highest and lowest average swing.  It is evident the Perth has a shorter and wider distribution indicating a large range of inswingers and outswingers.  Adelaide has a narrow range centred around 0 degrees, although there is a slight bump towards fairly big outswingers.  This data coupled with knowledge about how much a ball swings when it is a certain number of overs old can be exploited by England when choosing how many seamers to play and when to bowl them.

Imran Khan is a contributor to CricViz and the @cricketsavant

India can still take the series

The India v Australia Test series has been set up for the remainder of the rubber quite beautifully thanks to Australia’s comprehensive win in the Pune opener.

India began by losing the toss on a dry, excessively spin-friendly surface, but after bowling Australia out for 260, the hosts’ WinViz moved up to just shy of 80%. This looked reasonably justifiable given the Indian batsmen’s renowned prowess in their home conditions and, it appeared, no obvious match-winning spinner in the Aussie ranks.

But the extraordinary events of day two: India all out for 105, Australia 143-4 for a lead approaching 300, turned the match unexpectedly and decisively. WinViz had moved in one way only during that second day, and by the end of it Australia were 88%.

There was to be no twist in the tale on Saturday as the slow left-armer Steve O’Keefe once again proved Australia’s hero. He took his second six-wicket haul of the match and Australia won by 333 runs without recourse to days four or five. O’Keefe had cobbled together 14 wickets from his first four Tests. Now he has 26 from five and presumably heads to the Bengaluru Test this coming Saturday with a rare old spring in his step.

For every bit as brilliant as Australia were in Pune, India were very, very poor. The first thing they got wrong was the wicket, for this was a virtual dustbowl, full of cracks and loose clods of earth being dislodged from the opening exchanges. It is detrimental rather than helpful for India’s chance to play on tracks like this, and after the events of last week we surely won’t see another one like it for a long time.

The pitch held out for about an hour before the first signs of excessive turn and bounce emerged, but when the opportunity came, India did not bowl or catch as well as Australia did when it was their turn in the field.

Of the 40 wickets that fell in the match, 30 went to the spinners, of which all but three were to good-length deliveries. In other words, 67.5% of wickets were off good-length balls bowled by spinners. But, how often were the five spinners in the match finding a good length?

Interestingly, all of them radically improved their lengths in the second innings. But it is notable how poor Jayant Yadav’s control was in the first innings, while the most dramatic improver was Nathan Lyon (56.0%-87.6%, reflected in figures of 1-21 and 4-53 respectively).

This table shows the % of each spinner’s deliveries on a good length, in the first innings (first row) and second innings:

Ashwin
65.1%
78.6%

J Yadav
51.8%
70.5%

Jadeja
63.2%
79.1%

—————

O’Keefe
64.6%
74.4%

Lyon
56.o%
87.6%

Ravi Ashwin and Ravindra Jadeja are skilful enough to regularly bowl well over 70% of deliveries on a good length but it was a clear failing that they didn’t manage that until the second innings.

Incidentally, perhaps the reason why good-length deliveries were so much more effective for the spinners is that batsmen were pretty much looking to play back all the time and read the turn off the wicket. To do so against good-length spin bowling as opposed to back-of-a-length bowling is that much tougher.

Though the game was ebbing away from them by the time India started dropping catches in the second innings (Steve Smith three times, Matt Renshaw once), these failings served to drain the last vestiges of hope for the Virat Kohli’s men.

Australia’s players celebrate the crucial wicket of India’s captain Virat Kohli on the third and final day in India

One thing I want to look at in this match is the issue of luck. Some Indian fans were most insistent that Australia’s batsmen were more fortunate than India’s, and when covering day one for CricViz I did find myself looking up how many times players were missing or edging the ball without being dismissed.

So here is another simple table showing the percentage of balls edged or missed by the batsman in each innings who survived the most balls.

Renshaw
16.0%

Rahul
8.2%

Smith
16.3%

Pujara
8.6%

Perhaps there is some credence then in the theory that the cricketing gods did not look particularly favourably on India. However the old adage of “making your own luck” rings true to some extent. Notably, there were those dropped catches by India already detailed; in addition Yadav contrived to bowl David Warner on the first morning with a massive no-ball and there were plenty of questionable tactical decisions made by Kohli, not least the decision to take the new ball on the first evening against the free-hitting Mitchell Starc.

I would be reasonably confident that India can do the minimum required to come back and win the series now. It won’t be easy. They can do no worse than draw one and win two of the remaining Tests so pitch preparation will be vital. It would clearly be dangerous to replicate Pune again and lose the toss, while any wickets that are too flat and bring the draw strongly into the equation must also be avoided.

Whatever they say in public, Australia will certainly be unusually bullish about their chances. But Steven Smith’s leadership skills will surely be tested if India mount a strong response in Bengaluru this coming weekend. India need to improve sharply in all departments of their game – but they certainly have the capability to do so, and WinViz is sure to start in favour of the the home team on Saturday.

The (data) revolution in fielding

While batting and bowling have had quantifiable numbers associated with them, in order to help judge success or failure, fielding has remained untouched until very recently. Jonathan Liew from The Telegraph looks at how England are using a similar model as CricViz to monitor fielders’ performance.

Phil Oliver runs the stats company CricViz, which provides advanced data analysis to fans and broadcasters using a model Leamon helped develop. He believes that the next step is wearable technology, that would allow player’s movements to be tracked to the nearest pixel. Used in conjunction with a Hawkeye-style ball-tracking tool, it would allow us to settle debates that can now only be conjectured. Was that a genuine chance, or just out of his reach? Could the fielder have got to that ball quicker? And who really is the best fielder in the world? (According to CricViz data, Ben Stokes was the most effective fielder at last year’s World T20 by some distance.)