Muttiah Muralitharan successfully appeald for Sachin Tendulkar's wicket

Lift a finger, ump: in 91 Tests assessed, 23% of Muttiah Muralitharan's appeals were ruled in his favour

© AFP
15

Stats feature

Good shout, bad shout

The numbers behind the appeals

Charles Davis |

The appeal is one of cricket's most familiar and time-honoured rituals. It can be a solo or chorus; captains encourage the latter in the hope of tipping the balance of an umpire's judgement. Although in theory an umpire must always make the final decision on every dismissal, and can only do so after an appeal, in practice the great majority of appeals are restricted to certain modes of dismissal.

This article is a short study of the statistics of appeals. This is not an area that has been studied in the past, but it is easier now with detailed ball-by-ball records available, such as texts in ESPNcricinfo's archive. The study is limited to Test matches this century.

The method used involved searching all Test-match texts for mention of appeals, and filtering them into a separate file. The search went beyond the word "appeal", including terms like "shout" and any mentions of umpires' decisions. All lbw, caught-behind and caught-at-short-leg decisions were included, as were stumpings and run-outs, and appeals for other types of dismissals where they could be found. More than 25,000 appeals were catalogued from almost 700 Tests.

On average, about 44 appeals were found per Test. About 13 of these were successful. This works out to about 2.3* unsuccessful appeals for every successful appeal, or a success rate close to 30%.

About 77% of all appeals were for lbw, and 13% for caught behind. The rest were a mix of other dismissal types and catches taken elsewhere. It was a little surprising to find that only 3% of appeals were for catches at short leg, as these can be some of the most contentious. There were some appeals for multiple modes of dismissal, and occasionally one encountered an appeal that baffled the commentators. There were a couple of appeals for obstructing the field, unsuccessful of course, since no one has been dismissed in this fashion since 1951. The bowlers were Ricky Ponting and Virat Kohli, oddly enough.

The confidence man: Shane Warne had the persuasive powers to make the umpire agree with him

The confidence man: Shane Warne had the persuasive powers to make the umpire agree with him © Fairfax Media/Getty Images

Success rates for appeals varied according to the mode of dismissal. Only 19% of lbw appeals got the nod, but 61% of caught-behind appeals were successful, with many of the unsuccessful appeals for caught behind described as "half-hearted". About 60% of stumping appeals were given out. The low success of lbw appeals can be put down to optimistic bowlers and the complexity of the law: even when bowling teams are confident enough to review under the DRS system, only 20% of "not out" lbws were overturned.

I should mention one puzzle in the data: the number of appeals logged has dropped 10 to 20% since the start of 2010. It seems unlikely that the searches are missing mentions of appeals in recent years, since the detail in the texts is greater than ever. There has also been no decline in the proportion of wickets falling to spinners versus pace. Perhaps DRS has curtailed appealing to a certain extent; however, India, without the DRS, has seen a similar decline in the number of appeals.

It may be that the DRS has made players more aware of what constitutes a likely dismissal, but I can only speculate.

The 2001 Auckland Test between New Zealand and Pakistan contained a record 108 appeals

The 2001 Auckland Test between New Zealand and Pakistan contained a record 108 appeals © Getty Images

If you have the impression that spin bowlers are more prone to making appeals than pace bowlers, then this is supported by the data. The following table lists the breakdown according to bowling type.

Success rate of appeals according to bowling style
Bowling style Appeals per wicket Success%
Wristspin 1.36 22
Fingerspin 1.29 23
Medium 0.96 29
Fast-medium 0.91 32
Fast 0.77 35

Spin bowlers get the umpire's nod from only 17% of lbw appeals, whereas medium and pace bowlers get 21%. The success of lbw appeals is much the same for left-hand and right-hand batsmen (18.8% and 19.4 % respectively). There is similarly little difference for left- and right-arm bowlers (19.6% and 18.6% respectively).

The success rate of teams ranges from 24% (Bangladesh) to 35% (Australia). The important factors here tend to be the number of matches a team wins and the mix of spin- and pace-bowling wickets. India has an emphasis on spin, and as a result has a rather low return on appeals.

Team-wise success rate of appeals
Team Appeals per wicket Success%
Australia 0.79 35
South Africa 0.72 34
West Indies 0.86 30
New Zealand 0.90 30
Zimbabwe 0.98 27
England 1.05 27
Sri Lanka 1.19 27
India 1.32 25
Pakistan 1.38 24
Bangladesh 1.28 24

When it comes to individual bowlers, the most "appealing" are almost always spinners. The bowlers with the most appeals are listed in the table below. Note that for some bowlers, the assessment period does not cover the complete careers, since matches before 2000 are not included. Also worth noting: bowlers do not always take part in appeals, although they usually do.

Bowlers with the most appeals 2000-2016
Bowler Matches assessed Total appeals Appeals per wicket Success%
Muttiah Muralitharan 91 1110 1.38 23
Anil Kumble 83 898 1.71 21
Harbhajan Singh 100 776 1.42 23
Danish Kaneria 61 615 1.92 17
Shane Warne 77 608 1.10 27
James Anderson 119 578 0.80 34
Zaheer Khan 92 575 1.36 25
Rangana Herath 73 556 1.15 29
Daniel Vettori 97 525 1.22 26
Chaminda Vaas 85 483 1.17 31
Stuart Broad 98 473 0.91 31

Minimum 100 appeals since 2000. "Appeals/wkt" takes into account all wickets taken, including those not requiring appeals.

The most unsuccessful appeals by one bowler in a match: 29 by Murali against Pakistan in the final of the Asian Test Championship in Lahore in 2002. Murali took eight wickets in the match. He also made 29 failed appeals in Christchurch in 2006, where he took seven wickets. In fact, out of the six cases of more than 25 appeals in a match, five are by Murali. Undoubtedly his high strike rate is a factor; it could be argued that Murali made more appeals than other spinners because he had reason to.

Has DRS changed the way players approach appealing?

Has DRS changed the way players approach appealing? © AFP

I found it interesting that Shane Warne had a higher success rate than many other spinners. I always thought Warne was a very canny appealer. He had the ability to sense which decisions might be difficult for an umpire, or where an umpire might be prone to errors, and he saved his most intense appeals for these occasions. His appeals when the outcome was obvious, or his frivolous appeals, were rather less vocal.

At the higher end of success rates, we see only pace bowlers.

Bowlers with the highest rates of success in appeals
Bowler Matches assessed Total appeals Appeals per wicket Success%
Glenn McGrath 79 396 0.57 46
Dale Steyn 84 361 0.47 46
Mitchell Starc 28 116 0.55 45
Mitchell Johnson 73 312 0.58 43
Jason Gillespie 61 275 0.71 41
Brett Lee 76 333 0.62 41
Peter Siddle 61 218 0.62 41
Ben Hilfenhaus 27 104 0.61 40
Fidel Edwards 55 179 0.64 40
Jacques Kallis 142 234 0.55 39
Shane Watson 59 105 0.85 39
Vernon Philander 34 161 0.76 39

It is striking that the bowlers with the highest success rates - also among the most selective appealers on the list - are two of the greatest: Glenn McGrath and Dale Steyn, with 46%. The spin bowlers with the highest success rates are Pragyan Ojha on 35%, and R Ashwin on 34%. At the low end of the scale of successful appealers, spinners dominate.

Bowlers with the lowest rates of success in appeals
Bowler Matches assessed Total appeals Appeals per wicket Success%
Ashley Giles 53 350 2.01 14
Monty Panesar 50 429 2.11 16
Saqlain Mushtaq 32 351 1.99 16
Zulfiqar Babar 13 102 1.57 17
Danish Kaneria 61 615 1.92 17
Ravindra Jadeja 19 140 1.60 18
Enamul Haque 15 130 1.80 18
Nicky Boje 43 167 1.27 20
Chris Gayle 103 108 1.16 20
Anil Kumble 83 898 1.71 21
Ray Price 22 115 1.12 21
Sanath Jayasuriya 72 143 1.40 21

All of the bowlers in the table above are spinners, and those with lower strike rates tend to have lower success with appeals. That is, they appeal just as often but don't get the wickets. Those who have watched Giles and Panesar in action won't be too surprised to see them on top of this table. Saqlain Mushtaq (third on the list) was an inveterate appealer, and the leading appealer in the match with most appeals, New Zealand v Pakistan in Auckland in 2001 (108 appeals). The pace bowlers with the lowest success rates are Mohammad Sami and Ryan Sidebottom, on 22%, with Zaheer Khan on 25%.

Rave reviews: Umpire Bruce Oxenford (raising finger) has had the lowest proportion of referred decisions overturned

Rave reviews: Umpire Bruce Oxenford (raising finger) has had the lowest proportion of referred decisions overturned © Getty Images

The other participants in appeals are, of course, the umpires. For more recent Tests, it is possible to calculate the likelihood of decisions for all long-serving umpires, as shown in the following table.

Umpire decision rates 2011-16
Umpire Lbw% Caught behind%
Aleem Dar 14 59
Billy Bowden 16 67
Bruce Oxenford 16 52
Kumar Dharmasena 16 65
Ian Gould 19 58
Marais Erasmus 13 57
Nigel Llong 18 50
Paul Reiffel 17 46
Richard Kettleborough 13 61
Rod Tucker 14 52
Richard Illingworth** 14 51
Steve Davis 21 63

Percentages of appeals given out since June 2011. Minimum 200 appeals

These figures could be affected by "overturn" decisions in the DRS system, which have not been included in calculations at this stage. However, overturns cut both ways, so the figures may not be affected substantially. Most appeals, of course, do not lead to challenges and reviews; but in those that are reviewed - the more contentious ones - about 28% of the decisions are overturned in Tests. The umpire with the lowest proportion of overturns is Bruce Oxenford, with 18%.

There is more that could be said about the DRS, but perhaps that is a subject for another time.

02:55:12 GMT, December 14, 2016: The values in the third column in the table "Bowlers with the lowest rates of success in appeals" were wrong and have been corrected

*This figure was corrected from 2.9 in the original

** A previous version of this article referenced a different Illingworth

Charles Davis is a Melbourne-based statistician and author who has developed a large ball-by-ball database for Test matches. He blogs here

 

RELATED ARTICLES

 

LOGIN TO POST YOUR COMMENTS

  • POSTED BY Jim on | December 17, 2016, 11:50 GMT

    The stats provide the numbers. But what of the interpretation? The "Team-wise success rate of appeals" shows that the 4 sub-continent teams appeal less successfully, or more optimistically perhaps, than the others. What does this tell us? We only know that the least-successful appealers are spinners (by type) and sub-continent bowlers (by nation). The author assumes that the causal factor is the number of spinners, because the sub-contiment teams have more spinners. But the data don't tell us this. The causal factor could be that sub-continent players appeal more aggressively, which then makes the spinning stats look worse. We don't actually know which is the cause and which is the consequence.

  • POSTED BY Kashi on | December 15, 2016, 14:19 GMT

    Very Interesting Article indeed. Lots of analytical work with not much data. This raises an interesting topic. While catches, runs, wickets and all kind of statistics are reported, the no. of appeals and type of appeals should also be recorded going forward. Perhaps even for umpires. And all thee can be reported along with detailed score card.

  • POSTED BY Shakeel on | December 15, 2016, 9:26 GMT

    Man. you did a lot of work. thanks

  • POSTED BY COLIN on | December 15, 2016, 7:19 GMT

    Unfortunately SOUTHIESAAR's hypothesis that if you are white you will have a better success rate on appeals is too laughable to be worthy of serious comment other than to say that despite being the King of Spain the bowler with the worst record is most definitely caucasian.

  • POSTED BY Gaurav on | December 15, 2016, 4:01 GMT

    Very Nice and unique analysis!! However, it seems to me that there is some error in the table titled "Bowlers with the most appeals 2000-2016". For example, as per the table, James Anderson has made 578 appeals with 0.80 appeals per wicket. According to this statistics, Anderson has taken 722 wickets which is obviously not the case. Have I missed something very obvious here OR there is error in the analysis?

  • POSTED BY Bala on | December 14, 2016, 19:26 GMT

    "Those with lower strike rates tend to have lower success with appeals."

    Does this imply that the more successful bowlers are also more successful with appeals? That becomes self-perpetuating.

    Can the author share a correlation between the bowler's strike rate and appeal success rate? If there is a correlation, we would have unearthed a point to include in umpire training courses.

  • POSTED BY Vilayanur on | December 14, 2016, 13:59 GMT

    One glaring omission - race. Most fast bowlers are fromAustralia, South Africa, NZ and England, with some from WEst Indies and Pakistan. Most umpires are white.

    Fast bowlers (read white) get a greater percent of their appeals upheld. Spin bowlers (read people of color) get a lower percent of their appeals upheld.

    In the Table that showed highest success rate, EVERYONE except Philander is white. (Philander was at the ottom of this table). An overwhelming number was for Australia (who always appear to have an advantage with the umpires).

    In the Table with lowest success rate, ALL are people of color.

    Among spinners, Shane Warne has the second highest success rate.

    The issue of race and ithe role it plays on umpires decision needs to be studied. This article is a great resource that provides stats that speak to the truth.

    Please publish. Thanks.

  • POSTED BY superior frontal on | December 14, 2016, 13:56 GMT

    Very novel and nice analysis. One thing that may be interesting is to find %s by the type of appeal. For example when fast bowlers appeal for LBW, is their success rate higher than LBW in spinners' appeals?

  • POSTED BY DAVID on | December 14, 2016, 11:49 GMT

    Good to see old Ray Illingworth out in the middle again, in umpire's coat (refer final table. Pretty good effort for an 84-year-old.

  • POSTED BY Bala on | December 14, 2016, 2:12 GMT

    " This works out to about 2.9 unsuccessful appeals for every successful appeal, or a success rate close to 30%. "

    The success rate is 1 in 3.9 or close to 25%.

  • POSTED BY X on | December 13, 2016, 23:01 GMT

    The column "Total Appeals" in the table titled "Bowlers with the lowest rates of success in appeals" needs to be revised.

  • POSTED BY shahzaibq on | December 13, 2016, 18:14 GMT

    How about a similar analysis for the same bowlers who have taken at least 50 wickets both before AND after the introduction of DRS? I suspect their appeal/success ratios might have gone up, with ball-tracking coming into play. And one for umpires who have stood in both, 20 matches with DRS and 20 without DRS. I suspect that umpires who have used DRS might be more prone to giving marginal decisions even in subsequent games without DRS.

  • POSTED BY sahil on | December 13, 2016, 18:07 GMT

    For sure, its a unique article and one which involves a lot of effort. What's missed in this is IMHO is the geography involved. All the bowlers with high successful appeal percentages are from Aus/Sa. I guess the reason is in Aus/Sa there are a lot of dismissals caught at slips and Nicks carrying to kepper hence no need of appeals. However in sub continent, where most bowlers including spinners and seamers have a low success ratio, the mode of dismissals are more LBW and caught off spinner by close in fielders which means there has to be an appeal always!!

  • POSTED BY Soumya Das Gupta on | December 13, 2016, 15:53 GMT

    This is such an unique article. And real analytics

  • POSTED BY xxxxx on | December 13, 2016, 14:06 GMT

    It's fine for the author to go to all the effort if that's what he enjoys, but I suspect successful appealers are those who appeal when they actually believe it is out.