CricViz Analysis: Mohammed Shami in the Second Innings

Ben Jones analyses the Indian seamer, and his excellent second innings record.

India claimed another emphatic home win in Visakhapatnam today, after twin centuries from Rohit Sharma were too much for South Africa to handle. A wonderful first-innings partnership from Dean Elgar and Quinton de Kock gave the Proteas hope that maybe they would be able to cling on, with the assistance of some rain and good fortune, but those hopes unravelled in the second innings, as they collapsed to 191 all out.

Second innings collapses aren’t a rare thing in India, of course, given the way spin and uneven bounce only becomes more exaggerated as matches wear on. However, in recent times Virat Kohli’s primary weapon in the second innings has not been a spinner, but a seamer. Mohammed Shami has crowned himself King of the Second Innings, a crown he clung to yet more tightly in Vizag, his 5-35 blowing South Africa away.

Of course, Shami is a top class bowler, with no qualification necessary. He was arguably India’s best bowler in England, and in Australia was outshone only by the arrival of a generationally talented Jasprit Bumrah. Since the start of 2018, he averages 24.95 with the ball in Test cricket, an excellent record.

Yet what reveals more about his development as a bowler is the split in averages between first innings and second innings – in the second dig, the cost of his wickets plummets, and his strike rate of 32.1 less than half of its first innings equivalent. 63% of his wickets come in the second innings. That’s when Shami comes alive.

It’s a double-edged sword, in some ways, to have such a bias in favour of the second innings of the match. Since the start of 2018, Shami’s first innings bowling average is the worst in the world for any seamer with 15 wickets. There are very few elite performers who have been as ineffective at the start of the Test.

Yet in that same time period with the same wicket-constraints, Shami’s bowling average in the second innings is the fourth best in the world. Most bowlers see a fall in their average in the second innings, but the drop for Shami is extreme. What is it about his approach to Test bowling which makes him so lethal in the second innings but, relatively, so ineffective in the first?

One explanation could be that Shami targets the stumps more than most Indian seamers. Such a method would allow him to more consistently exploit the uneven or low bounce which tends to come about more regularly as a Test progresses, but earlier in the match when the bounce is true those deliveries are easier to put away. Any unexpected movement off the pitch and Shami’s deliveries are in a better place to exploit it than those from other Indian seamers, but without that movement, he’s sending down half-volleys.

Indeed, that idea holds true amongst a wider range of seamers. Of all the seamers to play at least 10 Tests since the start of last year, Shami targets the stumps more than anyone in the world. Those same principles of being there or thereabouts to exploit unexpected movement apply on a global level – nobody else is hitting that area more often than Shami.

Yet that is probably too broad a generalisation. Pat Cummins, the best second-innings seamer still playing the game given Morne Morkel’s retirement from international cricket, almost never targets the stumps, as we can see from the graphic above. Whilst Australian pitches don’t tend to break up or shoot and spit as much as Indian pitches, it does suggest that targeting the stumps is not a magical elixir for second innings bowling.

The other, perhaps more productive avenue to go down, is the idea that Shami has actually just been rather unlucky in the first innings of matches, and it’s pure random chance that his rewards have come in the second innings.

He has not produced significantly more chances in the second innings than in the first – since the start of 2018, Shami draws a false shot (an edge or a miss) with 22.4% of his deliveries in the first innings, and 23.5% in the second innings. He is forcing marginally more mistakes in the second innings of Tests, but certainly not enough to explain the vast difference in averages.

The actual deliveries he has bowled in either innings haven’t been hugely different in quality either. According to our Expected Wickets model (which uses historical ball-tracking data to calculate the probability of any given delivery taking a wicket), Shami’s Expected Strike Rate since the start of 2018 is 47.3, only marginally different from his Actual Strike Rate of 46.1; overall, his deliveries have roughly got the rewards they deserve.

However, when we look at the split for his Expected Strike Rate in each innings, we see that this isn’t the case. In the first innings, he’s been mightily unlucky, and in the second innings, that luck has come back around. Bowling deliveries of almost identical quality, Shami’s actual returns on the scorecard have fluctuated massively.

Ultimately, the most important aspect of this Indian attack – arguably the best attack in the world – is that it is so well rounded, and in possession of such depth, that a player can develop this clear specialism without it being seen to weaken the team in other areas. In a struggling team, Shami’s bias towards the second innings might not be as easy to manage, and could easily be interpreted as weakness in the first innings – in a winning team, he can just take a backseat for the first half of the game before springing into life later on. India, with this varied threat, are brutal to play against.

Ben Jones is an analyst at CricViz.

(Visited 559 times, 1 visits today)
3 replies
  1. katy
    katy says:

    Write the conclusion.

    The conclusion brings closure of the topic and sums up your overall ideas while providing a final perspective on your topic. Your conclusion should consist of three to five strong sentences. Simply review your main points and provide reinforcement of your thesis.


Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *