Authors: Vighnesh Subramanian
Consider this interesting statistic: Between 2013-19, Chennai Super Kings (CSK) has won a mere 35% of matches against Mumbai Indians (MI). In that same period, CSK’s win record against all other teams in IPL has been 50% or above. What is it that MI did differently?
Simple. MI studied data, identified the Achilles Heel of their opponent, and aimed their arrows at it. Oftentimes, the heel might not even be so vulnerable in isolation, but creative use of data analysis can present answers. CSK has a right-hand-heavy batting line-up, so MI decided to stack up their spin attack. Result – Ball after ball turns infuriatingly away from the right-hander, pressure is built, and there is no risk of a left-hander.
In his September 2020 The Cricket Nerd newsletter on how MI has been successfully using match-ups to ace the T20 game, Tim Wigmore writes, ‘Left-armer Krunal Pandya opened in the last three games against CSK, getting openers Shane Watson and Faf du Plessis out once apiece. In the middle overs, Rahul Chahar bowled immaculate leg breaks: in four games against CSK last season, he conceded just 4.28 an over. Revealingly, left-arm spinner Anukul Roy’s sole game for MI last season was in Chennai against CSK, providing a third option of turning the ball away from right-handers.’
What are match-ups?
Since the advent of data in sports, the term ‘match-ups’ has found its way into the layperson’s dictionary. Unlike advanced analytics that remains shrouded in mystery and purely in the realm of stats gurus, ‘match-ups’ is a basic enough concept for your everyday cricket connoisseur to understand and spout. To put it simply, it’s a 1 vs 1 comparison in a team sport to understand how one player fares against another.
Let’s say you are watching Kolkata Knight Riders (KKR) take on Sunrisers Hyderabad (SRH). KKR wins the toss, decides to bat, and blazes off to a brilliant start. You know that unless a few wickets tumble, the Sun ain’t gonna rise for Hyderabad. Yet, you see Warner holding on to his trump card Rashid Khan. You’re watching the match with your 60-year-old uncle who has his own commentary track to compete against the one playing on TV. “Useless Captaincy,” your uncle exclaims in exasperation. Traditional wisdom says wickets are the best form of containing your opponent. Any captain worth his salt should have sent his star spinner in to gobble up some wickets. Yet here is a captain doing just the opposite.
Why? What is the worst that could happen?
Simple. Rashid comes on, fails to capture wickets, and eventually doesn’t have enough overs in the bank to bowl to Andre Russell. Now that would put things into quite a conundrum, for we all witnessed the magic Andre Russell can weave with his bat (like he did against CSK).
This is where match-ups come into play. Let’s take a look at some of Russell’s stats in the below table.
When you see that you are up against players like Russell who can win you games by the scruff of the neck, you try to save your match-up bowlers to limit the damage and win you the game.
If you look at the table, Russell with a strike rate of 120, the opposition would love to take that down at any time of the year. If we look at IPL alone, in a span of 13 balls Rashid has got Russel out twice. Worth the gamble, if you want to contain runs.
No Rocket Science Here, Just Basic Cricket Science
Match-ups are not advanced analytics. Advanced analytics contains other things like ball tracking. Match-ups are all about looking at the right numbers and zooming in on the right combination. This is where sports analysts come in, to identify these crucial combinations and suggest the best use of the findings to the players. Prasanna Raman, fondly known as Pdogg, is one of the finest exponents of this. His ability to accurately identify these combinations has given more than a competitive edge to the South African team.
Another classic example of these match-ups can be seen in the 2016 ICC World Twenty20 series when Carlos Brathwaite smashed four sixes off Ben Stokes. In the whole of this series, England’s Joe Root bowled only two overs. One was against South Africa, where he went without a wicket. And another was in the finals against West Indies where he got both the openers Chris Gayle and Johnson Charles. But he didn’t get to bowl again. No doubt the bowler bowled well but isn’t it surprising to see a scorecard that read 1-0-9-2!?
Root bowled the second over of the game. First ball wicket of Charles, who tried to slog him and only managed to get a top edge. During the second ball, Gayle showed no respect for Root as the first ball was a smashing four. By the third ball, Gayle, who was booming with confidence, as usual, tried to smash one more but he was caught out. In a span of 3 balls, Root gets both the left-handed batsmen. Again, it’s not rocket science, that left-handers struggle against off-spin, but bringing on a part-timer against an explosive duo? Not something from the cricket books, but definitely a lesson from the data books.
Dhoni Does It Too
No twenty-twenty match-up or captaincy discussion is ever complete without a mention of MS Dhoni. Thanks to copious amounts of data, we know that Virat Kohli is weakest against left-arm pace, averaging at just 29.07 (way below his usual averages). Also, Kohli’s performance against Deepak Chahar and Imran Tahir is exceptional. That leaves Dhoni with just two options – Ravindra Jadeja and Sam Curran. Dhoni plays Curran against Kohli, and it pays off! True, Dhoni is a cricketing computer unto himself, but these basic inputs do go a long way in understanding what works and what doesn’t.
Match-ups are well and good but relying solely on data is not wise either. Like Pdogg said in an interview, “If I say I am 10 out 10 (on a scale of success) then I am lying to you. I can say that I am doing a good job if I am 6.5 out of 10.” Sometimes if the batsman/bowler is having a great day, you just have to sit back, admire and let the star of the show take it forward.
Tags: Analytics for Sports Cricket and Analytics