2020 IPL Mid-Season Analysis

Email: Sportsanalyticsadvantage@gmail.com

If this article has given you insight into the data that Sports Analytics Advantage can offer cricket teams around the world in formulating team strategies, selection, draft or auction plans, or any other work, please feel free to get in touch at sportsanalyticsadvantage@gmail.com.

12th October, 2020.

Following today's match between Royal Challengers Bangalore and Kolkata Knight Riders, we are now at the halfway point of the group stage of the IPL, with all teams having played seven of their 14 group matches.  Considering this, it's an excellent point to look at some relevant stats and insights which we can ascertain from the first half of the tournament so far.


Instead of IPL teams playing seven matches at home and seven matches away, there are three 'neutral' venues for the teams - Abu Dhabi, Dubai and Sharjah.  I have already discussed Sharjah's high-scoring nature in my previous T10 League analysis here, and this dynamic has again been in evidence throughout the IPL so far.

Here we can see that there isn't a great deal between overall batting data at Dubai and Abu Dhabi, but Sharjah is almost off the charts - it's a much higher scoring venue with both a higher runs per wicket and strike rate figure.

The bias towards batting first in the IPL so far has been a real surprise, and given this, it's interesting to look at some of the individual venue dynamics between batting first and second.  The chart below examines the data for the three venues for batting first and second:-

Interestingly, there is much less of a difference in runs per wicket and batting strike rate between 1st and 2nd innings at Abu Dhabi as opposed to Dubai and Sharjah, where there is a marked bias towards batting first.  The next question really should be, why is this the case?

One possible explanation focuses on Powerplay performance.  So far, the average Powerplay score batting first is 46.32 runs for 1.04 runs, while this is considerably worse for teams batting second.  Yes, the 43.61 average runs scored isn't much worse, but the 1.64 wickets lost certainly is.  Teams batting first have lost 2+ Powerplay wickets on eight occasions, while teams batting second have done so 15 times.  Losing wickets in the Powerplay has a considerable knock-on effect on scoring in later phases of a team's innings - I examined this exact discussion in my book - and perhaps chasing teams are going too hard too early.  Either that, or the pitches are playing much worse for the batting teams as the matches progress.

Not only this, but teams batting first have been able to find boundaries much more frequently, hitting boundaries on 18.13% of balls faced, almost 3% higher than teams batting second (15.49%).  Winning the boundary percentage count is imperative in T20 cricket - around 87% of IPL matches were won by the team with the highest boundary percentage between 2017-2019 seasons.

While dot-ball avoidance is less important than boundary-hitting - another topic examined in my book - it's also worth noting that teams batting first so far in this year's IPL have hit 32.14% dots compared to 37.25% dots for teams batting second - so the innings of teams batting second feature around six more dots on average than for innings of teams batting first.

Understanding potential bias towards pace and spin at the three venues is also useful.  The chart below looks at this:-

Here we can see that spin offers a considerable benefit in economy (lower batting strike rates) than pace at all three venues, with between 11-18 runs per 100 balls saving depending on the venue.  However, with the exception of Dubai, there is a trade-off with wicket taking, with spin being pretty unimpressive from a wicket-taking perspective at both Sharjah and Abu Dhabi.  

Across the matches so far, though, 63.43% of balls bowled were via pace bowling, and just 36.57% via spinners.  Should teams be bowling more spin?  It's certainly something to consider - and I'm surprised that more teams aren't structuring up differently, looking at starting XIs with at least three pace and spin options each - allowing flexibility and the opportunity to bowl 10+ overs via each bowling type depending on conditions.


Comparing teams here isn't 100% ideal because teams haven't played the same number of matches at each venue.  For example, Rajasthan Royals have played three matches at high-scoring Sharjah while Chennai Super Kings have only played there once - it's not quite a like for like comparison.  However, it's still useful to work out a pretty accurate guide to team dynamics, which is interesting to examine.

Given that we've already mentioned that a huge driver of team success is winning the boundary percentage count in matches, one way to examine whether teams have performed well or badly is via looking at their net boundary percentage (boundary percentage scored - boundary percentage conceded).  The chart below looks at this (sorted by current league position):-

Mumbai Indians and Delhi Capitals are streets ahead of all other teams when looking at net boundary percentage, and their position at the top of the league table at this stage is absolutely justified.  Both of Rajasthan Royals and Kings XI Punjab have a net boundary percentage of worse than -2%  and this is an unimpressive figure in a level-budget competition - their issues largely stem from bowling boundary concession figures around the 19% mark - while Sunrisers Hyderabad also aren't far from this -2% net boundary percentage either, although with a rather different dynamic which is discussed shortly.  The real outlier is Chennai Super Kings, who have a very marginally negative net boundary percentage despite being second bottom of the table.

This would perhaps indicate that Chennai Super Kings have been poor rotators of the strike this tournament - not necessarily a surprise given the average age of their team - and this was indeed the case when looking at teams batting boundary percentage versus non-boundary strike-rate, which can be seen in the chart below:-

Only Rajasthan Royals and Kolkata Knight Riders have produced a lower non-boundary strike-rate so far in the competition than Chennai Super Kings, while both the two leading teams, Mumbai Indians and Delhi Capitals, again have the most impressive scoring metrics, grouped close to the ideal top right-hand corner.

Much has been made of Sunrisers Hyderabad's approach - discussed by David Warner on occasion - of reducing their dot balls played and this is evidenced by their high non-boundary strike-rate.  However, their team boundary percentage is the worst in the competition by some distance, illustrating a rather different approach taken by Trevor Bayliss as head coach at Sunrisers compared to his generally more attacking methodology at England.

Sunrisers' high stability approach is further shown by the chart below, which looks at team batting balls per dismissal versus batting strike rate:-

They are grouped towards the bottom-right corner with Chennai Super Kings and Royal Challengers Bangalore, while current league table leaders Mumbai Indians and Delhi Capitals are towards the ideal top-right hand corner, summarised by high stability and strong scoring.  With the bat, matters are less positive for Rajasthan Royals and Kolkata Knight Riders at the halfway point of the tournament.


First of all, I want to look at the scoring data for the individual batsmen who have faced a minimum of 100 balls so far.  We can look at their scoring dynamics via assessing their boundary percentage and non-boundary strike-rates, which can be seen in the chart below:-

There are a number of fascinating insights which can be discussed here.  Firstly, AB De Villiers - as is often the case when looking at these metrics - is closest to the ideal top-right hand corner, with a high boundary percentage and non-boundary strike rate.  Marcus Stoinis is hot on his heels, also exhibiting impressive data, and this year's IPL so far has been a real breakthrough tournament for the Australian all-rounder - I'm happy to admit that I've considered him pretty over-rated by many observers previously when looking at my recruitment metrics, but he's proven me wrong so far with several brutal displays with the bat as a finisher.  

Interestingly, only nine players have been able to hit boundaries from at least 20% of balls faced, and one of these players, Shane Watson, has struggled to rotate the strike so far - again giving a nod to the previous discussion on Chennai Super Kings' struggle with non-boundary strike-rate.  On the flip side is Virat Kohli, who has the opposite dynamic.  It may surprise readers to note that Kohli has the worst boundary percentage of any batsman facing 100+ balls so far in this competition, but his non-boundary strike-rate is truly incredible - evidenced by the sheer number of twos which he runs.  Such a low boundary-hitting style isn't one I particularly advocate, but it's a style which seems to be working for him so far given the resurgence in Royal Challengers Bangalore as a force this season.

Moving on to bowlers, a simple chart looking at economy rate versus strike rate (balls per wicket) is useful to ascertain performance levels of bowlers bowling 100+ balls so far in the competition (spinners highlighted in red).  It should be noted that Pat Cummins, due to strike rate, was omitted from this chart:-

The spin trio of Washington Sundar, Axar Patel and Rashid Khan have performed superbly so far in the competition from an economy rate perspective, going for around five runs per over, with Rashid Khan the closest bowler to the ideal bottom left-hand corner.  Those three, plus Yuzvendra Chahal, have produced the best numbers for spinners so far.  

As for pacers, Jofra Archer, Jasprit Bumrah, Trent Boult and Kagiso Rabada have performed extremely well, with Rabada's current strike rate in the tournament of less than 10 balls per wicket simply stunning.



For the final area of assessment, there are several angles which I want to examine - age and salary.  It's interesting to look at the age dynamic, with there being particular discussion currently on the mantra of 'old blokes win stuff'

I'm not a particular buyer of that mantra, so it's certainly interesting to note a marked drop in batting strike rate among older IPL batsmen this year - albeit with a trade-off of a higher batting average:-

Here we can see exactly this, with batting performance looking pretty well-balanced between the 23-26 and 27-30 year old age bracket.  Young batsmen (18-22)  have struggled so far with a poor average, strike rate and boundary percentage, while the age bracket of 31-34 years of age produced mediocre returns.  Interestingly, this general dynamic for players around this age bracket was something that has been evidence generally in the last three years as well, as discussed in my book, and you can read the chapter discussing this here.  Essentially, there appears to be a reasonable argument that IPL teams retain batsmen for one expensive retention too many when they are either at peak age, or slightly past it.  The players aged 35+ tend to be previous elite players and therefore have a very high ceiling to decline from - so can still produce strong performance levels.  This year, there's a very slight tail-off in non-boundary strike-rate as players turn into their thirties but it's not as strong a drop as in previous years.

Performance based on salary is also interesting to assess.  Marquee batsmen so far have performed extremely well from a stability (batting average) point of view, and are excellent rotators of the strike despite fairly mediocre boundary percentages, as can be seen below:-

As discussed already, Virat Kohli is a big contributor to this dynamic but other batsmen such as David Warner and KL Rahul, plus to an extent MS Dhoni and Manish Pandey, also give weight to this.

There's a further interesting dynamic here with players priced in the second tier pricing (500-990 Lakh) having quite balanced returns with high strike rates and fairly acceptable batting averages, and batting strike rate drops from here as the price brackets get lower.  Many of the batsmen priced below 100 lakh are very young or inexperienced and generally this cheaper group of players produced very poor batting returns.  


Old certainly isn't gold when it comes to bowling output.  Older bowlers have really struggled so far in the IPL, as the table below illustrates (it should be noted that 35+ age bracket was not viable to be assessed individually due to only four bowlers fitting into this age bracket currently):-

So far in the tournament, bowling economy rates have worsened as each age bracket advances, which certainly isn't a ringing endorsement for the general level of bowlers around or older than peak age at the current time.  Have younger bowlers emerged stronger than older bowlers following a period of inactivity around lockdown?  It's an interesting debate which could have some credence behind it.  

Finally, from a financial perspective, it's very much worth noting that economy rates worsened as salary bracket groupings reduced:-

Essentially, by purchasing a marquee bowler, teams give themselves a much bigger chance of having strong bowling economy - only occasional bowler Glenn Maxwell has an economy rate of over 9 runs per over currently out of all the bowlers priced 800+ lakh.  Chris Morris, Rashid Khan and Bhuvneshwar Kumar all have economy rates below 7 runs per over at the time of writing from this elite price bracket.

However, from a balls per wicket perspective, the most expensive bowler bracket (24.02 balls per dismissal) had the worst performance, so there's this clear trade-off between strong economy and mediocre wicket-taking.  This could be because these bowlers are generally quite defensive bowlers, or a further explanation may focus on batsmen trying to see these players off without too much damage, before targeting weaker bowlers.  

It will be fascinating to see if many of these dynamics continue throughout the tournament...

If this article has given you insight into the data that Sports Analytics Advantage can offer cricket teams around the world in formulating team strategies, selection, draft or auction plans, or any other work, please feel free to get in touch at sportsanalyticsadvantage@gmail.com.