Here's An Idea: On Smarter Stats and Better Decision-Making
All statistics are equal, but some statistics are more equal than others.
On February 13, Jarrod Kimber set out to investigate the possible reasons why Dom Bess had been dropped by England, especially after his incredible run in Sri Lanka and in England's first Test against India. Bess had taken a total of 17 wickets. He had taken 17 wickets for 22 runs in his last three Tests. No mean feat. But it gets better, Virat Kohli, Ajinkya Rahane, Cheteshwar Pujara were among his wickets. Not too shabby for a young spinner.
So why was he dropped?
Despite taking a bucket load of wickets, some of them very important ones, Bess was struggling with his line. Even at crucial times when he was taking wickets, his line was inconsistent and he bowled far too many full tosses. Therefore, he did not bowl very well but got lucky lots of times. It happens a lot in sports; luck changes things so much that if we only focus on the results, we miss significant bits of information that can help in the making of important decisions in the future.
(Without a shadow of a doubt, Bess is a good bowler, a talented finger spinner. He bowls an excellent line, he can spin the ball and he can also get good bounce. All good things. But, he is not yet fully developed as an international cricketer, though he has all the skill and potential to be a great.)
Smart Stats
Focusing on the results makes us fall prey to outcome bias. Psychologists define outcome bias as an error often made in evaluating the quality of a decision when the outcome of that decision is already known. Simply speaking, it is when one tries to work backwards from the answer. One is prone to finding patterns where there are none just so they fit the result and make it make sense.
The general belief among most sports fans is that good processes yield good results and bad processes yield bad results. Therefore, the winning team is considered to have had a better overall performance, while the losing team is viewed differently.
One way to avoid this is by employing the use of smart statistics.
The most common ways, and probably easiest, through which we evaluate players after a tournament or series is by simply identifying the top run-getters and wicket-takers. Often, we use these blanket metrics to make calls of who should be on the list to make the flight on a Proteas ticket and who should not.
This is an ongoing conversation, especially right now, with the T20 World Cup around the corner. As things stand, lots of people are already making lists of top performers, earmarking them for the national team.
But, if our lists are with the World Cup in mind, the statistics need to be refined further. Maybe the top run-getter showed a weakness in a crucial area. Maybe the top wicket-taker had a very high economy. All these are important because, for example, a bowler that takes a wicket in an over that goes for 20 runs (6,4,6,4) might not be the best bet. On the other hand, the batter who played four balls and scored 20 runs might be a T20 asset.
In the same breath, a bowler who manages to bowl a decent number of dot balls in T20, even without taking many wickets, can be an asset. These are the small margins that determine victory in T20.
Therefore, our statistics need to be smarter, they need to be more nuanced. Instead of just speaking of runs, they should breakdown the runs into phases of the game, type of delivery faced, boundary runs, sixes and so on. Bowlers should not just be judged by the number of wickets, but also subtle contributions like dot balls and the percentage of deliveries that did not reach the boundary.
Smarter stats lead to smarter decision-making, and that leads to a competitive advantage.
The Breakdown: T20 Challenge Top Performers
(Image Courtesy of 12thMan Analytics)
If the question is, which bowlers had the greatest impact, instead of just focusing on the wicket-takers, it will be prudent to consider contributions from bowlers like Keshav Maharaj, Chris Morris and Lizaad Williams. They offered "intangible value." In this case, intangible value is value accrued from less obvious things like wickets or runs.
The trio's overall dot ball percentage was the highest throughout the tournament. According to a breakdown by Shaun Rheeder an analyst with 12th-Man Analytics Keshav Maharaj, Chris Morris and Lizaad Williams had the highest dot ball percentage 57.64%, 48.08% and 46.94 respectively.
In T20, like in other formats, strike rotation matters. At the end of the day, cricket is still cricket, and too many dot balls often lead to wickets. However, in T20s singles are an inferior currency when trying to chase a total or set a competitive one. A run rate that hovers around six or seven runs per over is not very helpful unless the team is chasing a very low total. Among the bowlers, Keshav Maharaj, Imraan Manack and Aaron Phangiso had the lowest boundaries hit off their bowling. With figures of 6.94%, 7.53% and 7.92% respectively.
Spin has been the Proteas' Achilles heel for a long time. It is no secret, bring on spin and the Proteas struggle. By holding this tournament at Kingsmead, CSA did the best they could to get players and teams onto a pitch that has a resemblance (however minor) to something the Proteas might encounter in India in October. It was a masterstroke.
Now, the easy metric to look at would be how many times a player got out to spin. However, what that metric does not look at is whether the player was just taking singles off-spinners and watching the action from the other end. Waiting for the pace to cash in. George Linde, Raynaard van Tonder and David Miller (171.43, 154.72 and 154.55) had the highest Strike Rate against spin.
According to an ESPNCricinfo report, during last year's IPL, teams scored an average of 11.07 runs an over. Another report suggested that BBL teams average 11.38runs at the death. What this means is that for a team to be competitive, they need exceptional bowling in the death overs. The lower the economy the better. In this tournament Lungi Ngidi, Anrich Nortje and Marco Jansen showed skill in the death (4.80, 6.32 and 6.56)
On the other side of this is that the batters who can put bowlers to the sword at the death. The top three strike rates were from George Linde, Migael Pretorius and Heinrich Klaasen (222.22, 192.86 and 192.00)
The above are some of the critical metrics that need to be considered. Going beyond the regular ones that focus on most runs and most wickets. At the end of the day, sometimes those most runs come with a low boundary percentage of those most wickets come at a high boundary percentage. The cost of those runs and wickets needs to be considered. After all, teams can only ride their luck for so long.
That said, smart statistics need to be balanced with other factors. They are one part of a number of things that come into play when picking players.
If you enjoy the content here and on Chadwick Drive, there are a couple of ways you can show your appreciation.
You can Buy Me A Coffee. You can buy several coffees if you like (simply change the number of coffees to your preferred amount). All coffees you buy will be greatly appreciated.
Alternatively, please encourage anyone you think may be interested to subscribe to this newsletter, the blog or both