The risk-and-reward equation, and why it works for England's Bazball

When Zak Crawley reached his hundred during a stunning onslaught on Australia's bowlers in the fourth Ashes Test at Old Trafford, this is how ESPNcricinfo's ball-by-ball commentators described the ball off which he reached the milestone:

In the end, the Manchester weather denied them a real shot at the Ashes, but that innings from Crawley was the very epitome of everything that defines England's approach under Bazball. And while 93 balls is an eye-popping number, the other stat in those couple of lines of commentary is just as revealing: when Crawley reached his century, he had a control percentage of 64. Let's talk a little more about that number.

For every ball that a batter faces, ESPNcricinfo's ball-by-ball scoring team records a binary control metric. This metric records how convincingly a batter played each delivery: all deliveries left alone (without resulting in a dismissal) or middled are marked in control; and those where the batter was beaten, got an edge or mistimed a shot are marked not in control. These can also be referred to as false shots played by the batter.

A caveat before proceeding any further: all not-in-control deliveries do not have equal wicket-taking potential. A delivery that just misses the outside edge by a whisker is probably more threatening than a ball which induces a mistimed cover-drive, with the ball dribbling harmlessly to the leg side. However, for the purpose of this exercise, both are marked as not-in-control. Over a long innings or a set of innings, this metric gives a fair indication of how assured the batter was at the crease.

Crawley's 64% control at the time of reaching his hundred meant that he mistimed, edged or was beaten 34 times out of the 93 balls he had faced at that point. By the time his innings eventually ended on 189 off 182 balls, his control percentage had improved to 70.88, which means he played 53 false shots. Though the control percentage improved, it was still the sixth-lowest among all centuries since 2013. Given that, on average, a top-order batter plays around 11.1 false shots per dismissal in the last five years, it's obvious that Crawley had the rub of the green going his way in that innings.

However, it is expected that a batter will play more than 11 false shots in a long innings (over a 200-ball innings, even 90% control means 20 false shots). What's even more interesting is the control numbers for England's batting unit since they've adopted this new approach.

Usually, Tests are won by the team which exhibits more control with the bat. In other words, they are won by the bowling team which consistently asks more questions of the opposition batters. Of the 173 Tests which have produced decisive results in the last five years (since August 2018), 123 (71.1%) have gone in favour of the team whose batters had the higher control percentage in the match.

That 50 Tests went the other way is illustrative of the quirky nature of the game - a batter could play flawlessly for his first 35 deliveries, but an error off his 36th could result in his dismissal, for a score of, say, 20, and a control percentage of 97.2. An opposition batter in the same game might achieve a control percentage of only 80, but could end up with a century. Extend the corresponding logic to most of the batters of each of those teams, and you could end up with the winning team scoring more runs but having a lower control percentage. Or a team could be going for quick runs aiming for a third-innings declaration, which could result in more false shots.

Even with those possibilities, though, in more than 70% of the games which had a result, the team with the higher control ended up as the winner, which suggests a reasonably strong correlation between those two factors.

That's where England's numbers since Bazball become interesting. They have achieved an enviable 13-4 win-loss record during this period (with one draw), but only in five of those 18 matches have they had a higher control percentage with the bat than their opponent. (A rider at this point, though: the sample sizes are still relatively small, as it's only a little more than a year since Brendon McCullum and Ben Stokes took charge as coach and captain respectively.)

That, in turn, means England, unlike other teams, have been winning a lot of Tests even when their batters have returned lower control percentages than their opponents. In 13 such Tests, they have won eight, lost four, and drawn one. Compare that with the results for all the other teams when they have returned lower control percentages: five wins, 28 defeats. England's win-loss ratio in such matches: 2.00; the win-loss ratio for all other teams: 0.179. That's a factor of 11.2. Perhaps no other metric illustrates more effectively just how different England's approach to Test cricket has been when compared to other teams in the last 14 months.

For comparison, in the period between January 2018 and May 2022, England had a 10-14 win-loss record in Tests when they had a lower control percentage than their opponents. That was still better than the overall 37-91 record for all teams in this period, but nowhere near the winning ratio they have racked up over the last year.

How is it that England have been winning despite lower control numbers?

The key to answer that question is to not just look at the balls when their batters played false shots, but also at the outcomes when they were in control. England's ultra-aggressive approach ensures that they try to maximise the runs they can score off deliveries that they are on top of: their strike rate off in-control deliveries is a whopping 84.16, compared to 54.71 by the opposition. (Their strike rates when playing false shots is also better than all other teams, but we'll come to that a little later.) This huge difference in strike rates compensates for the slightly lower control percentage - 78.59 to 82.29 - in these 18 Tests. In other words, the runs they score when in charge makes up for the excess risk they seemingly take in playing that brand of cricket.

That is best illustrated by looking at the total runs scored (off all deliveries faced, including the in-control balls) per false shot played. To calculate this, we divide the total bat runs scored by a team off all deliveries, by the total number of false shots (or not-in-control balls). So, if a team scores 300 bat runs in an innings and plays 100 false shots, their runs scored per false shot is three.

In these 18 Tests, that figure for England is 3.56; for their opponents, it is 2.94. Given that the ultimate aim in all cricket matches - even Tests - is to score more runs than the opposition, this shows England are actually managing their risk better than the opposition by getting more value per false shot. Coincidentally, in Crawley's 189, his runs per false shot was 3.57 (189 runs, 53 false shots), which almost exactly matches England's number in their last 18 Tests. Talk about following the Bazball template to the T!

Because of this approach, even when their batters commit errors, they mostly do so when attempting to score runs. Only 25.4% of their total false shots have come when playing defensively; the rest have come about when trying to look for runs. That is a much lower percentage than for most other teams: Sri Lanka and New Zealand are within ten percentage points, but for most of the other top teams, this percentage is over 40, which indicates a larger chunk of errors happen when not looking for runs.

It is not surprising, therefore, that England's batting strike rate of 46.79 for not-in-control deliveries since the beginning of June 2022 is the highest among all teams during this period. Australia are third at 39.64, about 15% lower than England, while India (36.52), Pakistan (35.72) and South Africa (35.49) are all in the mid-30s.

The other aspect of attacking batting is the effect it has on the opposition think-tank and their strategies. Unless the opposition have plenty of runs to play with or the conditions are extremely bowler-friendly, a flurry of boundaries forces the field to spread out, leaving fewer fielders in catching positions to snaffle the genuine edges.

The number of false shots played per dismissal is a fair indication of how many errors batters get away with, due to various reasons. In the period since June 2022, England have lost a wicket every 10.95 false shots, which is second only to New Zealand's 11.21, and higher than the corresponding number for Australia (10.73), India (9.97), South Africa (9.23) and Pakistan (8.47). In Ashes 2023, England lost a wicket every 11.5 false shots and scored 3.1 runs per false shot, compared to 10.67 and 2.87 for Australia. The combination of a relatively high runs per false shot and false shots per dismissal means England score more runs per dismissal, which almost always is the formula to winning Test matches.

All of this suggests that while England's approach looks risky with higher false-shots numbers - mostly attempting non-defensive shots - than you'd normally associate with a winning team, they make those risks work in their favour. For any other team, these control percentages would probably not lead to consistently winning matches, but it works for England, because of their drastically different approach to batting.

It's a fine balance, and one that requires a sensible approach to aggression, but so far England have managed to find that balance more often than not. The 13-4 win-loss is a strong testament to that.