This is a blog post from
When are T20 games won and lost? Commonly, we have a tendency to focus on things which happen towards the end of a run-chase. The closest thing in the cricket broadcast to a “state of the game” record is the ‘worm’, which compares the progress of each team on an over-by-over basis. The chart presented in this article is a chart of 'match progress' which shows the batting team’s position after each over. The example shown below is the match progress chart for the IPL game between the Mumbai Indians and Royal Challengers Bangalore that was played on April 14, 2017.
The chart evaluates the position of the batting team at the end of each over by comparing it to positions of all other equivalent T20 batting teams at the same point in a match. An equivalent T20 batting team in the first innings is determined by the pattern of wicket-loss at the end of each over. For instance, take two teams with two wickets down by the end of the 10th over. The position of the team would be the number of wickets lost by the team after each of the first 10 overs. So, a team which lost its first wicket in the 9th over and second wicket in its 10th over would not be equivalent to a team which lost its first wicket in the 3rd over and its second in the 8th over, and so on.
In the example in the chart above, after 12 overs, Royal Challengers were 76/1. This is 21 runs worse than the median total after 12 overs for a team which lost its first wicket in its 10th over. The median T20 score for a team with the pattern of wicket-loss found in the Royal Challengers' innings is 171. Consequently, the Royal Challengers ended their innings 29 runs under par.
In the 2nd innings, the position of the team is calculated considering not only the pattern of wicket-loss, but also the position of the team batting first relative to the relevant median in the first innings. For example, after 12 overs of the run chase, Mumbai Indians were 65/5. They needed about 10 runs per over for the last eight overs. But since the Royal Challengers had been 29 runs below par at the corresponding point in their innings, this meant that at 65/5, MI's position was about average for a team which had lost wickets in the way that MI had.
Teams which were five down and needed 10 runs per over for eight overs have won 11% of the time. This is the final piece of information provided in this chart under “WIN %”. Until Kieron Pollard took 19 runs in the 16th over of the run chase, the Mumbai Indians were still under water.
Teams have resources (wickets) to play with, which they use to make progress in a match. This chart attempts to match the rate at which a team uses up a resource with the amount of progress it has made in the match.
A team’s progress in the match is related to two things. First, it is related to the degree of progress other teams which have used up resources in the same way tend to make. Second, it is related, during the run-chase, to how much progress a team needs to make at the end of every over given the target it has been set. Samuel Badree’s hat-trick set the Indians back, but it did not cause them to collapse as the chart shows. On the other hand, the Royal Challengers under-performed significantly with the bat.
Every single delivery in a T20 game is significant. What’s more, every single delivery is equally significant. This type of chart helps to trace the progress of the match keeping this central fact in mind. For instance, it shows that in the game discussed in this article, even the loss of four early wickets did not cause MI's match position to be as far below par as that of RCB's at 76/1 after 12 overs in their innings. After 12 overs MI had an 11% chance of winning (7 out of 60 teams in this position have won). If we assume that MI's 6th-wicket pair would continue to go for the runs (a safe assumption, since the alternative is a defeat), then, maintaining the asking rate of 10 after 13, 14 or 15 overs without losing wickets steadily improves the chasing team's chances (as the chart shows).
Conventionally, all the attention would be on Pollard's innings (because, at first glance, it looks outstanding). The innings was terrific, but it was rendered successful by RCB having scored far fewer runs than they should have given the resources they spent.
In other words, if the chasing team does not lose resources, they will go on to win. Instances in which the risk-taking involved in chasing steep targets results in neither runs nor wickets are rare. But the fact that the asking rate was maintained tells us that the runs were coming. This in turn suggests that after having been so far below par, RCB basically needed to bowl the visitors out in order to win. Now, had RCB produced a par total, that asking rate would have been steeper. The win expectation for a team needing 13 runs per over for the last eight overs with 5 wickets in hand is about 2%.
The match-progress chart presented here provides a picture of the whole match at a glance. It does this by showing the position of each team relative to an average T20 performance, and comparing the ongoing match situation to comparable situations in past matches. A systematic explanation of the same is available here , as are progress charts for every T20 game so far for which over-by-over records are available.
