Does the follow-on work in its current form?
Despite gaining a 309-run first-innings lead, Pakistan did not enforce the follow-on against Australia in Abu Dhabi recently. When captains choose not to enforce the follow-on, they are considered to be conservative. However, the follow-on is a peculiar idea which does not lend itself easily to such tags. Batting in the third innings with a huge lead is perhaps not conservative because the batting team is willing to burn resources (ten second-innings wickets) chasing quick runs to set a fourth-innings target. By enforcing the follow-on, a team is taking advantage of a rule that allows it to conserve these resources for later use.
But why is the follow-on rule necessary? Is 200 runs a good threshold, given changing scoring rates (to take just one change)? Before considering these questions, here are some figures about follow-ons.
The follow-on has been enforced 302 times in 2139 Tests from 1877 to 2014. The team enforcing the follow-on has won 230, lost three and drawn 69 Tests. The follow-on was enforced on 40 occasions though the lead was less than 200. The rules for the follow-on have varied, just as the length of the over and the length of Tests have varied over the history of the longest form of the game. Twenty-seven of those 40 Tests were won by the side enforcing the follow-on. Thirteen ended in a draw.
The rate of enforcement of the follow-on has dropped in the 21st century compared to that in the 20th century after World War II, though the opportunities to enforce the follow-on have remained available at about the same rate.
|Span||Follow-on enforced||Follow-on not enforced||Enforcement rate*|
|1877-2014 (2139 Tests)||14%||4%||79%|
|1877-1944 (274 Tests)||22%||2%||91%|
|1945-2000 (1206 Tests)||14%||2%||85%|
|2000-2014 (659 Tests)||12%||7%||63%|
*Enforcement rate is the percentage of Tests in which the follow-on was available and was actually enforced.
However, while the enforcement rate has dropped, the success rate of teams that chose not to enforce the follow-on has improved. In the 21st century, teams not enforcing the follow-on with a lead of at least 200 have won Tests at about the same rate as teams enforcing the follow-on have.
The follow-on exists to allow teams that have a substantial lead to pick up 20 wickets without wasting time adding unnecessary runs to their lead. In eras when two runs per over was the normal scoring rate, it was often the case that the team batting first wouldn't have enough time to bat a second time, set a target and then take ten wickets. With covered wickets and faster scoring, does the follow-on make sense in its existing form? Assuming that scoring rates have improved by 0.5 runs per over (from 2.5 to 3.0), building a 200-run lead after a 400-run first innings takes 40 overs or nearly half a day less than it used to. In a game in which the team batting first makes 550 and the team batting second makes 400, scoring at three per over as opposed to 2.5 means using 63 fewer overs. Given that wickets don't deteriorate as much as they used to, would it not make sense to allow the team with the lead to have the option of batting fourth?
Thirty years ago, a Test with such scores would be written off confidently as a draw after the first innings. Today, a team making 550 batting first and conceding 400, for a 150-run first-innings lead, is in a situation where it runs a greater risk of losing the Test than the team batting second. To win, the team batting first has to make a bold declaration and take ten wickets in about 100 overs. A lot of teams manage to do this. The team that gave up the first-innings lead can save itself simply by killing the scoring in the third innings.
What if the team batting second could be forced to bat again, despite getting within 200 runs of the team batting first? On wickets that don't decline as much as they used to, wouldn't any team trade chasing 160 in the final 30 overs of the match with trying to take four or five wickets in the final 30? Instead of time on the fourth day being used up by both teams (batsmen playing against part-timers) while the match progresses to a point where a declaration may be safely made, would it not be better spent in the team with the lead chasing wickets?
Teams with large leads currently use up their second innings wickets trying to add runs to this lead, which typically makes it impossible for their opponents to consider victory as a realistic option. It is a profligate use of resources. The follow-on is irrelevant in these cases, as there is usually plenty of time to win.
For example, take a team which leads by 300 after two completed innings before the end of day three. This team is so far ahead in the game, that it really doesn't matter what it chooses to do. The follow-on rule, however, does hurt the team that has a lead of 150 after two completed innings an hour after lunch on the fourth day, and has no choice but to bat again and try and increase their lead from 150 to 300-325.
The declining use of the follow-on suggests two things. First, that time is not a factor in Tests in the way that it used to be. Second, that teams that lead by 200 or more have plenty of time to win irrespective of what they choose to do in the third innings of the match (this is why they win at about the same rate no matter what they do about the follow-on).
On the other hand, in matches where time is a factor, the follow-on threshold of 200 runs just gets in the way. Eliminating the follow-on threshold, but retaining the follow-on itself may persuade teams to chase wickets. It is often argued that teams bat a second time to give their bowlers some time to rest. For these teams, time is typically not a factor. Would teams give up the opportunity to win Tests simply to give their bowlers a rest? I think not.
Any team with a lead of any size should have the option of enforcing the follow-on in the third innings of a Test. Giving teams this opportunity will increase their ability to take 20 wickets.