Finetuning D/L method for Twenty20s

I think the Duckworth-Lewis system for Twenty20 games needs reviewing. This occurred to me after England's World Twenty20 match against West Indies in June. To recap: England scored 161 in their 20 overs and, following a rain-delay, West Indies were set a revised target of 80 from nine overs. The number of Powerplay overs was reduced from six to three.

This was all mathematically logical, but the revised target presented West Indies with an easier target. To understand why, think about it this way. It is as if West Indies were told: you have already batted for eleven overs, the score is 81 for 0, and you have a further 80 runs to make to win. Not only that, but three of the remaining nine overs are Powerplay overs. Which team wouldn't accept that with open arms?

The reason why this target was too soft boils down to two things that make Twenty20 different from the 50-over game 1. Powerplay overs are much more valuable in most Twenty20 games than most 50-over games. 2. High scoring-rates can be more easily maintained for the lower number of overs you get in a reduced Twenty20 game. So, in the England-West Indies match, a fairer target would have been possibly nearer 85 or even 90 with no Powerplay overs left.

Here are two ways the ICC could tweak the D/L system for Twenty20 games. 1. Reduce the number of Powerplay overs in a linear rather than a proportionate way. By that I mean that if the number of overs remaining is halved, the number of Powerplay overs should not be halved but reduced by the number of total overs reduced. For example, if the number of overs is reduced by one, from 20 to 19, the number of Powerplay overs should be reduced by one, from six to five. This seems to me to be logical and mathematically justifiable 2. Consider recalibrating the maths slightly for the shortened version of the game to require proportionately higher scoring-rates as the number of overs remaining is reduced. I accept that the second suggestion is, for a layman like me, less easy to justify mathematically, but the first seems to me to be logical, mathematically understandable, and obvious. The D/L system has worked well in ODIs even though it is incomprehensible to the average person. I can live with that as long as it is logical. But if it comes to the point where the system is both incomprehensible and illogical then that is surely the time for it to be changed.