Here’s a field of research one wishes ardently were as absurd as it sounds. You should avoid crowding in zig-zag formations, apparently, and opt instead for standing in rows.

In a typical suicide blast in Pakistan, we have at least four times more punctured wounds, two times more injuries on the lower limbs, and five times more injuries on the human torso, compared to a non suicidal blast,” says Usmani.

According to the scientist, suicide attacks claimed by Lashkar-e-Jhangvi (LJ) have 35 per cent laceration and eight per cent punctured wounds rate, while those claimed by Harkatul Mujahideen (HM) have 31 per cent abrasion rate compared to 13 per cent of the LJ. On the other hand, the laceration rate of attacks claimed by the Baloch Liberation Army (BLA) and the HM is quite similar. The BLA suicide attacks have more than 30 per cent torso injuries compared to eight per cent in the HM and 20 per cent LJ attacks. Usmani claims: “If you give me the description of wounds, I can provide you with the name of the probable culprits”.

Usmani’s programme can also predict whether security measures put in place to thwart terrorist attacks at high-profile targets, such as the Presidential House, are effective. In most cases, Usmani says, his analysis is that “the heavy road blocks and barbed wires put in place [for security] will in fact act as shrapnel and may seriously damage or destroy the buildings”.

“What a world we live in,” writes Ricochet reader Damian Counsell, who sent this item to me, “where this kind of computer modelling is of benefit to humanity.” Quite.