"Thus, a lot of calculations lead to success, and few calculations lead to failure." — The Art of War by Sun Tzu

It is rare in Pakistan's democratic history for a party to have been so confident of its chances that it has relied solely on campaigning to win an election, given the intricacy of the country's electoral dynamics and the unpredictable nature of the various "external" stakeholders' roles. All the appearances of a boisterous public dispute aside, every election has actually been conducted like a chess match.

Every political party runs an election on two fronts, even though the majority of regular voters may not be aware of this. First, there is the performative one, which calls for well-known figures to walk or cry on stage while attempting to win over voters with ludicrous promises and/or alluring "bayaania" (story, if preferred).

The other is more understated and is being waged by "scientists" who understand the election process and work behind the scenes to put their party in the best possible position to win even before a single vote is cast.

All political parties strive to gain a comfortable majority in as many of the constituencies they are running for office by combining these two tactics. They understand that they cannot win everywhere in reality, but they can use data to turn past setbacks into future triumphs. "Swing constituencies" and methods for turning them around are useful in this situation.

"Swing constituencies" are those that one can win over by influencing minute shifts in the preferences of the electorate. Every election in Pakistan produces a sizable number of "swing constituencies" that, in the event of a slight variation in voting trends, may have favored one of the two front-runners.

Swing constituencies’ are those that can be won by influencing very small changes in voter preferences. In 2018, there were 81 constituencies where the difference in votes received by the winner and the runner-up was less than 5pc of the total ballots cast