This is a difficult question to answer without getting super technical. But there has always been a balancing act between playing optimally (the game theoretical strategy which if perfected would be an unbeatable strategy) versus exploitably (playing in a way that tries to maximize against an opponent's particular strategy which has mistakes). Most top players nowadays have veered towards the optimal route, however, there are many instances where exploitable play is more profitable against certain opponents.

I'll try to give a basic example. Let's say there is $100 in a pot and we have $100 remaining. If I were playing a game theory optimal strategy, I would bet strong hands 67% of the time and bet $100 and bluff 33% of the time and bet $100. This is because I am giving my opponent 2:1 odds to call, so he needs to beat me 1 in 3 to be profitable, and I want to maximize how often I can make this bet and also do it in a way so he is indifferent to calling or folding with his mediocre hands. And my opponent knows he must call with at least 50% of his hands, otherwise when I bluff I am winning money since I risked $100 to win $100, so I need him to fold at least 50% for those bluffs to make money.

But if I know my opponent is folding too frequently then I can just bet every hand I have and bluff way more often and win money. Or if I know he is calling too frequently I will never bluff and always have strong hands. These adjustments are playing the player, and it helps to get someone to win the most money he can, but it is an exploitable and beatable strategy if the opponent properly adjusts.

Short answer: I try to play the odds most of the time