Willie the Duck
Hero
Mike Mearls would talk about 3rd editions spiked chain, an exploit that wasn’t really leveraged by most until after the edition had been out. Something that could leverage a ton of rather cheesy optimizations and yet discovered by groups only after they’d been playing.
Does 5e have any of them? I tried to think of one but nothing really satisfies. I was thinking maybe the optional rule flanking.
Is it more dependent upon the group?
It depends on the thresholds. If the premise is simply an exploit* specific to the edition that people recognize after they've gotten to know the rules, then there are probably several.
*And the definition for this is going to be murky. Let's say that it is: something you do when you realize the rules make notably effective that gives you an at-least significant leg up on the person who just goes along with traditional patterns like sword&board fighter, picking feats and classes individually based on their appeal rather than specific synergy, etc.. Cheese is even harder to define, so let's leave that at a 'I know it when I see it' level.
- One one level (and most comparable to spiked chains), I think it took everyone a hot second in 2014 to look at quarterstaves being versatile, another to notice that Polearm Master included quarterstaves, maybe a third to add on a third component like Shillelagh or dueling fighting style or the like, and boom we had one-handed quarterstaff and shield little whirlwinds of attacks. Given the balance of melee weapon combatants compared to ranged and spells and turning into bears and the like, it's hard to call it overpowered (although still frustrating for the person who wanted two-weapon fighting or sword and shield to be good play options), it sure feels like an edition-specific-rules-confluence bit of cheese that people grokked a bit after the game came out.
- Wish-Simulacrum cheese also takes a brief moment for people to find, and then everyone knows about it. It's such an obvious thing that I half-expect it was the developers just wanting to make the neigh-infinite-wish-loop obvious and every DM to ban it and get the whole thing out of the way
- Cha-class synergy in general was a pretty obvious 'exploit' that got discovered quickly. Full-casters to rapidly increase paladin smite slots or warlock dips to enhance paladin/bard ranged at-wills showed up almost immediately. Coffeelocking obviously took until Xanathars to get going, but then everyone found it. Lore bard3 (sometimes with knowledge cleric 1) dips with anything to become a skill-monster as well.
- Combining the bonus-action-attack feats (PAM and CBE) with the -5/+10-attack feats (GWM and SS) also probably counts.