Same nightclub, but now a few of the doors are faulty.
LQT1: the IKs exit door is half-jammed. People can’t leave as fast as they should late in the night, so phase 3 stretches out — but evenly, just a slow, uniform closing. Since exercise is exactly when you need that exit to speed up, this one tends to decompensate with sport or swimming.
LQT2: the IKr door is outright glitchy, opening and sticking at random. The exit turns chaotic, and sometimes the door pops back open and lets a second small wave of people back in before the room has fully cleared — that’s exactly the bump you see on the orange curve (the EAD, early afterdepolarization). If that second wave is strong enough, it can fire off a stray new action potential, which is the classic starting point for torsades de pointes. Typical trigger: a sudden noise, an alarm going off, a strong emotion.
LQT3: here the problem isn’t the exit but an entry door (Na+) that won’t shut properly. A steady trickle of Na+ keeps leaking in throughout the plateau, so the room stays “lit up” longer — hence the abnormally prolonged plateau. That leak shows up mostly when everything’s quiet, so sleep and bradycardia are the classic risk moments.
The common thread across all three: the room always takes longer to close down (repolarize), so the QT stretches out on the ECG — and the more unstable and unpredictable that closing gets (as in LQT2), the higher the risk of torsades.