The mainstream media coverage of overshoot in publications like the New York Times is intentionally misleading in order to push the (false) narrative that herd immunity requires infecting and killing far more people than you'd naively think. (Their numbers are only accurate if you assume no social distancing at all, which probably isn't even possible at this point.)

Basically, the number of infections peaks at the point where - due to some combination of immunity and measures like social distancing - each person infects at most one other person on average. However, there are still a whole bunch of infected and susceptible people spreading the disease, meaning that quite a lot more people then become infected than is actually necessary to end the outbreak. If you assume no attempt to flatten the curve, then a lot more people become infected than the herd immunity threshold, and that's what the NYT and co have focused on. However, one consequence is that flattening the curve without fully surpressing the outbreak reduces overall infections and deaths - the peak is lower and happens after fewer total infections, making it possible to build a whole bunch of herd immunity without all the extra infections and deaths from massively overshooting it.

Thank you for your explanation. That makes a lot of sense, and was a lot better than anythign i could find.