(Disclaimer: I've always been in favor of overbooking and have personally played the VDB game often. I've only been IDB'ed twice.)

I think the logic is this (in favor of overbooking): on certain routes and on certain days and times, there's both a very high demand as well some degree of no-show activity. Airlines know they can fill every seat. Thus the specific ticket sales leading to the overbook situation - say, the last 15 tickets they sell - are being sold at max revenue to business travelers. Because these sales make up a part of the whole revenue model for the flight, they enable the airline to compete for price-sensitive leisure travelers as well.

Without overbooking, the airline would still want to sell to those business travelers, so they'd raise the bottom leisure fares and keep more seats upon closer to the flight date.

The reason IDB rates are low en masse are threefold:
- There are a large enough body of flights happening every day where IDB, VDB, overbooking, etc. are not a topic at all. Someone with knowledge of the data could reliably predict flights where the IDB rate is 0 out of millions and millions of passengers flown.
- Data analytics is far better than it used to be. Even the oversold flights are usually oversold by just the right amount. Odds of the airline missing by 10-20 seats are rare. If they miss by 3-4 seats, getting the VDBs is usually easy. (The UA3411 situation had nothing to do with overbooking, really.)
- Most overbooks don't result in extreme delays for the bumped passengers. A more typical delay is 3-4 hours, and there's always someone willing to take a $400 voucher for that kind of delay.

In other words, I suspect an overbooking ban would affect some routes by a lot, others not at all. It would also affect a business traveler's ability to find availability close to a flight's departure: these are the exact travelers the airline *wants* to accommodate.