Welcome to the law of large numbers. Winning the lottery is a rare event--more than one in a million. A hard drive failure is a rare event--most drives can have trillions of successful reads and writes before they fail. Yet we've all seen the people holding up those oversized checks and who among us has never had a hard drive fail on us?

These guys see hundreds of pitches. There's hundreds of VCs in the valley. Something can be rare--less than one in a hundred and still be likely. Rare for a single instance can be likely and expected across a large number of instances.

There's a bit of confusion here with "the law of large numbers". People do win the lottery, but because it's very rare, only very few people do so. Hard drives may fail if they remain in use over a very long time, but only a few drives will remain so.

Accordingly, the "law of large numbers" as you report it, means that either only a handful of those hundreds of VCs in SV have witnessed inappropriate pitch behaviour, or this inappropriate pitch behaviour is not rare.

Regardless of the exact quantities, it remains true that an event is either unlikely to be witnessed, or it is not rare. That an event will be witnessed even though it is rare doesn't mean that it will be witnessed often.

The law of large numbers doesn't say that rare things happen often, is what I'm saying. Because that would be an oxymoron, not a paradox (which that "law" is meant to illustrate).