I've been trying to get a handle on this because I'm just not buying the 300 victims per molester. I found the following very interesting document that I recommed everyone who is interested in this subject should read. http://www.missingkids.com/download/nc70.pdf
This is pretty authoritative and one thing it makes clear is that there is a vast difference between situational molesters and pedophiles. While it may be that as Waiting says, there are some 'super predator' pedophiles that have high numbers, the situational molesters do not.
I think the following quote is telling, "What is the difference between child molester and a pedophile?
Unfortunately, the two terms are often used interchangeably, which causes some confusion. The term "child molester" is ultimately very generic, and means only that an individual has committed a sexual offense against a minor. The term "pedophile" is often used to describe individuals who prefer sexual relations with children. It is more helpful to refer to "situational" and "preferential" pedophiles, a distinction made by the FBI Behavioral Science Unit. The situational offender is a person who prefers sexual relationships with an age peer, but who, when experiencing high levels of stress or for other reasons, turns to a minor for sexual contact. Incest offenders are often situational offenders. Many will offend against only one victim during their lifetimes, although others may have a series of victims during their lifespan.
On the other hand, preferential pedophiles have had few, if any, sexual relationships with age peers, and have a life long interest in children, most often males. They may literally have hundreds of victims during their life, and they typically pose a much greater risk to the community at large." http://www.svsotp.com/pedofaq.htm
To make any sense of these numbers we really have to have some idea of what kind of people we're dealing with. I suspect most are going to be "situational" offenders.
One additional comment, I have found Amazing's approach to this matter interstesting. However, I find it somewhat disconcerting that as soon as the numbers dropped he threw his 'conservatism' out the window. If the 'conservatism' had any value whatsoever then it should still apply.