I disagree.
Denyhosts is a very effective datasource for blocking.
Fail2ban works the same way, but for more authentication mechanisms.
If you apply the DShield method: correlate the same attacker across multiple sites and then block if it exceeds the threshold, it could be very effective.
ThreatSTOP are actually working on a way to be the aggregator and correlator for fail2ban.
It's not as simple as with denyhosts, which is strictly for ssh logins, since fail2ban is for multiple targets, and not everyone cares who is trying to exploit every target type, but it is useful, IMNSHO.