We are in the process of automating ZAP to run regularly against a variety of test applications and will publish the results here as and when they are in a suitable state.
Our aim is to make ZAP as effective as possible against real world apps. Test apps are useful tools but we have found that some apps test for issues that are:
Having said that, any of the tests that ZAP currently fails on are a good candidate for fixing. If you are interested in enhancing ZAP to find these issues then see Contribute: Coding: Scan Rules
Google Firing Range - a test bed for automated web application security scanners (unmaintained)
Google Security Crawl Maze - a comprehensive testbed for web security crawlers (maintained)
OWASP Benchmark - a test suite designed to verify the speed and accuracy of software vulnerability detection tools
Websites Vulnerable to SSTI - a set of simple servers which are vulnerable to Server Side Template Injection (maintained)
Yahoo Webseclab - a set of web security test cases (unmaintained)