When measuring the effectiveness via reduction of issues, how do you account the natural stability introduced by reducing the updates/week each developer ships?

When measuring the effectiveness via reduction of issues, how do you account the natural stability introduced by reducing the updates/week each developer ships? When devs are tasked with code reviews and/or QA, this is time that could have been spent on their own features. In other words, if the product is stable today, and everyone's on vacation (no new updates), the product will generally remain stable save for unforeseen usage patterns.