Restorative justice defines a social movement of the past three decades. First I argue for a positive evidence-based vision of restorative justice. Then I problematize that vision. Restorative justice has scrambled the eggs of criminal justice to create a new omelet that some say is clearly defined, that consistently accomplishes modest statistically significant reductions of reoffending across large enough data sets, that is highly cost-effective, and that is emotionally intelligent justice (Sherman, 2003). Restorative justice is relational justice that conduces to emotional intelligence in how it serves the needs of victims in particular (Strang, 2002), but also offenders, their families, and their communities (Braithwaite, 2002). Restorative justice is also perceptually much fairer than conventional justice for all kinds of participants in justice processes and this contributes to effectiveness (Tyler, Sherman, Strang, Barnes, & Woods, 2007). One argument goes that all of this is true because restorative justice is a more evidence-based design with a more coherent theoretical foundation than alternatives. This also means that with strong standard-setting and training, the better restorative justice programs can deliver much more in all these ways than quick and dirty restorative justice by poorly trained people, of which there is a lot. Effect-sizes for restorative justice do improve more strongly than for other interventions when more hours of preparation and more restorative components are invested in them (Lipsey, 2009, pp. 141–142).