Would this study design be sufficient? I would like to study the efficacy of countermeasure versus a polygraph:

Use 30 polygraph operators, 10 of each skill level (beginner, intermediate and advanced). Randomly select 20 research participants. Train 10 of the participants in countermeasures to serve as the experimental group and the 10 untrained to serve as the control group. Randomly assign the participants of each group to each of the three skill levels operators. Use an ANOVA to see if the results indicate that countermeasures are statistically significant in defeating the polygraph. Utilize a Tukey's post hoc in order to determine the variables (if any) that did effect the outcome of the polygraph. Thoughts?