According to the individual zones of optimal functioning (IZOF) model, an athlete's performance is successful when his or her pre-competition anxiety is within or near the individually optimal zone. When anxiety falls outside the optimal zone, performance deteriorates. The model also suggests that skilled athletes are aware of, and are able to accurately recall and anticipate, their pre-competition anxiety. A meta-analysis of 19 studies from 1978 to 1997 (146 effect sizes based on 6387 participants) was conducted to examine the validity of the assumptions regarding the in-out of the zone notion and the accuracy of recalls and anticipatory measures of anxiety. The findings provide fairly good empirical support for the IZOF anxiety model, with an overall effect size (d) for the in-out of the zone notion of d = +0.44 (41 effect sizes, n = 3175). In other words, the performance of athletes who were within their individually optimal zones were almost one-half a standard deviation unit better than that of athletes who were outside their zones. Furthermore, both effect sizes (r(w)) for accuracy of precompetition anxiety measures, recall (r(w) = +0.71, 24 effect sizes, n = 369) and anticipatory (r(w) = +0.69, 81 effect sizes, n = 2843), exceeded the 'large effect' suggested for correlations by Cohen. The implications for future research extending the IZOF model to a wider range of positive and negative emotions are discussed.