Predictive Capabilities of Avalanche Models for Solar Flares

We assess the predictive capabilities of various classes of avalanche models for solar flares. We demonstrate that avalanche models cannot generally be used to predict specific events due to their high sensitivity to their embedded stochastic process. We show that deterministically driven models can nevertheless alleviate this caveat and be efficiently used for large events predictions. Our results promote a new approach for large (typically X-class) solar flares predictions based on simple and computationally inexpensive avalanche models.