Research into errors in mathematics classrooms is often centred on student error. Whilst investigating teacher emotional expressions, using data from experienced teachers and affective pathways, I encountered examples of teacher mathematical error occurring in association with expressed emotions. Similarly, I have observed instances of new teachers making errors and hence have begun exploring the implications of teacher error management. In this paper, after describing illustrative examples, I suggest a model based on a continuum of affectively driven strategies that are likely to be familiar to secondary mathematics teachers. Each affectively driven strategy has implications for teacher student relationships and for the learning of mathematics. I discuss some of these implications in terms of longer term affective impact on classroom climate and for students. I offer the model for further discussion in relation to developing growth mindsets.