Using administrative data on 3rd through 8th grade teachers from Houston Independent School District (HISD), we examine how teacher performance responds to the receipt of cash payments for high value-added scores.  The system is structured as a series of rank-order tournaments with teachers needing to exceed certain thresholds to receive the award. This allows us to use a regression-discontinuity design to estimate the impact of winning an award on teachers' future value-added. We find significant improvements for teachers that exceed the high (75th percentile) award threshold but no impact on the low (median) threshold, though teachers who did not win an award in a subject the prior year are positively affected by the median award.  The high threshold impacts also exhibit a negative gradient with experience - we find large effects for teachers with less than 6 years of experience but no impact for those with more than 11.  Finally, we find evidence that teachers who win the high award respond positively to both the award itself and its monetary value.