I will explain why a random walk on $\text{SL}_{2}(\mathbb{R})/\text{SL}_{2}(\mathbb{Z})$ equidistributes with an explicit rate toward the Haar measure, provided the walk is not trapped in a finite orbit and the driving measure is supported by algebraic matrices generating a Zariski-dense subgroup. The argument is based on a multislicing theorem which extends Bourgain's projection theorem and presents independent interest. Joint work with Weikun He.