“…One way to address this challenge is to proxy players’ relative ability using contestants’ ranking (see e.g. Cohen-Zada et al, 2017b; Jackson & Mosurski, 1997; Leach, 2003; Page & Coates, 2017; Rottoff et al, 2011), the number of wins in the regular season (as in Morgulev et al, 2020a), or using betting odds (as in Berger & Nieken, 2016; Malueg & Yates, 2010). For example, M&Y use betting odds to identify matches of players with equal skills and then use that sample to disentangle strategic effect models from psychological momentum in best-of-three tennis matches.…”