This study investigated the relationships between internal and external training load metrics across a 2-week ‘in-season’ microcycle in squash. 134 on-court and 32 off-court ‘conditioning’ sessions were completed by fifteen elite squash players with an average (±SD) of 11 ± 3 per player. During every session, external load was captured using a tri-axial accelerometer to calculate Playerload; i.e., the instantaneous rate of change of acceleration across 3-dimensional planes. Internal load was measured using heart rate (HR), global (sRPE) and differential RPE (dRPE-Legs, dRPE-Breathing). Additionally, HR was used to calculate Banister’s, Edward’s and TEAM TRIMPs. Across 166 training sessions, Playerload was moderately correlated with TRIMP-Banister (r = 0.43 [95% CI: 0.29-0.55], p < 0.001) and TRIMP-Edwards (r = 0.50 [0.37-0.61], p < 0.001). Association of Playerload with TRIMP-TEAM (r = 0.24 [0.09-0.38], p = 0.001) was small. There was a moderate correlation between sRPE and Playerload (r = 0.46 [0.33-0.57], p < 0.001). Association of sRPE was large with TRIMP-Banister (r = 0.68 [0.59-0.76], p = 0.001), very large with TRIMP-Edwards (r = 0.79 [0.72-0.84], p < 0.001) and moderate with TRIMP-TEAM (r = 0.44 [0.31-0.56], p < 0.001). Both dRPE-Legs (r = 0.95 [0.93-0.96], p < 0.001) and dRPE-Breathing (r = 0.92 [0.89-0.94], p < 0.001) demonstrated nearly perfect correlations with sRPE and with each other (r = 0.91 [0.88-0.93], p < 0.001). Collection of both internal and external training load data is recommended to fully appreciate the physical demands of squash training. During a training microcycle containing a variety of training sessions, interpreting internal or external metrics in isolation may underestimate or overestimate the training stress a player is experiencing.