Purpose: To analyze whether exercise training conducted at night disturbs sleep and affects nocturnal cardiac autonomic control in high-level female athletes. Methods: A total of 18 high-level female soccer players (mean [SD] age 20.4 [2.1] y) wore actigraphs and heart-rate (HR) monitors during night sleep throughout night training days (n = 8) and resting days (n = 8), for 3 consecutive weeks. This was a longitudinal study that measured internal training load, sleep, nocturnal cardiac autonomic activity, and well-being ratings prior to training sessions. Results: Training load varied across training days (eg, training impulse range, mean [SD]; effect size, ES [95% confidence interval]: 72.9 [18.8] to 138.4 [29.6] a.u.; F_4,62 = 32.331; ${\eta }_{\mathrm{p}}^2=.673$ [.001–.16], large effect; P < .001). However, no differences in subjective well-being ratings were observed, although ES was large. Total sleep time (training days vs resting days: 07:17 [00:47] h vs 07:51 [00:42] h; ES = 0.742 [0.59–0.92], P = .005; moderate effect) and sleep-onset time (00:58 [00:19] h vs 00:44 [00:16] h; ES = 0.802 [0.68–0.94], P = .001; moderate effect) were negatively affected after night training. In addition, small effects were detected for wake-up time, time in bed, and sleep latency (P > .05). No differences were detected in HR variability during sleep (range of lnRMSSD: 4.3 [0.4] to 4.5 [0.4] ln[ms] vs 4.6 [0.3] to 4.5 [0.4] ln[ms]; F_3,52 = 2.148; P > .05; ${\eta }_{\mathrm{p}}^2=.112$ [.01–.25], medium effect), but HR during sleep was significantly higher after training days (range of HR: 56 [4] to 63 [7] beats/min vs 54 [4] to 57 [6] beats/min; F_2,32 = 15.956; P < .001; ${\eta }_{\mathrm{p}}^2=.484$ [.20–.63], large effect). Conclusion: Overall, the results indicate that exercise training conducted at night may disturb sleep and affect HR, whereas limited effects can be expected in HR variability assessed during sleep in high-level female soccer players.