The aim of the study was to retrospectively investigate differences in training duration and intensity between runners of which submaximal test performance decreased (D), unchanged (U) and increased (I). Eighteen competitive runners were monitored over 46 weeks by daily training logs and 7 submaximal treadmill tests. Training duration was recorded in minutes and intensity was determined by session Ratings of Perceived Exertion (sRPE). Duration in intensity zones was calculated over 6 weeks according to the classifications: Zone 1, sRPE ≤ 13; zone 2, 14–16; zone 3, sRPE ≥ 17. Also, total training duration was calculated. Change, compared to the first test, in submaximal heart rate (HR) and heart rate recovery (HRR60s) was categorised into D, U and I. Previously established day-to-day variation in HR of 4bpm (Achten and Jeukendrup 2003) and HRR60s of 6bpm (Daanen et al.. 2012) were considered as U. 426 weeks of training and 71 tests were examined using multilevel analyses. Total training duration was 8.7 ± 1.1 h/week averaged over 6 week periods. Duration in zone 1, 2 and 3 was 47 ± 11%, 39 ± 11% and 14 ± 7%, respectively. Number of tests with HR classified as D, U and I was 23, 40 and 8. HRR60s classified as D, U and I was 19, 41 and 11. Runners who showed a decreased HR trained 4.6 h more in 6 weeks compared to runners who showed no change. However, no differences between the HR groups were found for intensity. Runners who showed a decreased HRR60s trained 2.4 h more in zone 3 compared to runners who showed no change. No differences in total duration and duration in zone 1 and 2 were found between the HRR60s groups. It can be concluded that runners who train more (irrespective of intensity) show a decreased HR, indicating improved submaximal running performance. Training at high intensity seems to be related to a decreased HRR60s indicating worsened submaximal running performance. These findings can help coaches to prescribe appropriate training.