Although patient outcomes after allogeneic hematopoietic cell transplantation (allo-HCT) have significantly improved in recent years, complications and associated mortality remain substantial. Although many transplants are performed worldwide, the number of patients enrolled prospectively into clinical trials is small. Patient and physician preferences often override treatment assignments in randomized transplant trials, biasing the common intention-to-treat analyses. Large retrospective and observational database studies are likely to detect the real effect of allo-HCT. However, they may be subject to immortal time and other biases derived from heterogeneity of allo-HCT indications and approaches and differences in referral or institutional policies affecting patient selection. Timing of the transplant procedure may be fundamental but studies commencing at start of transplant may neglect the influence of pretransplant therapies. Conversely, a prolonged lag period between the decision and execution of transplant may artificially 'improve' the outcome by 'natural' selection weeding out patients relapsing or dying before transplant. Finally, comparative nonrandomized transplantation trials often suffer from unbalanced assignment for therapy arms. We herein present common clinical dilemmas discussing proper application of available evidence in daily clinical practice. Pitfalls and caveats frequent in clinical studies of allo-SCT are highlighted to promote a balanced interpretation of available data.