Background: Patellofemoral instability (PFI) is a common condition in the pediatric population, requiring accurate assessment of patellofemoral joint morphology. While magnetic resonance imaging (MRI) offers detailed visualization of osseous, chondral, and soft tissue structures, Merchant view radiographs (M-XR) are widely used given their accessibility and cost-effectiveness and for detection of small patellar or condylar fragments that are not well-depicted on MRI. Agreement of key patellofemoral parameters between the 2 imaging modalities remains unclear. Purpose: To assess the agreement between sulcus angle, lateral patellofemoral angle (LPFA), and patellar tilt in pediatric patients with PFI, both before and after medial patellofemoral ligament (MPFL) reconstruction, on MRI and M-XR. Study Design: Cohort study (diagnosis); Level of evidence, 3. Methods: A retrospective cohort study was conducted on 84 pediatric patients ≤18 years who underwent MPFL reconstruction. Of these, 74 patients had preoperative MRI and M-XR and 44 had postoperative imaging. Sulcus angle, LPFA, and patellar tilt were measured, agreement between imaging modalities was assessed using single-measure, 2-way random-effects intraclass correlation coefficients (ICC[2,1]). Threshold values were applied to classify normal or abnormal measurements; agreement was assessed using Cohen kappa. Subgroup analyses were performed in patients with and without patella alta. Correlation was analyzed using Spearman coefficients, and paired t tests were used to evaluate differences in mean values of paired images. Results: The cohort had a mean age of 15 ± 2 years and included 51 (60.7%) female patients. Across all parameters, MRI and M-XR demonstrated poor intermodality agreement (ICC < 0.50) and weak positive correlations (Spearman rho, 0.27 to 0.42). Agreement between MRI and M-XR binary classification was poor for pre- and postoperative sulcus angle (kappa, 0.117 and 0.095, respectively), pre- and postoperative patella tilt (kappa, 0.421 and −0.044, respectively) and preoperative LPFA (kappa, 0.132), though postoperative LPFA demonstrated excellent agreement (kappa, 1.00). In 32 patients with preoperative patella alta (CDI ≥ 1.3), agreement in preoperative sulcus angle and LPFA on MRI and M-XR remained poor. Conclusion: Our study demonstrated that MRI and M-XR yield significantly different measurements of patellofemoral morphology in pediatric patients with PFI, with poor statistical agreement. These modalities should not be used interchangeably for clinical decision making or postoperative follow-up. Establishing novel cutoff values for radiographic-derived measurements applied to MRI is warranted.