Background Malalignment is a known risk factor for patellar instability and can be assessed using radiographic metrics such as tibial tubercle-trochlear groove (TTTG) distance and tibial tubercle to posterior cruciate ligament (TT-PCL) distance. The most commonly cited threshold is TTTG>20mm indicating malalignment, however this is based on CT scans, and differences have been reported between MRI and CT measurements. Furthermore, substantial variability has been reported in thresholds for abnormal values based on imaging modality and patient size. Therefore this study aimed to identify MRI-based thresholds for measurements of malalignment and the accuracy of these measurements in identifying patellar instability, as well as to determine the impact of patient sex and height on thresholds for instability. Methods A retrospective analysis was conducted using an institutional database to identify knee MRIs of patients with symptomatic patellar instability and age and sex matched controls. Clinical data were recorded and measurements of TTTG and TT-PCL distances were performed on axial knee MRI images. Receiver Operating Characteristic (ROC) curve analysis was performed to determine the area under the curve (AUC) and confidence intervals, with AUC >0.7 indicating a good diagnostic test. Youden’s J statistic was used to establish thresholds for detecting patellar instability. Subgroup analysis was performed based on sex (male vs. female) and height (above vs. below the median height). Results A total of 240 patients were included in the study (64% female, mean age: 27 ± 9 years). The overall AUC and 95% confidence interval for TTTG distance was 0.75 (0.69, 0.81), while TT-PCL yielded values below 0.7 for all subgroups. The study cohort had a median height of 170.2 cm (177.8 cm in males and 165.1 cm in females). The identified cutoff values for TTTG distance were 13.7 mm for “short” females, 14.7 mm for “tall” females, 16.1 mm for “short” males, and 20.1 mm for “tall” males. The AUC for TTTG distance was 0.79 (0.7, 0.88) for patients below the overall median height and 0.68 (0.57, 0.77) for those above. AUC for TTTG distance was good in both females and males, at 0.78 (0.7, 0.85) and 0.7 (0.57, 0.82), respectively. TTTG distance in males had greater predictive ability in males below the median male height (0.79 (0.61, 0.94)) versus males above the median height (0.56 (0.39, 0.72)). Among women, the TTTG cutoff had a similar efficacy among those with heights below and above the female median (AUC = 0.77 (0.64, 0.88) and 0.76 (0.64, 0.86), respectively). Conclusion TTTG distance on MRI is an effective measurement for detecting malalignment related to patellar instability, while TT-PCL thresholds demonstrated low efficacy. In all cases, the cutoff value was below the reported value of 20mm. We additionally identified varying thresholds for TTTG distance based on sex and height, with the lowest diagnostic value among patients with above median heights, particularly in male patients. Notably, calculated cutoff values varied between male and female subgroups. These findings highlight a need for incorporating patient specific factors when using these thresholds in clinical practice.