Pre-myopia, a non-myopic refractive state, is a key concern for myopia prevention because of its association with a significantly higher risk of myopia in children under 3 years of age. Amid the myopia pandemic, its onset at younger ages is increasing, yet research on screening methods for myopia and pre-myopia in preschool children remains limited. This study aimed to establish effective noncycloplegic screening methods for myopia and pre-myopia in preschool children. This cross-sectional study included 16 kindergartens in Shanghai, China. Uncorrected distance visual acuity (UDVA) was recorded using a logMAR visual acuity chart. Pre- and post-cycloplegic refractions were obtained using an auto-refractor (TopconKR-800). Noncycloplegic axial length (AL) and corneal curvature radius (CR) were measured using the IOL Master-700. Logistic regression models were developed to establish accurate noncycloplegic screening methods for myopia and pre-myopia. A total of 1,308 children with a mean age of 4.3 ± 0.9 years were included; among them 640 (48.9%) were girls. The myopia prevalence rate was 2.8% (n = 36), and the prevalence of pre-myopia was 21.9% (n = 286). Pre-myopia screening (cycloplegic spherical equivalent [SE] ≤ -0.5 < SE ≤0.75 diopters [D]) using UDVA exhibited an area under the receiver operating curve (AUC) of 0.52, noncycloplegic SE had an AUC of 0.70 and AL had an AUC of 0.63. The accuracy of combining the SE and AL/CR ratio was among the best with the least number of checks used, and the AUC was 0.74 for pre-myopia screening and 0.94 for myopia screening (cycloplegic SE ≤ -0.5 D). The addition of UDVA did not further improve the accuracy. Using UDVA alone did not achieve good accuracy in pre-myopia or myopia screening of young children. Under non-cycloplegic conditions, the combination of AL/CR and SE demonstrated favorable results for pre-myopia and myopia screening of preschool children.