Introduction: Executive functions (inhibition, attention shifting, updating) are linked to math intelligence in school students and adults. This link is particularly important because performance in school mathematics is predictive of various competencies later in life. While some researchers argue that tests of executive functions and math intelligence measure the same underlying construct, others argue that they measure distinct but correlated constructs. Also, evidence on the differentiation of cognitive skills over time exists. Clarifying the relation between executive functions and math intelligence is, however, challenging, especially because preschoolers cannot fill in commonly used questionnaires that require them to read. As a consequence, researchers have to resort to behavioral, verbal, apparatus-, or computer-based assessments of executive functions. Objectives/Methodology: With this meta-analysis of 29 studies containing 268 effect sizes, we examined the link between executive functions and math intelligence for a total sample of 25,510 preschool children. Specifically, we synthesized the corresponding correlations and sought to clarify which executive function assessments were used for preschool children and how the assessment characteristics may moderate the correlation between executive functions and mathematical skills. Results: Utilizing three-level random-effects meta-analysis, we found a moderate correlation between executive functions and mathematical skills in preschool children, r = 0.35. The type of assessment (behavioral, verbal, apparatus-, or computer-based assessments) did not moderate this relation. Differentiating between the three executive functions revealed average correlations of r = 0.30 between math and inhibition, r = 0.38 between math and attention shifting, and r = 0.36 between math and updating. These analyses will be supplemented by further moderator analyses. Conclusions: Our findings support the significant link between executive functions and mathematical skills in preschoolers—yet, the average correlations do not suggest that both measures are identical. Results will be discussed against the background of deployed assessments and testing environments.