Peach fruit set is affected by cumulative chill and spring frost. A spring frost occurred on 29 Mar. 2015 at the U.S. Department of Agriculture, Agricultural Research Service (USDA-ARS) Byron station after 3 weeks of bloom, reducing fruit set and resulting in many buttons (abnormally small fruit with dead embryos). Fruit set was rated in 2014, 2015, and 2016 and button set rated in 2015 using the same scale (0 = no fruit/button to 9 = 1–2 fruit/button at every node). The overall fruit set rating was substantially different in the 3 years, averaging 5.61 in 2014, 2.61 in 2015, and 6.04 in 2016. Buttons and skin-damaged fruit in 2015 varied among peach genotypes. Comparison of fruit and button set ratings showed that there was no difference between cultivars and selections, but some significant differences in fruit set for four ripening months, among the 3 years, and among the nine chilling classes, respectively. Among the cultivars, the most common button set rating was 0–3. For example, ‘Sunprince’, ‘Loring’, and ‘Carored’ trees had a high button set rating, whereas ‘Flameprince’, ‘Julyprince’, and ‘Contender’ trees were low. As for peach selections, BY04P1690n was among those with the highest button set rating. In the population derived from a cross of button-prone BY04P1690n and button-free BY99P3866w, fruit and button counts from 10 long fruiting shoots ranged from 4 to 53 fruit (21.63 on average) and 2 to 27 buttons (10.39 on average). The peach button rate ranged from 5.36% to 87.10% (30.70% on average). The range, distribution, and percentage of the button counts suggested that, if buttoning was genetically controlled, it appeared quantitative. Further assessment is needed.