Background Angular stable plates were introduced two decades ago as a promising treatment for fixation of displaced fractures of the proximal humerus (PHF). However, high rates of adverse events and reoperations have been reported. One frequent reason is secondary penetration of screws into the glenohumeral joint, due to sinking of the fracture or avascular head necrosis. To prevent joint penetrations angular stable plates with smooth locking pegs instead of locking screws have been developed. The aim of the present study was to investigate whether blunt pegs instead of pointed screws reduced the risk of secondary penetration into the glenohumeral joint during fracture healing after operatively treated PHFs. Methods From two different patient cohorts with displaced PHFs (60 treated with PHILOS plate with screws and 50 with ALPS-PHP plate with pegs), two groups were matched according to fracture type AO/OTA 11-B2 and 11-C2 and age (55–85 years). They were followed up at 3, 6 and 12 months. Primary outcome was radiographic signs of peg or screw penetrations into the glenohumeral joint at 12 months. Secondary outcomes were Oxford shoulder score (OSS) and Constant Score (CS) and radiographic signs of avascular humeral head necrosis (AVN). Results Eighteen PHILOS patients with B2 and C2 fractures could be matched with a corresponding group of 18 operated with ALPS-PHP with pegs. The number of penetrations of pegs and screws were equal between the two groups and the development of avascular head necrosis did not differ either. The functional outcomes for both OSS and CS at 12 months was clearly in favor of patients without joint penetrations in both groups. Conclusion We found no differences in the number of screw or peg penetrations in the PHILOS and ALPS-PHP group and the occurrence of AVN was equal. Joint penetrations led to inferior functional outcomes at 1 year. The ClinicalTrials.gov identifier 20/11/12 prospectively for the Philos Group is NCT01737060, and for the ALPS group 11/03/20 retrospectively is NCT04622852.