There is a significant difference in the decadal variation between June and July-August precipitation over Haihe River Basin, especially after 2002, the decadal variation characteristics of June and July-August precipitation over Haihe River Basin are opposite, so it is necessary to establish prediction models respectively. Based on the idea of year-to-year increment, several important factors related to annual increment of mid-summer rainfall over Haihe are selected through correlation analysis. Sea level pressure index (SLPI) of key areas in middle and high latitudes of Europe and Asia in the previous winter, Nino3 index in June and the difference between June and January of Nino3 index as ENSO evolution speed are used as key factors to establish the multivariate linear regression equation. Then forecast experiment of mid-summer precipitation over Haihe Basin in 2022 is conducted based on the predicted Nino3 index in June by models. The comparison between year-to-year increment model and climate model results initialed in March show that year-to-year increment has high prediction skill especially in flooding years. Then the failure hindcast case is carefully studied through the contribution of each predictor. The main factor is SLPI which reveals the relationship between East Asian winter monsoon and summer monsoon. The relationship is strongly relied on the following tropical Pacific and Indian sea surface temperature anomalies (SSTa) evolution. Nevertheless the tropical SSTA especially in western Indian during June exhibit unique feature may disturb the contribution of SLPI. It suggests that the nearly prediction of SSTa in key area highly associated with Haihe late summer rainfall should be paid attention.