In the House, Demo­crats would need to net 30 seats to win back a ma­jor­ity, but in a wave elec­tion, un­der­dogs of­ten pre­vail. The Cook Polit­ic­al Re­port rates 31 GOP-held seats as com­pet­it­ive (either as toss-ups or lean Re­pub­lic­an). Of those 31 dis­tricts, 23 are based in urb­an or sub­urb­an areas where Trump’s brand of pop­u­lism is un­likely to be a selling point.

There are two main ar­gu­ments that Re­pub­lic­ans are re­ly­ing on to per­suade them­selves that Trump’s nom­in­a­tion wouldn’t be cata­stroph­ic. One is that even pres­id­en­tial can­did­ates who lose in land­slides don’t usu­ally bring their party down with them. Demo­crats picked up Sen­ate seats in George McGov­ern’s and Wal­ter Mondale’s em­bar­rass­ing elec­tions. And Re­pub­lic­ans only lost one Sen­ate seat in Barry Gold­wa­ter’s epic 1964 de­feat. But the party of all three los­ing can­did­ates suffered sig­ni­fic­ant House losses, and it’s hard to see in today’s po­lar­ized polit­ic­al en­vir­on­ment how Trump’s di­vis­ive can­did­acy wouldn’t ripple down-bal­lot.

Second, some GOP strategists be­lieve that Trump’s abil­ity to at­tract new voters could give a pop­u­list boost to es­tab­lished sen­at­ors. In Ohio, as the the­ory goes, the es­tab­lish­ment-friendly Port­man has already locked down col­lege-edu­cated Re­pub­lic­ans and could be­ne­fit from Trump’s re­ori­ent­ing of blue-col­lar voters in the GOP’s corner. But this the­ory ig­nores Trump’s dam­age to the GOP brand, which could make col­lege-edu­cated swing voters less com­fort­able vot­ing for Port­man or any Re­pub­lic­an.