IIRC, the real Iowa result had Sanders make a net gain of four points compared to the final RCP average for the state, and Sanders overperformed his final RCP average in New Hampshire by eight net points. Hopefully this means that pollsters have a knack for overpolling Clinton supporters and underpolling Sanders supporters, which could actually translate to a slight lead for him in the real result (even if it's already ORC).
It'll also depend on how energized people are in Nevada if he can manage to perform as well as he does with young people and tighten the margins with AA's and tie or outright win the Hispanic voter he will win Nevada.