> Also, wouldn't the tolerance be doubled when you add them in series? Or does it still average out to +/- 5%? Neither. Let R_{1, ideal}, R_{2, ideal} be the

> Also, wouldn't the tolerance be doubled when you add them in series? Or does it still average out to +/- 5%? Neither. Let R_{1, ideal}, R_{2, ideal} be the "ideal" resistances; both with the same tolerance t (in your example t = 0.05). This means that the real resistances R_{1, real}, R_{2, real} satisfy (1-t) R_{1, ideal} ≤ R_{1, real} ≤ (1+t) R_{1, ideal} (1-t) R_{2, ideal} ≤ R_{2, real} ≤ (1+t) R_{2, ideal} Adding these inequalities yields (1-t) (R_{1, ideal} + R_{2, ideal}) ≤ R_{1, real} + R_{2, real} ≤ (1+t) (R_{1, ideal} + R_{2, ideal}) So connecting two resistors with identical tolerance in series simply keeps the tolerance identical.