RL is Return Loss and measures the loss of reflected power. As we do not want a lot of reflected power, the greater the value (loss) the better the result. Y axis maps with a low value at the top and larger on the bottom so a passing result will graph below the limit line.