A new probability measure Q on the same outcomes can be built by reweighting: each outcome x counts with a weight dQ/dP(x). The weight exp(μx − μ²/2) turns standard normal samples into N(μ, 1) samples without moving a single point. Slide μ and watch the samples keep their places while the weights, and with them the weighted histogram, change. Pricing works the same way: the real-world measure P and the risk-neutral measure Q share every path and differ only in the weights, and Girsanov's theorem gives the weight for whole Brownian paths.