Some empirical data, with each point the average of 50 trials: <br /> <img src="http://cs.smith.edu/~orourke/MathOverflow/MinDistSquare.jpg" alt="MinDistSquare" /> <br /> The curve $1/(\sqrt{2} n)$ (displayed) fits rather well. Congrats to Douglas Zare and Ori Gurel-Gurevich!