I did misunderstand your previous comment. It's conceivable that people who guess correctly had experienced delay hypersensitivity reactions at the injection site (with vaccine, but not with placebo), but it's also possible that the group of correct guessers was small, say 3-6, and with a smaller number it's easier to obtain a series of correct guesses by chance alone. It's like flipping a coin and having it come up heads 3-4 (or more) times in a row. Statistics can do that. However, I think it's more likely to be the delayed hypersensitivity reactions.