The idea that something will happen because it is likely to.
It is often mistaken to mean that something is more likely if the opposite has happened a lot (called the "gambler's fallacy", where fallacy means wrong reasoning).
Example: you toss 3 tails in a row.
• so far it is 100% tails, but the Law of Averages says that if you keep tossing coins then eventually about 50% of all tosses will be heads.
• The "gambler's fallacy" is that heads is now more likely (it isn't). You see, the coin has no idea what happened before, so every toss is still a 50% chance of heads or tails.
So in the long run random events tend to average out at the expected value, but that does not help us predict the next value at all.