Recently I found a nice article on Wikipedia. It is about Parrondo’s paradox. This is a mathematical paradox that can be applied in games. Suppose that you have 2 games. You play it many, many times. But in the end you will loose the game. This paradox is stating that if you combine such 2 games, it might happen that you can win eventually.
How does this work? It is kind of complicated to explain, so I will give an example. Like I already told, we have 2 games:
- If you play this game, you will always loose 1 Euro.
- If you play this game, you will loose 5 Euro if the money you have at this moment is an odd number. If it is an even number, you earn 3 Euro.
Very simple rules in those 2 games. If I start with 100 Euro, I will loose it all in 100 times if I am playing only game 1. If I am playing only game 2, I also will loose all my money after 100 games.
What happens if I combine these two games?
Suppose I start with game 1, after that I play game 2, then again game 1 and so on. I will again loose all my money, because I first loose 1 euro, then I will loose 5 Euros, so I end with 94 Euros, then again 1 euro, again 5, and so on until I got an empty wallet.
Now I will try to start with game 2 and then game 1, game 2 and so on. Then I first earn 3 Euro. So I have now 103 Euro. Then I loose 1 euro, so I end with 102 Euro. But then I again earn 3 Euro.
Because I always end before starting with game 2 on an even number, I always earn more money than I loose. So I am a happy person now.
That is the paradox. You can gain by combinations of different games. In fact, by changing the order, I manipulated the outcome. Is that not what we as testers are supposed to do? Try to search for combinations that reveals problems in testing product.
Like Elisabeth Hendrickson tells it in her book Explore It!, we should try to search for hidden bears. This paradox is one of them.