Three Winning Predictions
A man had been receiving emails predicting which horse would win a race.
The predictions were right three times in a row, and he was about to pay a large fee for the next one.
But when he learned that emails from the same sender had stopped arriving for a friend partway through, he decided not to pay.
What did the man suspect?
Question your assumptions, one by one.
Choose questions to gather clues, then build your theory.
Ask your AI to host, then ask questions and work out the mystery.
View Hints
Hints
Use as many hints as you need, one at a time.
Hint 1
The predictions the man received really were correct three times in a row.
Hint 2
The man did not know what predictions other people had received.
Hint 3
The first prediction email was sent to a large number of people.
Hint 4
Different recipients were told that different horses would win the same race.
Hint 5
The predictions were distributed so that someone would be right whichever horse won.
Hint 6
People whose prediction was wrong were not sent the next prediction email.
Hint 7
If different predictions keep being sent only to those whose previous ones were right, the people left at the end will all have received three correct predictions in a row.
Check the Answer
Answer
The sender had been emailing predictions to a large number of people, naming different horses as the winner for different recipients.
They then sent the next prediction only to those whose previous prediction had been right.
Repeating this process left a group who had received three correct predictions in a row. The sender intended to gain their trust and charge them a large fee for the next prediction.
When the man heard that the emails to his friend had stopped partway through, he suspected that he too was one of the people left over by this process.
Explanation
The sender began by emailing free horse-racing predictions to a large number of people.
But the predictions were not the same for everyone.
One person was told that horse A would win, another that horse B would win, and so on, with predictions covering every runner distributed among the recipients. Whichever horse won, someone would have received a correct prediction.
After each race, the sender stopped emailing those who had received an incorrect prediction.
Only those who had received the right prediction remained, and different horses were again named as the winner for different recipients in the next race.
After repeating this for three races, the sender tried to sell the next prediction to the people who remained, for a large fee.
The man was one of those people.
Every prediction he had received had been right, so he had come to believe that the sender’s predictions were reliable.
His friend, however, had received an incorrect prediction partway through and had received no further emails.
Learning that the emails to his friend had stopped made the man realise that the sender might be selecting who continued to receive them.
The three correct predictions were real, but that alone did not mean the sender was good at predicting winners.
With that in mind, the man decided not to pay the fee.
If the solution makes sense to you, add your vote. One vote per browser.