News for hackers. Stuff that matters.
Re:Don't bet on the most likely winner by MaxBarraclough on Monday July 06, 2020 @04:41AM (#23745869)
But it's not. From the article:

> you want to bet on the mispriced horse, not the horse most likely to win

Re:Don't bet on the most likely winner by tsimionescu on Monday July 06, 2020 @07:22AM (#23746701)
If one horse is given 1:50 odds but should have been given 1:100, how many races would you have to bet on to make money on those odds?
Re:Don't bet on the most likely winner by mmmmmbop on Monday July 06, 2020 @12:45PM (#23749714)
Let's assume the odds are fair such that the odds of all horses sum up to 1. Betting on all other horses would give you 1:1.02 odds whereas they should have been 1:1.01. While there is no way to answer for sure how many races you would have to bet on to make money, we can look at the probabilities. There's a 95% likelihood that the horse will lose 5 races in a row. If you start with $100 and reinvest your winnings every time you'd earn a profit of $10 with 95% likelihood, meaning there's a positive expectation value. On the other hand, if the 1:50 odds were right, the probability of it losing 5 times in a row would be around 90%, leading to a neutral expectation value.
Re:Don't bet on the most likely winner by MCneill27 on Monday July 06, 2020 @12:53PM (#23749790)
Except investing isn't like a horse race. Each horse has a different finish line, and those finish lines are in flux during the race. Crucially, you can change your bet in the middle of the race.