Originally Posted by
Pokerjoe
You're thinking of it wrong.
When we bet on sports, whether we know it or not, what we're doing is saying "I make the midpoint of the distribution curve of results differently than the market does."
Think of it this way: If one poker player has pocket aces and his opponent has kings and they go all in preflop, which player should you bet on? Let's say the price was 5-1. The odds on KK beating AA are ~4-1, so you'd bet on the dog. If you only think in terms of "predicting", you'd have to bet AA every time, no matter the odds, which would be really, really dumb, and which is, in fact, the psychology behind squares overbetting faves.
If you laid 5-1 on that race and I took the bet and we ran it out 4 times and I lost them all, most bystanders would be snickering and calling me an idiot. Then on the 5th race maybe I catch a king on the river and those same bystanders would go crazy, calling me a luck box, and claiming it's all unpredictable. Yet the truth is that the kings are supposed to win one in five races, and how they win, king on the river, four-flush board, whatever, doesn't matter, that's random, that's unpredictable, that's irrelevant, and that's not what you should be focusing on. It doesn't matter how you win or lose bets. There are no bad beats. It's absolutely predictable that if you play enough football games, one will be decided by a missed extra point on the game's last play. That isn't a bad beat, it isn't "unpredictable!" it's actually perfectly predictable.
Stop trying to predict results, and just try to estimate chances, and then compare that estimate to the market's estimate.
The market estimates an average score for the game, which we call "the line." When your estimated average score varies from the line, you bet. If you are better at estimating average scores than the market, you profit. If not, the juice eats you up. That's the game.