This is another good point, there are people out there who are above average at selecting winners. It's a slightly self defeating idea in some ways however, because if they are really that good at predicting outcomes then A) they would likely just make it into a profitable career and keep their ideas to themselves or b) be found by bookmakers who would take their long term winning streaks into consideration when determining fair odds. We all need to remember that bookmakers are using some of the most sophisticated prediction software, with decades worth of historical analysis, when automatically pricing and updating the bets they will take - with a margin of safety added on top. However, if you are content and willing to grind out low profits across many bets, this kind of strategy might work - just make sure you track the profitability outcome over time.
Yes, I understand that due to the mathematical advantage over the long distance, the coolest player will most likely lose to the bookmaker, but apparently there are segments when a person is maximally involved in a certain sport and can calculate the odds more accurately than the bookmaker. And by the way, we all know that any software sometimes makes dramatic mistakes, so a person always has a chance (not at a distance, but at separate intervals).