Fading the public is one of the oldest and most used methods for picking games, and it can sometimes be profitable, but don't let it be an excuse to avoid actually handicapping the games and making an informed play. The logic is this: sportsbooks make money, right? And sportsbooks arguably have the best resources at their disposal to handicap games, i.e. data, models, etc. right? So if the books are setting a line that draws a disproportionate amount of public money on one side, then it stands to reason that they know something the public doesn't, and are going to bury the public for a nice pay day, right? Sometimes yes, but often no.
Think of it this way. Whenever the books set a line that draws a lopsided amount of money, the books are making a bet on that game. Just like you or me. But unlike you or me, the books don't have to hit 53+ percent to be profitable. Depending on the breakdown, they only need to hit a significantly lower percentage of the time.
For example, let's say the books set a line that draws 75% of public money on Team A, and 25% on Team B, at the standard price of -110. So:
Team A -7.5 -110 75%
Team B +7.5 -110 25%
And for simplicity's sake, lets say there is $100 total "to-win" dollars wagered, $75 on Team A and $25 on Team B.
If the public side wins, then the book pays out $75 to the winners, but collects $27.50 from the losers, who laid $27.50 to win $25 at the -110 price. So if the public side wins, the book loses $47.50.
If the non-public side wins, then the books pay out $25 to the winners, and collect $82.50 from the losers, who laid $82.50 to win $75 at the -110 price. So if the non-public side wins, the book profits $57.50.
So, the book stands to either make $57.50, or lose $47.50. In such a case, the book only needs the non-public play to hit more than 45.2% of the time to make a profit.
Moral of the story is that, although we can all recall times when public side gets buried and it seems clear in hindsight that the books knew something, a lot of this is confirmation bias. In reality all the books are doing is making a bet, and it's a bet that can hit less than 50% and still make money. You, on the other hand, cannot make money hitting less than 50%, and so "tailing" the book by fading the public can be a dangerous proposition.