Perhaps the BEST B1G Forum anywhere, here at College Football Fan Site, CFB51!!!

But betting against the masses is often a good idea.

Yeah, sports betting is just like poker -- a skill game with a high variance due to inherent luck.

It's also like poker in that the house doesn't care who wins. In sports betting the goal is simply to have 50/50 betting on each side of a wager, and in poker it's just to have action going. Sports betting it's the vig where the house makes its money, poker it's the rake.

The problem with sports betting, though, is that you can't make money with a 1-2% advantage, because you need to win 55%+ of your bets to beat the vig. Which means the skill isn't in betting every game and hoping that you can win >55%, it's in identifying where the spread or O/U is significantly wrong and that you can figure that out with >55% certainty.

Way back in 2006 I spent an entire season tracking picks on my old blog. I picked every Big Ten game--minus Purdue because I couldn't separate emotions--which is to say I *picked* Purdue but didn't count it to my season stats because I knew I couldn't be objective... I also would look around at the national scene and pick games where I thought there was something particularly interesting about the matchup or the line.

Over the course of a season, I would pick 8-10 games per week, and I finished the year with a 62.5% record against the spread. Which I was, and am, pretty proud of.

The problem was that when I did the math, if I had been betting $100/game, over 3+ months of work my entire winnings for the year would have been ~$1K. Based on the number of hours of analysis, the risk, etc, I thought to myself "this is a grind--it's just not worth it."

• Credit/collections: Suppose I have 10 people bet $100 each on PU and another 10 bet $100 each on tOSU. I'll owe $1,000 to the 10 winners and be owed $1,100 by the 10 losers so I'll make $100 but only if everyone pays me. If one of the losers skips town I'll collect $990 from the other losers, pay $1,000 to the winners, and be out $10. This is why the mob has to rough people up.
• Unbalanced bets: In the above examples I had equal bets so I made money either way but instead if I have 15 people bet $100 each on tOSU and only five bet $100 each on PU then I have a problem. If PU covers I get $1,650 from the 15 losers and owe $500 to the five winners so I'm up $1,150 but if tOSU covers I get $550 from the five losers and owe $1,500 to the 15 winners so I'm out $950. In order to cover this I have to call my bigger bookie and bet between $950 and $1,045 on tOSU. That severely reduces my profit margin because the unbalanced bets effectively don't count towards that 5% that I would make on bets placed through me.
• Volume: Even with no collections problems and perfectly balanced bets the bookie still only makes 5% of the total bet through him. To make $500 a week you need $10,000 a week in 100% collectible and perfectly balanced bets. At $50-$100 per bet/gambler that is 100-200 gamblers. That strikes me as an immense amount of effort and risk to make $500 a week.

I have a former coworker who was a bookie on the side, and we were talking about this.

He said that when you see a line that is getting bet heavily enough one direction that it's moving the line, bet the opposite way. The masses think they're smarter than the oddsmakers, and it's usually not the case.

I'd love to see if there is actually any truth to that, though.

I get the impression that the various "schemes" to "Beat the Odds" are WORK, for relatively little payout.

If something were simple and easy, the "market" would already have reacted to it and compensated accordingly.  It reminds me of GRQ schemes in the stock market, based off some analytic gizmo.