If you live in Australia and own a television, you can’t have failed to notice them. There they are in the middle of every sporting event. Relentless offers to bet, bet, bet, with various sweeteners to encourage you to sign up for an account, and bet bigger and more often. You know the sort of thing: money back if your team loses by less than 6 points. Get a bonus bet if they lead in the second half and lose.
The problem that the bookmakers have is that they are all essentially selling the same product while competing for a finite number of punting dollars. In a world of smartphones, their competitors are only ever a click away, so they need to offer something to stand out from the crowd, in the hope that you will open an account and give them all your money.
That’s what they want, mind you. But what if there was a way to turn these offers into an almost risk free opportunity to profit?
Take this typical example offered by a number of books at various times during the AFL season: money back up to $50 on AFL head to head bets if your team leads at the end of a quarter but loses.
For the sake of this example, let’s assume that two bookies have this offer running and that there is a real coin flip of a game this round where two teams are about even chances to win, paying $1.90 each.
If we played this how the bookies intended, and just backed one team to win we’d be risking $50 to possibly win $45 ($50 x 1.90 = $95). If the bet loses we’d maybe get our money back, but there’s also the chance of losing our $50.
But here’s a better strategy: how about we back both teams with different bookies offering this promotion, in the hope of hitting the promotional refund. $50 on Team A at Bookie A, and $50 on Team B at Bookie B. With these two bets, we’ve paid out $100, and one of our bets will win, returning $95, while the other will lose.
So we’re guaranteed to lose $5 no matter who wins. **
Why would we want a guaranteed loss? Well, what we’re hoping for is that the losing team will have led at the end of a quarter, in which case our losing bet will be refunded, resulting in an overall profit of $45, all for just $5 risk.
Instead of placing a $50 bet at odds of $1.90 that a particular team will win, we’ve effectively placed a $5 bet that the losing team will lead at a break, at odds of $10 (the $5 we risked, plus the $45 profit means an effective return of $50 on our synthetic $5 bet).
How likely is it that a team will lead at the end of a quarter and go on to lose? Well, to answer that we can use the data available at afltables.com and a data scaping tool to get the quarter by quarter breakdown of every AFL game since 2000. It turns out that this happens in about 40% of games:
So the “true” odds of our $5 bet on a 40% chance should be 1 / 0.4 = $2.50. As we’re effectively getting $10 that’s clearly massive value (we only need one in every 10 bets to succeed just to break even).
Now, not all offers are quite as good as this one (as you can see from those stats the offer of a refund if your team leads at half time or at 3/4 time still represents value, but will hit much less frequently), but the basic idea holds: back all the outcomes, across multiple books with promotional offers, for the smallest risk possible, and you can’t lose.
** Yes, so there is also the rare chance that the match may end in a draw; in this case most Aussie bookies pay out using the dead heat rule, in which case the loss will be about the same.