New odds boosts
Browsing through the odds boosts of soccer matches lately, I’ve noticed an increasing trend for multiples on player statistics coming up as the featured boosts. You’ll regularly see offers of two players to commit two or more fouls, two players to have multiple shots on target, three players to all win a foul, and so on.
They sound pretty likely to happen in the course of a game and the odds on them are attractive.
These bets can sound very attractive to a lot of players, and it’s not hard to see why they draw interest. They sound pretty likely to happen in the course of a game and the odds on them are attractive.
But let’s take a closer inspection. Let’s try and get a handle on how these compare to other odds boosts and see how we can work out if we’re facing a decent offer or not.
Place your bets
Here’s a classic example from PaddyPower on a Brighton vs. Fulham game: Danny Welbeck and Raul Jiminez to have two or more shots on target, each boosted from 9/1 to 11/1.
A quick glance at the stats for the season and you might assume both are about evens to have one or more shots on target. The bookmaker odds on either one individually won’t be anywhere near that, but remember, this is an odds boost, so let’s think about it. At evens each, to both have one or more shots on target they’d have fair odds of about 2/1. All good so far.
So, what are the fair odds for both to have two shots on target? Is it 4/1? 6/1? 8/1? All would be intuitive answers, and the odds boost seems like far too good a price to ignore. The correct answer? It’s over 40/1. That leaves the “boost” as one of the very worst bets the bookie will have laid on the game.
The Poisson trap
Why are the numbers so different? Well, it’s thanks to something in math called the Poisson Distribution.
First, if both players have about a 50% chance to have a shot on target, that usually implies less than one shot on target on average per game due to the way distributions work. It’s closer to 0.7 expected shots per game. Still, not a big deal. It still works out at a fifty-fifty chance.
The problem is that the progression doesn’t increase smoothly. The odds of having two shots on target is not simply twice that of having one. Instead, it follows a distribution which is curved, rather than linear.
you already know that a team scoring six goals is not half as likely as a team scoring three
You almost certainly already understand this for other things in football. You already know that a team scoring six goals is not half as likely as a team scoring three; it has a much lower chance of happening. Teams score exactly three goals all the time – it’s happened 36 times already this Premier League season at the time of writing. But only once so far has anyone scored exactly six. This is perfectly normal and logical. It doesn’t seem unusual because it isn’t. The distribution isn’t linear.
This also applies to everything else in soccer, which people find less intuitive to understand. Scores are common things to see and think about. Even a casual soccer fan will see dozens, perhaps hundreds of them a week. This is not true of player shots on target or fouls won.
Because of this distribution curve, a player who is evens to have one shot on target is not 2/1 to have two. They’re actually closer to 11/2 – a 15.4% likelihood, to be precise. Multiply those two together, and you have 15.4% × 15.4% = 2.37%, which is a little over 41/1.
This only gets more severe the further we go up. How would you feel if, instead, the bet was three shots on target each? Ask your friends what they think the odds should be, and it wouldn’t be a major surprise to hear some fans estimate 8/1, or even lower. A bookie offering a boost to 40/1 here would look like they were giving away money to many punters. In fact, the fair odds are about 816/1.
Avoiding the trap
So, how can we avoid getting suckered in by these when having a punt? Well, first, think about the appeal of the bet: “the odds can’t be that much higher, it’s only one more foul.” You wouldn’t fancy backing a team to get double the number of corners that you think they’ll get, so don’t think the same for fouls won or committed – even if it is only one more.
Poisson plays a part in things like corners, too, of course. The difference between over five corners and over six corners can be much larger than it appears, although not as much as with lower-frequency events.
Some odds boost bets similar to this are quite reasonable, and as a very, very general rule they tend to be for things like team fouls and corners – things that have high totals over the course of a game. Rarely are good bets spotted in players to have multiple shots on target or fouls, because the expectations are usually quite low. An increase of one can be a lot.
there’s a limit to how greedy sportsbooks should really be
In general though, consider what you think the total should be in a match before lumping on simply because it sounds plausible. “Bookie creates odds boost in their favor that is intended to seem attractive” is not going to be the biggest betting scandal of the year. That’s what odds boosts are there to do, after all. But there’s a limit to how greedy sportsbooks should really be, and regularly offering specials with a 75% edge is definitely over the line.