Gambling, expected value, matched betting and CFDs


After about 2 months of matched betting I’m beginning to see considerable gains being made from my ‘gambling’ activities. I was at first sceptical about matched betting as many of the uninitiated often are. Gambling is usually associated with financial loss and yet I, like many others who try their hands at this method of betting are able to profit at the expense of the large bookmakers. (If you aren’t already doing so, I highly recommend you give this side hustle a go).

Lately I began to think about my forays into advantage gambling in the past….

I can remember going with my friends to the casino and dropping a couple of hundred occasionally on the blackjack tables. Every now and then I’d get back a bit but overall, I definitely lost. In my late twenties I learnt about poker (a late bloomer, I know) and began watching the pros play on YouTube. The life of a Poker playing Pro such as Phil Ivy or Tom Dwan looked quite glamourous and cool. I read about online players who had only been at it for a year who’d built million-dollar bankrolls from playing at penny tables within a year. Drawn by the appeal and my interest in the math of the game, I began playing and although I enjoyed it, it was time consuming and I’d also have to admit that it was not profitable for me.  Whether I was just an average player and couldn’t overcome the rake or my strategy was poor – I couldn’t really say for sure.  Looking back on all these gambling activities, I’d say all up they have probably cost on the order of $2-3 thousand. Admittedly, the reason people gamble is not necessarily for profit – it’s a bit of thrill. Not many people going into casino on Friday night expect to come out winning on a regular basis – you’re paying for entertainment.

After reflecting on this I began to realise that my matched betting activities are in a way really only recouping some of the losses I’ve made thus far. I then began to wonder how much the average person spent on gambling through their lifetimes and how I compare. I know many people who buy a scratchy or lotto ticket. Although technically this is gambling, it isn’t frowned upon the same way as say slot machines or blackjack. If you bought one ticket for a lottery a week for a decade, the cost sure does add up and runs into the thousands quite easily though. So how much is average? Well the amount is quite astounding actually. Also, I’m proud that I’m well below it.

Apparently, in the US, people lose an average of over a hundred thousand dollars through gambling in their lifetime. Let that sink in!  . Perhaps unsurprisingly, it is actually more prevalent in lower income earners as well.

I then thought about the definition of gambling. Gambling is defined by Wikipedia as:

“the wagering of money or something of value (referred to as “the stakes”) on an event with an uncertain outcome, with the primary intent of winning money or material goods”

When you think about, trading currencies and even buying stocks could conceivably be thought of as gambling. Consider the decision to buy a particular stock of a company. Most people who buy stocks, buy them with a belief that they will be worth more in the future and increase their net worth or at least pay a dividend that they can then collect. The return the stock delivers is uncertain, even though history may give a good idea. The primary intent in owning a stock is to increase wealth through material gain. I think this makes a fairly compelling argument that buy a stock is in a way gambling. However, I would stop short grouping together buying stocks in Google and playing slot machines. Not all forms of gambling are equal.

Given that stocks achieve a return of 8 % or so annually over extended periods of time, they on average have a positive expectation. It’s the Expected Value (EV), which is what separates the different forms of gambling from one another. As much as matched betting is touted as being risk-free etc, there is actually small risks associated and qualifying losses (albeit small) are bound to be encountered to obtain many bonus bets anyway. However, unlike slot machines or playing blackjack (without card counting), matched betting, even when done by a novice, will probably yield a positive return or EV. No investment is really ever guaranteed – even the US treasuries have some risk and uncertainty in their return – there isn’t a 0% chance the US will default. Investments are actually on a spectrum – the idea is to pick the ones with positive expectation and a return profile that suits your risk appetite. Getting back to casino’s -the house has a mathematical edge – your expected value at any game is negative on average in the long run.

To me CFDs, which involve taking on leveraged trades on financial securities are basically on the side of the spectrum where slot machines and other negative EV forms of gambling are found. I’m sure many people have made a killing on them, but on average these forms of trading actually put the trader/investor at a disadvantage and are far riskier and less profitable (lower EV) than actually trading the commodity or stock as is. I was drawn to them back in my days of wanting to trade currencies. They give small traders with little capital access to the markets but it comes at a price. To borrow money the trader is charged interest. Although it’s the LIBOR and usually quite low, it eats in to profit margins. If a stock yields a return on average of 8% per annum and LIBOR is 4%, you lose half your potential profit trading it as a CFD. Compare this to owning the stock and selling it, where the charge is less than a ¼% for a reasonable trade size ( i.e. 10000 currency units), a much smaller drag on profit.

So why is matched betting different from many other forms of gambling? It has a positive expectation – it gives a long run edge. Looking at the expected value of a wager is the best way to understand this. I find this concept a really good way of evaluating whether gambling offers are something worth throwing money at. It also helps to rationalise qualifying loses during matched betting and hopefully convince the uninitiated of the legitimacy of matched betting as an income stream. (I find most people don’t believe you when you say it’s possible to make money off the bookies.)

Example: Let’s say a bookmaker offers a bonus bet if your horse runs 2nd or 3rd but doesn’t win, limited to $ 50 bets.

The odds-on favourite is $ 2.5 to win and $ 1.10 to place. The first thing to realise about betting is that bookies odds are actually quite accurate and there is much research to prove this. This is why building a system to beat the bookies at their own numbers is near pointless. So, let’s assume these prices probably reflect the true odds reasonably well (which they do). At $ 2.5, it’s basically a (1/2.5) 40% chance to win. At $ 1.10 to place, it’s 1/1.10 = 90% to place. Thus, the probability of finishing 2nd or 3rd is around 50% (90-40). So basically, on average, every two times I make this bet, I win a bonus bet. Given I’ll probably lose about 10% of a bet each time I qualify, in fees and odds differences between the exchange and bookie, I can make 80% of the bet size in bonus bets per 2 bets. Or 40% of the bet size in bonus bets per bet. Given that bonus bets can be converted to about 60-80% of their value in cash depending on range of factors. Your sitting on an EV of around 50-60 % of your bet size per bet. To put it another way, your making about $25 every time you bet on a race with an up to $ 50 bonus bet offer at these odds, if you bet $50. Mathematically you can prove matched betting in the long run has positive expectation. Your expected value (EV) is $25 per $ 50 bet. When you think about it doesn’t actually make much sense to bet more than $ 50, since the profit is diminished by the larger costs of odds differences and exchange commissions.

Recently I was looking at some online casino offers and determining whether or not the deposit bonuses offered represented a profitable opportunity. Due to the play through requirement, I established that I’d have an expected value of $ 300 if I were to attempt it. However, there would be considerable deviation. My return could vary anywhere between – $200 and $ 800 when I ran some simulations in Microsoft Excel. Expected value tells you what the long run average will be – there can be considerable deviations from the average over small samples for certain games – particularly certain forms of betting – such as no lay accas or each way betting.  In the case of this deposit bonus offer by an online casino, I decided in the end that depositing $ 1000 and have a small chance of a loss probably wasn’t worth it. Especially considering the number of hands of blackjack I’d have to play through (60 hours or so I estimated) to earn the expected value of $ 300 per deposit bonus. Gambling can be profitable, but it must be a positive EV play when you consider the outcomes of the wager.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s