After a five-month experiment following tipster models to engage in sports betting, I’ve discovered that it’s a tedious and emotionally-challenging way to make money. Success is possible, but only for those who are disciplined, rational and committed for the long term.
Gambling, as people know, is a great way to lose money. But betting – a more structured approach to gambling that doesn’t rely on luck or chance – does offer a way to make money.
Put simply, betting, also known as trading, refers to an agreement between two parties where one party makes a prediction and loses or makes money if his prediction turns to be true.
To see if I could make money from betting on sports, which lots of bloggers I’d read had suggested was possible, I embarked on a five-month experiment. Between May and October 2016, I followed two different tipster models, which told me exactly which bets to make.
With no previous experience in sports betting, I thought the models would make the sports betting process easy and stress free.
How wrong I was.
While good models have proven results over a long period, they are time consuming, tedious and emotionally challenging. Plus, you will have need to risk a decent amount of your own money.
If you have a disciplined, rational, contrarian mind and are willing to commit for at least a few years, then a tipster model might be a nice way to make a bit of extra money.
What is it?
At a glance
- Tipster models recommend bets to make, with the aim of earning a small profit on your total outlay.
- They compare bookmaker odds to their own odds, and recommend bets on perceived value.
- Each model generally follows a single sport or code.
A tipster model will recommend bets across a range of matches, events or sports. The model might be specifically for:
- one type of bet (e.g. win/lose)
- one sport (e.g. tennis)
- a range of sports and markets (e.g. the batsman with the most runs for cricket, under 41 points for an NRL match, or premiership winner for the EPL).
Tipsters develop a ‘model’ by analysing a wide range of data and coming up with a likelihood for something happening. Sometimes the analysis is purely data driven, other times it has some personal intuition included.
Recommendations typically come with four components:
- the amount
- the selection
- a price
- the best available bookmaker.
A typical bet would look like this:
In this example, you would place at 1 unit bet on Kohli, to be the top Indian batsman, with Sportsbet for $4.00.
Let’s look at each of these components in more detail.
The market is what you are betting on. In the above example, the market is for a batsman to score the most runs in a cricket match. What market you bet on depends on the model you follow. Football, horse racing, and tennis are pretty popular markets to bet on, but there are heaps of options to choose from.
The bookmaker is where the recommended bet can be found. Sometimes it is at more than one bookmaker. Depending on the model, you may need accounts with lots of different bookmakers.
Before you start betting you need a ‘bank’, which is the total amount of money you are committing to betting. A unit is a percentage of your bank. Tipsters generally recommend that one unit equal 1% or 2% of your bank. So, if you had $1,000 as your bank, one unit would be $10 or $20.
Bet size ranges between 0.25 to 5 units depending on the service and market. The greater the chance of winning, the more units that will be committed to the bet.
Why do they use units? Tipster models make money by placing many bets over a long time. By risking only relatively small amounts on any given bet there is time for the model to produce results.
It is important to understand that a one unit bet is 1% or 2% of your total bank.
If you bet 10% of your bank (e.g. 1 unit = $100 from a $1,000 bank), you are gambling. On bad weeks a model will lose 10–15% of the bank (e.g. $100–150 of a $1,000 bank), which isn’t very pleasant but very much recoverable. If you were using 10% of your bank for one unit you would lose everything in a single bad week.
The price is the odds that the model thinks is good value.
All betting odds have an implied value. To work it out, you divide one by the odds.
If a bookmaker offers $2.00 for NSW to beat Queensland, then the implied likelihood is: 1 / 2.00 = 0.5
This is implying that there is a 50% chance NSW will beat Queensland.
If NSW’s odds were $3.50, the implied likelihood is: 1 / 3.5 = 0.28
This implies that NSW have a 28% chance of winning.
Note, this is the implied likelihood, but not the actual likelihood.
Tipster models come up with their own likelihood for a certain outcome, and in doing so find value bets.
Let’s say that a model works out that NSW has a 48% chance of winning (or a price of $2.08). This will then be compared with a bookmaker’s implied value (assume they have set the likelihood at 40% or $2.50). Here you can see there is a discrepancy between the model’s price and the price that you can get at a bookmaker.
Tipster models make money by exploiting these discrepancies across many bets placed over a long time.
So how does this all work then?
If you bet $10 one hundred times on NSW to win with an implied likelihood of 40%, you would, in theory, win 40 times and lose 60 times.
So you would win: ((10 x 2.50 – 10) x 40) = $600
and lose: (10 x 60) = $600
Hmm, that didn’t go so well did it!
But, let’s now assume you’re following a killer model and it has told you that the actual likelihood of NSW winning is 48%. So, you won’t only win 40 out of 100 times but 48.
In this case you win: ((10 x 2.50 – 10) x 48) = $720
and lose: (10 x 52) = $520
Now you’ve made a handy $200 profit, which on having bet $1,000 is a 20% return. That’s almost as good as property investing! Almost…
Now imagine if you didn’t only bet $10 each time but $100, you would have made $2,000. This is good, but considering it might take an entire season to place 100 bets it’s not amazing. If you had of bet $500 each time you would have made $10,000, which is starting to get fairly rewarding.
I know what you’re thinking. That sounds pretty easy, and $10,000 would buy me lunch and gym memberships for three years, where do I sign up?
The thing is, I’ve only really told you the good parts so far.
Here’s the not as good parts:
- The model might be wrong.
- The model might be right but going through a prolonged period of wrongness (perhaps you get the 52 wrong bets all in a row, before any of the 48 right ones).
- You will be recommended a bet that seems so unlikely that you simply cannot place it, but it does go on to win.
- You regularly won’t be able to get the odds that the model tells you to get.
- You will almost certainly get sucked in by flashy bookmaker ads and bet on things other than just the model. Nothing good will come of this.
- You will spend multiple hours each weekend on your phone, outside a café (where your friends are drinking coffee and eating some hipster baked good that you’re not really sure the name of), looking for 4G reception so you can place a group of bets.
- You will get disheartened at one point and take a break only to miss a spell of winning bets.
- You will not be able to place every bet because it will only be available at an obscure bookmaker that you never sign up to.
So, do models work? They can.
Here’s a screenshot of results of Steve, the Daily25 blogger, after five years of following tipster models – he’s made $400,000.
Now compare that with my results after five months.
This five-month period roughly aligns with the most recent downward turn on Steve’s graph.
Yes, it can work but it isn’t easy, quick or glamorous. Because:
1. You will need to sign up to a lot of bookmakers. It’s time consuming and god only knows what this does to your credit rating or likelihood of identity theft.
2. It takes time – a lot of time – to receive and place bets. Some services are simple and straight forward and you’ll be done in 10 minutes. Others not so much and you’ll spend an hour during lunch time trying to find markets and prices.
3. It requires you to use and risk your own money. You must pay for a service and bet with your own money. To cover the cost of a tipster model, your bank needs to be $5000. I doubt you would actually make money with this amount but it would cover the cost of subscription. Models only aim for 5% to 10% returns. If you placed 100 bets at average amount of $50 you would bet $5,000 in total. Assuming you make a 10% return, you’ll only get $500. I doubt you will find a service that costs less than $500 for the time needed to place 100 bets. Tipster services are capital intensive.
4. Most importantly, it is emotionally straining. I was using $100 for one unit and some weeks lost over $1000. Conversely, some weeks I won $1000. During some games, I can go from a definite $500 win to losing $300 in two minutes.
– Models generally lose more often than they win (win rate about 40%).
– They might get ahead by winning only two or three bets out of 10.
– They sometimes have long periods of losing; my longest losing streak so far is three consecutive weeks.
It is hard to feel like you are constantly losing.
It affects your mood, which affects how you interact with people in the real world i.e. your partner, kids or employer.
You’re also hardwired to cutting loose and run. This is a good thing; it’s why you haven’t been eaten by a lion yet. But it doesn’t help when trying to follow a tipster model.
You need to be disciplined and able to compartmentalise your emotions.
- Good, honest models have a long history of fairly decent returns.
- Some models are fairly easy to follow.
- For me, it has been a really interesting learning experience and has exposed me to a whole new world. Due to betting, I have learned more about statistics, probability, computer programing and human behaviour.
The not so good
- It is emotionally tough – don’t underestimate how tough it is!
- You need a decent amount of your own money to bet with.
- It requires a lengthy commitment i.e. three years.
- It can actually be quite time consuming and finicky.
Should you do it?
- If you are a rational person that can go against the crowd and have good control of your emotions, then a tipster service might work for you.
- If you are even slightly emotional, impulsive, undisciplined, have limited capital or get easily addicted to things do not try this.
I recommend you look at Daily25 for an honest reflection of the realities of sports betting (on a bad month this guy loses $50,000).
Don’t get too excited about the returns that tipsters claim they make.
They may be accurate, but only in theory. Try to find out how much effort it will be for you to follow their service. One service I followed had quite remarkable results but was much harder and more stressful than others I have tried.
- Steve from Dailyprofit – the same Steve as Daily25 – is very open and transparent. I have followed the NRL, AFL, NFL, NBA and NHL models.
- Championbets is professional and also records results in an open way. I have followed Matt’s Sports Bets and the tennis package.
Lastly, this type of thing would really lend itself to automation for many reasons. Many of the downsides could be eliminated with a simple script.
I’m convinced that a lot of the people using the services are already doing this and it’s the only way I would try sports trading again.