Inside sports wagering: A Q&A with the trader setting your odds (Part One)

Welcome to volume 4 of Winning Formula. Today, we’re flipping the format with a Q&A interview with a sports trader who sets the odds for some of your favourite sports. A straight Q&A was the best format since it was an information-dense chat.

Australia is arguably the sports gambling capital of the world. According to the AIFS National Gambling Trends Study published in 2023, 72.8% of Australian adults participated in some form of gambling, with 38% gambling at least once per week. Of regular online punters, 58.5% of them gamble on the AFL.

With the AFL season kicking off tonight, it’s the perfect time to understand better how odds are set.

I always assumed it was some supercomputer on the other side of the app. I can’t believe it’s just a room of 30-somethings, usually blokes, who love punting that are setting the odds for everyone else.

In this article, you’ll learn how odds are set and managed till the final whistle, the profit margin a sports betting company is trying to hit, why your friend's account keeps getting banned and how much information each betting company has on you.

In the next edition, we go deeper and explore how million-dollar bets are placed, the common biases that sportsbettors have, how sportsbetting has changed in the last 5 years and some hot takes on the future of sports betting.

It’s also important to note that some names and examples were changed in these answers to protect the source's identity. It’s fair to say my source risked their career for this interview. 

Table of Contents

1. The Role of a Trader

2. Why odds change

3. Data sources & the trader’s edge

4. What’s it like during a live game

5. Sport specialists

6. Margins and profitability

7. Sharp money vs recreational punting

8. Why accounts get banned

9. How does a trader monitor live bets?

The Role of a Trader

Q: Tell me about your role as a trader in a betting agency. How does it work?

A: We're called traders, and it's a 24/7 operation. Some companies have global teams because they need to cover sports 24 hours a day. 

Some will have day and night traders, but someone must be on duty 24 hours a day, 7 days a week, 365 days a year because markets are always open. 

Bets come in, sports news happens, and there's always live content. Trading is non-stop. To simplify it, there are a few key things we do. 

One is called price origination. This is where we come up with the price of an outcome, whether it's who's going to win a match, who's going to score a certain number of points, or even who's going to win a coin flip with heads or tails.

Part of that is math, and part of it is data.

Traders need a basic understanding of math and probability. You don't have to be a genius or particularly skilled with spreadsheets or calculators - finishing high school level math is sufficient. 

In the industry, traders typically have one of two skills when they get a job: either they're good at math or at understanding a certain sport. If you've got both, that's awesome. If you've got one, you can learn the other.

Q: So you could get hired if you're great at understanding a specific sport, even if you're not a math wizard?

A:  Exactly. If you know who Dustin Martin is and understand how whenever he gets tagged, that will impact his disposals, that is brilliant. 

Or if you've never heard of Dustin Martin but can work out that in his last ten matches he’s averaged 27 disposals, and he’s about to face a team that typically concedes more to strong midfielders, then you can deduce his stats might go up. 

So it can be all about a deep knowledge of the sport, or it can be about analysing the figures.

Why odds change

Q: What other responsibilities do traders have beyond setting the initial odds?

A: The second main function is price management. When we put a price up on the site, from that moment until the end of the game or until that selection is settled, we have to manage the price. This is the more complex part.

With live betting, price management is very intense because then you're trying to manage something with updates, scores, and live information coming in. 

If it's pre-match, it's more straightforward, and we consider factors like the weight of money, who's betting on certain things, lineup news, late player withdrawals, injuries, weather considerations for certain sports, and even coach statements about game plans before the game.

Q: What's the final part of a trader's responsibility?

A: After price origination and price management, we handle settlements. 

As a trader, once games are done, you must settle all the odds and all the bets - win, loss, or void. So it's a cycle: originate prices, manage them pre-match and live, and once it's done, settle the bets.

Q: So that means you're responsible for paying people out for winning and collecting revenue?

A: Yes. We're tasked with verifying outcomes. For example, how many disposals did Dustin Martin have: 25, 30, 35, 40? We have to key that in, and it flows through to our system.

Q: That explains why there can sometimes be disputes about stats.

A: Yes, stats will change sometimes, and traders can input things incorrectly, too. 

For example, the coin might have been heads, but we accidentally clicked tails. When that happens, customers are impacted, and we have to rectify it.

Data sources & the trader’s edge

Q: What does that look like from your perspective when gathering information? Are you watching games live?

A: For live data, most betting companies have a partnership with data providers. We get data feeds that the average punter doesn’t get. The big difference here is the speed of data. 

Most people don’t realise I’m also checking the same news sites and feeds like Twitter that the public reads for up-to-date information before and during the game.

What’s it like during a live game

Q: Take me inside the room where you work. Are you creating the odds for several games simultaneously or focusing on one game at a time?

A: It depends on the sport and the company you work for. 

Every company is a little different. Big companies will have in-house traders, while some smaller companies outsource everything. They'll say, "I want your table tennis odds," and accept the risks if there are errors. They're just trying to make their margin.

In-house traders are used to prevent errors and provide insight into the human element - my full-time job is creating odds for a single sport, which has benefits. 

The bigger the sport is, the less spread your focus is. A major AFL game might involve one person. If it's a final or a grand final, multiple people probably manage that game's price. 

If it's a smaller sport like college basketball, you might be looking at five games simultaneously.

It's a balance of how much information is coming in that needs to be managed. It comes down to effort versus reward - how much can I (as a trader) impact one college basketball game versus handling five games at once? We need to consider how much money flows into each match. More money equals more focus.

Q: Is there a significant difference in betting volume between popular and less popular teams?

A: Absolutely. What would you rather watch? If it's more high-profile, there's probably more money on it. 

I don't have precise data, but I suspect a relationship between people watching and betting on a game. These days, not many people bet on things they don't watch.

Recreational bettors will look at the best game of the week, or the game in primetime, like Geelong vs. Collingwood, and decide to bet on that. 

They're not analysing which game offers better value; they just want to make the game they're watching more fun by having a bet on it.

Sport specialists

Q: Do traders typically specialise in one sport, or are they versatile across different sports?

A: It depends on the person's skill set and knowledge. Some know all sports and can add value across the board. I personally only know AFL well. If you put me on Rugby or the NBA, I'd know the fundamentals, but I wouldn't be able to add the same value.

For example, if the star forward gets injured at the Lakers, I wouldn't know who will score in his place.

That's where the human element comes in. We can't teach a computer easily the second and third order ramifications of a change. If a star NBA player is injured or getting targeted, a trader who covers that sport well understands the player who will score more in their place and can set those odds well.

Margins and profitability

Q: How are the odds set so they’re in the company's favor? What kind of margin are betting companies typically aiming for?

A: If we took no margin, it would just be a 50-50 battle, like a coin flip, and the odds by the bookie would be $2.00. 

Most companies are probably trying to make around 15% margin over time. There are tips and tricks to disguise that margin so customers might not realise it. 

Casinos do something similar. When you play roulette, the margin (or house edge) is about ~5%, but it's fairly well hidden, and the game plays quickly.

Sharp money vs recreational punting

Q: Can you tell the difference between smart money and recreational money? Is smart money just very large bets?

A: It's not always big money. There are a lot of giveaways that help us identify sharp bettors.

The first step is trying to determine their margin. If someone has a winning record against us, that's a good indicator.

We can see the profit margin on a person's account when bets come in. If someone is consistently winning, that tells us they're good. The comparison is, if you worked at a bank, you'd probably want to know how much money someone has in their account. That's our most identifiable factor - what's their winning or losing margin with us.

But it gets more complex. You can have people who've won a lot of money, but it might have been luck, or they might have bet on errors we made. And vice versa - someone might have lost a lot, but they're making smart bets across a wide range of markets.

For the math-oriented folks, there's something called expected margin. It's simple math. If I bet something to happen at $2 odds, but when the game starts, the true price or closing price is $1.50 odds, my expected margin is positive. I've bet at higher odds than what the market determined was fair at close.

So the expected margin is up based on the prices I got versus what they closed at. 

That particular bet could still lose, but my process was sharp. If I keep getting those prices, I should win 5% in a thousand-bet sample size. Today, I might lose, but my approach is solid.

Why accounts get banned

Q: What other behaviors help you identify sharp bettors?

A: Customer behavior is another indicator. If you're consistently betting on our errors, you're pretty sharp.

Some people sit there waiting for us to make an error, and because there are plenty of errors, they can do it all day.

Q: Those accounts eventually get restricted, I'm assuming?

A: Basically, yes. We also look for arbitrage betting - people who bet when our odds are out of line with the market. 

Sometimes, we'll get arbitraged when we don't want to, and sometimes, we'll get arbitraged when we do want to because we want to lay money on a certain outcome.

It's pretty easy to spot because we have odds comparison tools showing prices across different bookmakers. 

So if Marcus Bontempelli is $5 to score two goals at one company but $2.5 at most others another, we immediately ask why that company has a high price but are also looking for bets that could take advantage of this disparity.

When someone places an arbitrage bet, we know almost immediately because so many people are doing it that we'll see five similar bets come in within 20 seconds. We’ll flag these accounts and another team at the company will look into it.

Q: That explains why I hear friends complain about their accounts being limited even though they thought they were being clever.

A: Yes, because they're betting on errors.

You don't get banned for betting recreationally on average markets. You get banned when consistently trying to win by exploiting mistakes or inefficiencies. It becomes really obvious.

How does a trader monitor live bets?

Q: How do you monitor incoming bets? Do you have a system that flags certain wagers?

A: Yes, a screen shows bets coming in, almost like a Twitter feed. It displays certain information for each bet but doesn't show the customer's name due to privacy considerations.

The display might show things like the customer's state of location, the bet, the amount of the bet, the price they received, the game they are betting on, and any previous notes on that customer from other traders in the company.

Q: Where do those notes come from?

A: From what someone has already input. If it's your first bet, we have no information on you. At this stage, you're just at default settings, like creating a new character in The Sims.

As time passes, we might add notes: "This customer bet on an error," or "This customer had a thousand dollars on this market, but it looks recreational.” 

We start to build a profile of the customers' patterns or typical behavior as they place bets. Different teams within the company work together on this. We'll pass on notes like "That bet was fine; they could have got a better price elsewhere" or "That bet was an error; they're just here to exploit mistakes."

We can then flag things for another team at the company to look into more if something feels very suspicious.

Q: What percentage of bettors would you classify as sharp versus recreational?

A: Every company will be different. At newer companies with lots of promotions, a higher percentage of customers might be sharp because they're trying to take advantage of a new operator that has minimal information. At long-standing companies with good recreational customer experiences, the ratio is different.

At established bookmakers, it might be 98% recreational to 2% sharp. That's enough to notice the difference. And we get to a point where we don't care what that 98% of recreational bettors do - they don't even come up on our radar.

For example, on AFL Grand Final Day, a million bets might be placed, but what comes to the trader's screen might be only 2,000. We have filter rules to show us only certain bets. It's like when you go to an arcade and play a game - the leaderboard doesn't show every single person's score, just the top performers. That's what we want to see.

We can't monitor every single bet because there are too many. We focus on the ones we think are worthwhile or concerning. If you bet $20, we don't care. We'll pay attention if you bet $1,000 and your account is flagged.

Well, that’s part one of volume 4 on sports betting. Stay tuned for more!

The next edition will cover how gambling trends are changing, common bias’ our trader sees that costs punters loads of money, how someone who wants to place a one-million-dollar bet does it and spicy predictions on the future of sports betting.

These newsletters grow with personal referrals, so please forward them to someone who loves sports, business and in this case, punting on sport!

If you have any feedback or ideas for a newsletter, don't hesitate to contact me on LinkedIn, reply to this email or email me here or by replying to this email.

‘Til next time,

👋 Will