Conversations with Daniel Chiang: A data analytics approach to sports betting.

Sports betting is all the rage now, particularly during the COVID-19 pandemic, since you can do it easily and efficiently online. For most people, the image we get when thinking about a sports bettor is someone smoking a cigar and making his picks based on gut feelings. That’s the Hollywood version. The truth is much different, particularly in the 21st Century.

Last month, professional sports bettor and founder of Line Sniper, Daniel Chiang, successfully cashed in his $30k ticket to win $300k after the Tampa Bay Rays made it to the 2020 MLB World Series. Now, Daniel holds several future bet NFL tickets that were bought back in August including:

  • He holds a 10k to win 300k ticket on the Pittsburgh Steelers to win the Superbowl.
  • He holds a 35k to win 400k ticket on the New Orleans Saints to win the Super Bowl.
  • He holds a 20k to win 110k ticket on Patrick Mahomes to win MVP.
  • He holds a 5k to win 55k ticket on Russell Wilson to win MVP.

He set aside some time to discuss his analytics-based approach to making wagers with SCINQ Sports.

Let’s start with some background. How did you get your start betting?

Initially, I came from the financial world. I went to Wharton School of Business for my undergraduate, and I worked in equity hedge funds after college. While at the hedge fund, I started playing high-stakes backgammon. That’s how I ended up running into an individual who was one of the best bettors in the world. I figured out very quickly that a lot of different skills that I’d been utilizing — data analysis and modeling — in the hedge fund world, he was doing in the sports betting world. We teamed up and I applied some of my analytical and data skills from finance. We ended up being extremely successful. That’s how I came into the sports betting world.

Can you take me back to your first sports but not that first first big one just first sports bet. What was that like?

I remember my first sports bet pretty distinctly — unfortunately, just not the team. It was William Hill at Baldwin Park. A World Cup game. Denmark vs Finland under two goals. It was very exciting and very interesting because I was making these bets but really had very little clue about the high level sports betting that was available at the time. It’s kind of one of those things that I was doing for entertainment. I ended up winning luckily enough, purely by chance. From that point on I started trying to figure out: Is there a way to analyze sports betting similar to how I looked at financial markets? 

And how did you progress to high stakes sports betting?

One of the interesting things about betting despite it being such a big industry is that, at the time there were very few resources out there teaching people how to bet. I’m talking about really teaching people how to progress within the industry because by definition the best bettors in the world don’t like divulging their secret because that’s where they get their edge. If I hadn’t run into the individual playing high level backgammon and become very close friends, I would have never gone into high level sports betting. I simply observed what he was doing. I worked on sports betting models. That’s how I learned about the industry and learned about how to gain an edge in sports betting.

Which sports do you like to bet on most?

My favorite sport to bet on — and generally the most profitable for me — is football, both college and NFL. We also bet on MLB which we were quite good at. Interestingly, the NBA quickly became my least favorite sport to bet on, despite it by being my favorite sport.

Why is it a difficult sport to bet on?

Simply because during the regular season in the NBA, it is very difficult for me to get a feel for the team and the players. In the NFL where every game matters, you really see the players play each drive as close to 100% as humanly possible during the regular season. In the NBA, I found oftentimes that during games you were handicapped in certain ways. You very quickly figure out that for X, Y, and Z reasons a player has given up after the first quarter or one of the wasn’t playing seriously for the first three quarters… It just became extremely difficult to bet on basketball.


For people who aren’t familiar with sports betting, it involves a lot of acting on hunches and lots of luck. It’s what you see in movies. But in truth, like you say, there’s a lot of research that can be involved, especially the higher you go. Can you walk us through your process? How do you get a thesis going and then you know how you test it out.

People tend to believe much too highly of their opinion of the game or set of circumstances. Take the most recent Sunday Night Football game for example. The Bucs vs the Saints. Going into that game, you hear Brady is one of the best quarterbacks of all time, you can’t lose that game. Other people are saying that the Saints are definitely going to win. 

At the highest level of sports betting everything is a probability distribution or set of possible outcomes. The most important thing is to find a good price on the bet that you are placing. 

That’s where a lot of the data sciences and unique models work. For example for the Tampa Bay Rays, we had a Monte Carlo simulation running on the entire season based on over 10,000 simulations, utilizing our unique inputs. We simulated every game of the season tens of thousands of times. Ultimately, it came out that the Tampa Bay Rays were going to win a certain percentage of the time. 

What type of models do you tend to use? Also, how much weight do you put on your modelling data?

Different sports bettors will tell you different things. It depends on what type of game you’re betting on. For example, live betting might be slightly different.

In general, you should be putting somewhere in the range at 75 to 85% of the weight on what the model and the numbers are telling you to do and a much smaller percentage of the weight on your gut intuition.

Do you believe in something like momentum? If you speak to someone who follows analytics strictly it’s something that basically doesn’t exist.

There is analysis you can run in terms of how a team with two wins or three wins performs relative to the spread. There’s a lot you can do to quantify qualitative factors.

I do believe that there is a degree of judgement to use. As long as you understand the baseline of what your quantitative factors are telling you, you can adjust slightly for the qualitative factors.

Where I have a bigger issue — you see it in NFL coaching quite a bit nowadays — is people take the qualitative factors as the primary thing, and then adjust the quantitative factors against it. I would take the opposite approach — take the quantitative factors as your baseline and you can adjust your qualitative a little bit.

Can you tell me about Line Sniper?

Line Sniper focuses on educational materials that teach individuals how to bet on sports bet. We also focus on providing individuals with the data and analysis for the betting decisions themselves. We are rolling out a bunch of interesting features that really haven’t been rivaled by anyone else right now. For example, I’ll be bringing individuals from my own personal network on to the site. These are sports bettors who’ve made over seven figures. They’ll be analyzing games.

Success! You're on the list.

Leave a Reply

%d bloggers like this: