The Craft of Sports Betting Professionals

by Ed Feng, founder of The Power Rank

As a sports bettor, you’re always looking to improve your process. Instead of picks from those oh so honest touts, you want to learn from the professionals, the best of the best.

Or maybe you have this voyeuristic need to see how the best operate. You want to know what it takes to climb the mountain of sports betting success (hint: it’s a lot of work).

To help you along this journey, I’ve listened through hours of podcast interviews to find the best. This document distills the essence of these interviews with long time sports betting professionals:

  • AbnormallyDist – academic turned professional sports bettor
  • Haralabob Voulgaris – NBA betting legend
  • Alan Boston – college basketball betting legend
  • Rufus Peabody – quantitative bettor known for golf and Super Bowl props

Hard work is an essential component to thrive in any field. That’s as obvious as the horned, three-eyed demons that haunt Vegas Dave’s sleep every night. While these professional bettors are no different, these summaries give a sense for how they work smart as well.

While the methods of these professionals differ, their beliefs converge in certain aspects of betting. One common thread is closing line value.

Let’s take a look.

Abnormally Dist

AbnormallyDist (AD) had a cushy academic job waiting for him in August. After earning a Ph.D. in machine learning, he expected to make a career in the halls of academia.

However, AD had been messing with sports betting models, and he couldn’t resist going to Las Vegas for one last trip before his job started. During the fourth week of the NFL preseason, he bet teasers and won big.

AD still went to the first few days of his academic job. However, he couldn’t stand the endless hours of meetings. He quit and became a professional sports bettor.

From AD’s appearance on the Circles Off podcast, his care and attention to detail stand out. Let’s look at two examples.

First, he emphasizes that winning at sports betting is not a matter of running numbers and betting based on those results. He said that he runs his golf numbers on Sunday night, and then he puts in eight hours on Monday to find the bets that he likes.

AD has a similar process for the NFL. While he doesn’t think he can beat the markets on Sunday morning hours before kickoff, he likes his chances early in the week. He talked about the eight hours he works on his NFL bets on Sunday, and then forgets about them for the rest of the week.

Second, he had a nuanced take on closing line value. Instead of asking only whether you get CLV, he instead looks to understand why markets move.

The Circles Off podcast was released in April of 2021, and back then, the golf markets opened on Sunday. A day later, the influential site Data Golf released its predictions, which has substantial market influence.

On Tuesday, Bookmaker opens its golf markets. Because of their high limits, the market immediately shifts towards the odds at Bookmaker.

Because of this sequence of events, AD had a process for making his bets. If his bet agrees with Data Golf, he tries to make it early. If a bet doesn’t agree with Data Golf, he waits because he might get a better price.

AD also had some interesting thoughts on how making a model for player props can help with game results. He told a story of how he bet on the Jets as a huge underdog against the Ravens. Part of his analysis was how well the Jets defense defended runs to the right, and the Ravens often ran to this side.

However, he missed one thing. The best Ravens running back is QB Lamar Jackson. The Ravens ran Jackson left, and the QB racked up 100 rushing yards in the first quarter. Had AD remembered about Jackson as a runner, he might not have bet on the Jets. He did realize his mistake before kickoff, and he compensated for this error by betting on Jackson’s rushing props.

This is probably my favorite sports betting podcast ever, as AD discussed a number of other fascinating topics:

  • Picks versus learning the process behind finding good bets
  • Crypto, real estate and how he invests his money
  • How and why he moves the smallest golf markets

Abnormally Distributed on the Circles Off podcast.

Haralabos Voulgaris

While working at an airport in his twenties, Haralabob (Bob) Voulgaris put down an NBA futures bet. As described in Nate Silver’s book The Signal and the Noise, he wagered about $80k on the Los Angeles Lakers to win the NBA championship in 2000.

The Lakers, led by Shaquille O’Neal and Kobe Bryant, had a Game 7 against Portland in the Western Conference finals, and Portland had a 16 point lead with two minutes left in the 3rd quarter. The Lakers future seemed doomed. 

However, the Lakers stormed back to win the game, capped by the iconic alley-oop dunk from Bryant to O’Neal. Voulgaris now had a bankroll, and he never looked back as a professional bettor.

Bob started out as a mostly subjective bettor that consulted data to confirm his hunches. On an episode of Be Better Bettors, he talks about how he watched NBA games early in his career to pick up on coaching tendencies. 

He also originated CFL bets, and he looked at different fields in Canada to get an edge. It’s easier to score a touchdown in a rectangular end zone that is 20 yards deep than a smaller end zone in other stadiums, so this impacted his bets on totals.

As Bob progressed as a bettor, he focused on the NBA and hired a quant to build a predictive model. At this point, he used his model to originate bets, but he had no problem overriding the model based on his subjective analysis. Bob watched games every night during the NBA season.

Later in his career, the model got better with each passing year, and Bob felt more burned out from the daily grind of betting the NBA. At this point, he originated bets solely from the model.

There were three other points in the Voulgaris interview that stood out to me.

First, Bob talked about getting accounts. Ego was not the reason he let ESPN write a feature article on him (ok, ego probably played a role). He let ESPN publish this article to get accounts, and he obtained one of his largest accounts from someone who read this article.

Bob talked about getting beards, or famous people who could place large bets due to their historical ineptness at sports betting. He would play poker to meet these types of people. In a typical arrangement, the beard would get a free roll. Bob puts up the money and covers losses, and the beard gets 20% of the profits.

Second, Bob told a cautionary tale about closing line value. During one NBA season, he had a beard that could bet $500k on NBA games with a bookie right before tip off. With such a large bet size, Bob wanted to keep this account for as long as possible. 

To seem like a square to this bookie, Bob would bet on games the other way to move the market. For example, suppose he bet $500k with the bookie on Lakers -3.5 against the Clippers. He would bet Clippers +3.5, and his stature in the market would move this to Lakers -3 by tip off. This made his $500k bet look square.

A casual observer might view Clippers +3.5 as the sharp side, but the professional Voulgaris would disagree.

Third, Bob told one story from his time working for the Dallas Mavericks. After a string of frustrating losses, one assistant coach would go into a rage. This hormonal male showed his anger by throwing chairs. One day, this assistant asked Bob why these losses didn’t make him more upset.

Bob replied that if he thought getting mad would help the Mavericks win, he would be the most upset person in the organization. This story reflects his passion for stoic philosophy and also the necessary criteria that a pro bettor needs to take a long term view and not let variance dictate one’s actions.

Haralabob Voulgaris on Be Better Bettors.

Smarter sports betting in under 5 minutes

The Power Rank’s sports betting email newsletter discusses why I make certain NFL and college football bets. In addition, 7-Nugget Saturday is a curated list of sports betting tips and news.

  • Valuable
  • Concise
  • Entertaining

These are the goals of each correspondence.

As a bonus for signing up, you’ll get a free download of the book The Logic of Sports Betting, by Ed Miller and Matthew Davidow.

To get these free resources, enter your best email and click on “Sign up now!”








 

 
 

Alan Boston

Alan Boston is a college basketball betting legend. His handicapping was the best part of the 2001 book The Odds by Chad Millman, and he’s still betting college hoops today.

Alan does this without analytics. Or so he said on The Football Analytics Show podcast.

That might seem like showing up to Bet Bash in a medieval Renaissance jester outfit. However, a deeper contemplation of Alan Boston’s words reveals that analytics has its fingerprints all over his work.

For example, Alan talked about how a team tends to perform at the same level every year. The same coach has the same resources, which usually means recruiting the same caliber of players.

To Alan, this is logic. To a more numbers oriented analyst, you can look at this through data.

The Power Rank’s college basketball team rankings take margin of victory and adjust for schedule. Each team gets a rating, which gives an expected margin of victory (point spread) against an average team on a neutral site. This rating leads to game predictions and tournament win probabilities.

In the last two seasons not impacted by COVID, the R-squared for team rating in consecutive seasons is 0.687. College basketball teams tend to persist from season to season, which supports Alan’s logic of how teams perform at roughly the same level every season.

As another example, consider home court advantage. Alan has talked about how its value has declined over the years. 

You can show this decline by digging into the data. To calculate home court, I look at the margin of victory for the home teams over the past six seasons. For the 2021-22 season, I used a home court of 2.9 points, down from 3.5 points ten years ago.

Alan has his own way of determining this. A few years ago, he realized a large fraction of his bets on home teams were losing. That made him go back and lower his values for home court, which depends on the venue.

Alan also does not strive for closing line value. If the market moves in his direction 90% of the time, that means that he is doing the same thing as other people. Alan strives to think differently. As he said on The Football Analytics Show, “greatest doesn’t come from copying.”

While Alan is most well known for betting on college basketball, he has also handicapped college football. On an episode of Be Better Bettors, he talked about his college football work with the legendary Bill Walters.

At first, Alan didn’t think he wanted to work with Billy based on his reputation. At their first meeting, Alan called Billy a condescending asshole. Billy didn’t flinch.

Despite this awkward start, the two did come up with an agreement. Billy paid Alan a $12k salary for the college football season and let him bet $12k on each of Alan’s plays.

You can catch Alan Boston here:

Rufus Peabody

As Mark DeRosa walked to his car, he noticed a guy was following him. This guy with blond hair didn’t look that threatening, but he did seem to be getting closer. It also seemed creepy that Mark, a professional sports bettor, had seen this guy in sports books around Las Vegas. Mark put his hand on the mace he carried.

This encounter was the first time Rufus Peabody talked with his future betting partner Mark. Rufus has an economics degree from Yale, and he came out to the desert after college to work for Las Vegas Sports Consultants, the legendary group started by Roxy Roxborough.

Rufus then decided to join this sports betting group with Mark and others like Zach White. In the second of two episodes of the Risk of Ruin podcast on this group, Rufus talked about how he provided a quantitative edge for this group. In particular, two stories stand out.

First, Rufus scraped a bunch of weather data at the airport closest to MLB stadiums. He wanted to determine whether the wind blew out towards the outfield. An outward wind suggested betting the over.

At first, the model performed well. But then it started to lose. In addition, Rufus caught some differences between the data he scraped from the airports and the reports from live broadcasts. He soon realized the model was bogus because winds can swirl near stadiums. A reading at the airport does not reflect the actual wind in a stadium.

While this predictive model showed promise before revealing its true nature, Rufus also had the opposite experience. He developed a college football second half model, and his back testing suggested value. When he started betting, the model got crushed for the first few weeks. His betting partners lost interest in the plays the model suggested.

Rufus then moved to Costa Rica with his girlfriend. Despite what his partners thought, he continued to bet with his college football model. It started to win, and his betting partners wanted back in.

In more recent years, Rufus has focused more on Super Bowl props and golf.

On an episode of The Football Analytics Show, he talked about football player props. He starts by projecting game level quantities such as number of pass attempts, and he discusses how these projections depend on game state (ahead or behind in the game).

Then Rufus digs into each player and projects quantities such as number of routes and target percentage. Then he looks at the distribution for receiving yards. In other words, the probability that Chris Godwin gains zero yards, the probability he gains one yard, and so on.

It’s important to look at this distribution in football, as the mean and median are not the same. For example, an explosive pass play of 80 yards can raise a player’s average receiving yards. However, the market value is the median value of receiving yards, as the price to go over and under are about the same. 

The take home message is this: the average projection for yards is larger than the market value. Peabody accounts for this discrepancy when he bets Super Bowl props.

On an episode of the Establish The Run podcast, Rufus talked about his golf calculations. While he didn’t say a word about his player ratings, Rufus did reveal some of the details he considers in his projections. For example:

  • A golfer might perform differently when he plays in a top group over the weekend. The top golfers tee off in the afternoon when the conditions tend to be worse.
  • The course matters, but do not double count. For example, a long ball hitter should perform well at the Masters. Rufus accounts for this strength off the tee before calculating course fit at Augusta National.

To listen to Rufus Peabody:

Smarter sports betting in under 5 minutes

The Power Rank’s sports betting email newsletter discusses why I make certain NFL and college football bets. In addition, 7-Nugget Saturday is a curated list of sports betting tips and news.

  • Valuable
  • Concise
  • Entertaining

These are the goals of each correspondence.

As a bonus for signing up, you’ll get a free download of the book The Logic of Sports Betting, by Ed Miller and Matthew Davidow.

To get these free resources, enter your best email and click on “Sign up now!”