Aaron Kessler is a Las Vegas oddsmaker at the Golden Nugget. Aside from monitoring the movements of lines during the season, Kessler also spearheads the release of the college football ‘Games of the Year’ during the summer.
It began as a fun way to kill time in the dog days of summer—and quite frankly, it still serves that purpose. But the Golden Nugget’s point-spread barrage is now as much as the offseason checklist as the release of spring football games, the release of Phil Steele’s football bible and conference media days.
That signature event only recently came and went. The Golden Nugget released more than 200 point spreads on June 13 (h/t SB Nation) and is currently taking action on said games.
“It's gone from a little novelty with 25 games to a yearly event,” Kessler said. “It's easily our signature event at the Nugget.”
The attention and increased coverage of this yearly release has also prompted more action for the sportsbook. No longer an exercise simply for a handful of diehards, fans—experienced gamblers and novices alike—flock to Vegas each year to be a part of the madness.
“We've gone up 10 times in handle since our first year,” Kessler said. “And it just keeps getting better.”
Handle is a term sportsbooks use to gauge action, or the total amount of money being wagered on the buffet of games available. As coverage of this yearly release has increased thanks in large part to social media—and also the slow acceptance to point spreads at such places like Bleacher Report and elsewhere—the handle (and coverage) will only continue to flourish.
Before the spreads are released, however, they have to be crafted. The process, which will seem somewhat familiar to the experienced gambler, begins by simply deciding on what regular-season college football games deserve to get lines.
This is unique from their regular-season process, of course.
During the season, Week 1 point spreads will be produced for all Week 1 games, Week 1 will end, and then the process will start over with Week 2. Not here, though. It’s up to the Golden Nugget to decide which games it wants to include.
“First, Tony Miller—our sportsbook director—comes up with a list of games to use,” Kessler said on generating the lines. “I look over the list, search for omissions, and we agree on which games to feature and how many to use.”
Some of these decisions are no-brainers. The Iron Bowl, the Red River Rivalry, Alabama-LSU, Ohio State-Michigan and other marquee matchups and rivalries are givens each year. But this list includes much more than the obvious. It's a sign of how much it has grown.
High handle-producing teams such as Notre Dame are always featured prominently, but the Nugget has also added some less obvious matchups to its catalog more recently. For example, Boise State’s 2014 clash with Connecticut was given a line this year. (The Broncos opened as 10-point favorites, in case you were curious or interested in taking sides.)
Once the sportsbook has the games it wants, the work kicks into another gear.
“Then comes the fun part,” Kessler said. “I have to generate a power rating for each team we are booking.”
Power ratings are a lot like the power rankings you dive through weekly. The only difference is these ratings serve a distinct purpose beyond angering fan bases and dousing comment sections in flames.
These ratings allow oddsmakers to consistently handicap teams and games across the entire season, and they are especially integral to this process. Although the numbers for these ratings were not included—the Nugget has to keep some of the "secret sauce" secret, after all—Kessler’s unadjusted computer ratings were as follows.
There are various ways and methods to create these (more on that in a bit), but they are critical for oddsmakers in the line-creation process. Once they are in place, from No. 1 to No. 126, the spreads pretty much fall into the place. You take the larger number (higher-ranked team), subtract the smaller number (the lower-ranked team), add in home-field advantage and you get your spread.
Once this is complete, the numbers are compared to create—you guessed it—more numbers. Only these are the numbers you’re accustomed to.
“I make a list of all the games with computer-only lines in a spreadsheet,” Kessler said. “Then, I'll go through and adjust lines based on my opinion. These adjustments from the raw numbers can be huge.”
While the ratings serve as a guide, Kessler’s opinion looms large in this particular process. Because the season is still off in the horizon, the Golden Nugget has the difficult task of handicapping teams with enormous questions.
How will Everett Golson look in his return to Notre Dame? What kind of impact will Jacob Coker have at Alabama? How will Clemson replace so many key weapons on offense?
These are just a handful of the countless items that can flux a line one direction or another. And while it’s numbers-based with the ratings, the timing of this release requires a personal touch. But it won’t just be Kessler’s.
Once Kessler has the lines just the way he likes them, he’ll initiate the checks and balances. In this instance, that process begins and ends with outside consultant Bruce Marshall.
Marshall has worked in the industry for some time at places such as The Gold Sheet and elsewhere. More importantly, this is someone Kessler trusts and works remarkably well with.
“We literally lock ourselves in a room and don't come out until all lines are done,” Kessler said. “The whole of creating the ‘Games of the Year’ takes about 72 hours from beginning to end.”
Before the lines officially hit the floor, sportsbook director Tony Miller will take one last look and have final say. At that point, the Golden Nugget will open its doors—as it did on June 13—and take real-life wagers on college football games, some of which are more than five months away.
It’s a unique process, although it still follows some very key principles. These principles carry over to the week-to-week grind of regular season oddsmaking.
In-Season Line-Making: Risks, Power Ratings and The Business of Understanding You
At 6 p.m. ET each and every Sunday during the season, the Wynn sportsbook releases its college football point spreads for the week ahead. For those in tune with this process—the individuals pacing below the sportsbook board or those refreshing a webpage from afar—it is a 15-week ritual.
Other online sportsbooks may posts lines leading up to the first Vegas release, although it’s simply not the same. The evaluation process doesn’t truly begin until a Vegas entity is involved. It’s at that point that casino competitors will take note of early movement while gamblers scramble to hit on spreads they like.
“The Wynn is able to take on a little more risk as long as they adjust their lines more aggressively,” Fuhrman said on the release of college football odds. “It’s a copycat system, and that theme stays true through much of the week.”
The closer you get to the actual game, the bigger the bet a sportsbook may take. That doesn’t impact the average gambler, of course, but it does impact the way sportsbooks operate. The big, smart money—the kind of money that can move point spreads quickly—doesn’t usually come until late. Prior to that, a sportsbook might be reluctant to take it simply because the betting market has not matured.
Before you get to the adjusting phase, however, you must have something to adjust. For Todd Fuhrman, his adjusting while at Caesars started well in advance.
“I would set my handicapping number for the following Saturday the night before any game was played during the week prior,” Fuhrman said on the line-creation process. “This way I could look at where the market overreacted or underreacted for one week’s worth of results.”
To do this, Fuhrman would go to his power ratings, the most consistent and misunderstood part of this process. These ratings serve as the key figure of the oddsmaking blueprint, a way to compute (and order) teams against one another using a variety of factors.
Paul Bessire, architect of the popular website The Prediction Machine, is a numbers junkie. With multiple finance degrees, Bessire’s website is built for the gambler, stat freak or a combination of the two.
To predict outcomes of various contests in various sports, Bessire will plug in enormous sums of data and simulate it 50,000 times. He will then report his findings along with a confidence in the consistency across these simulations.
It’s a unique way to hopefully crack the process of oddsmaking, an exercise that seeks value that might not immediately jump out. But, even with his endless streams to crunch data, Bessire appreciates the math that goes into the creation process.
“Even though the names associated with the power ratings are generally similar to the computer polls, they’re basically the opposite,” Bessire said. “Computer polls just look at strength of schedule and wins and losses, power ratings ignore wins and losses. They’re looking at the amount of points put together and margin of victory.”
The Prediction Machine’s current top 10 looks slightly different than Kessler’s, which is nothing out of the ordinary. There are different processes, formulas and ways to compute these figures depending on the situation and what exactly is included. What doesn’t change is that they are updated regularly—weekly, really—which makes the process extremely fluid.
The Prediction Machine
“Everyone seems to have a different version of them, so there’s no one way to do it,” Bessire said. “It’s really an interesting and relatively complex system that takes some decent computing power to pull off, although it used to be far more difficult than it is today.”
The number associated with each team represents a value. Compare that value to another team’s and you’ll have a point spread on a neutral site.
What do these numbers look like? Again, it depends on where you look (and many places would much rather not share their numbers). The Linemakers—as seen on Sporting News—has its power ratings from December 11 of last year available for the public. While these are not current, they do provide a window into what these numbers actually are.
|5 (Tie)||Ohio State||119.5|
|10 (Tie)||Oklahoma State||116|
Take one value, subtract the other, and you've got a spread. Then home-field advantages must be considered, which isn't always as easy as tacking on three points to the home team. This is where extra math is necessary.
Oregon, Alabama and LSU, for example, will be valued much differently at home than most teams. A night game at Baton Rouge could be even handicapped differently than a day game, and thus the complexity of the odds creation process begins to evolve.
There’s also the issue of dealing with teams—Notre Dame, Texas, Alabama and others—that will be bet simply because who they are. These “public” teams will often be treated differently by oddsmakers simply because they are guaranteed to take action regardless of who they play, where they play and when they play.
“Xs and Os really aren’t being taken into account when creating these odds, and that’s not something many people realize,” Fuhrman said. “It’s data, numbers and an understanding of the market.”
Power ratings serve a valuable purpose in the creation process, but there’s far more that goes into it. Understanding behavior, managing risk and determining how the public will react to a given game are integral.
Combine these ingredients, and you have a general sense of what goes into the creation of a single point spread. That and a firm understanding of what your competition is offering.
In the end, sportsbooks are looking to generate equal betting on teams in a given game. If that’s the case—and it rarely is—it can enjoy its cut (the juice) and cash in regardless of who wins. In order for that to happen, however, the point spreads need to be solid, and the bets need to come in as the book anticipates.
And even then, it can all change in a matter of minutes.
Adam Kramer is the College Football National Lead Writer for Bleacher Report. All quotes were obtained firsthand.