How are point spreads made for NFL games? Veteran Vegas oddsmakers explain

LAS VEGAS — Oddsmakers’ opinions of the NFL might be the most pure and honest of anyone around. When they’re wrong, they can’t just throw a jersey in a trash can for social media views and move on.

If oddsmakers are wrong, it can cost their bosses millions of dollars. Or billions, in the case of a Super Bowl.

When you see a point spread for any NFL game, and in particular the Super Bowl, it’s not set in a casual manner. It can’t be. There’s too much money at stake.

“It’s 100 percent to set the right line,” said Lamarr Mitchell, the director of trading at MGM Resorts and BetMGM. “If you don’t, it just becomes gambling for the house.”

Chris Andrews is the sports book manager at the South Point in Las Vegas, and has been setting point spreads since 1979. Every week in the fall and early winter he’ll put up odds on an NFL game and when a sharp bettor confidently wagers on a team, Andrews will wonder if he made a mistake in setting the line.

“They’re going to bet. Once they bet, did I miss something?” Andrews said. “You’re thinking, ‘I missed something because they’re betting $10,000, $20,000.’ They’re not betting for the hell of it. So you go back and look. What did [the bettor] see that I didn’t see?”

Mitchell experiences the same thing. Usually early in an NFL week, often on Monday, he’ll see some of the professional players taking a shot at the lines, looking for edges.

“When you open it initially you will know if it’s the right number based on where the professional players come in on it,” Mitchell said.

Setting the line for the Super Bowl between the San Francisco 49ers and Kansas City Chiefs was an interesting exercise. Oddsmakers started projecting lines for potential Super Bowl matchups before the conference championship games, and the 49ers were going to be a 3-point favorite over the Chiefs. But by the end of the conference championship round, after the Chiefs dominated at the Baltimore Ravens and the 49ers needed a big comeback to get by the Detroit Lions, it opened at 49ers -2.

A lot of Chiefs bets came in — recreational money from casual bettors, said Tristan Davis, a senior trader at BetMGM — and the line went to 49ers -1.5. Then the line bounced back to 49ers -2 after a couple of days due to sharp money, Davis said.

“I think the eyes of the public, they see the Chiefs being so impressive,” Davis said. “We got a couple of sharp guys who saw some value on the 49ers. In my opinion [the opening spread] was just a little overreaction on the Chiefs’ win against Baltimore. And the 49ers going down 17 points to Detroit probably didn’t help their cause either.”

Setting a point spread for 272 NFL regular-season games and 13 playoff games — we’re not even getting into basketball, hockey, preseason football, futures odds or any of the thousands of offerings sports books now have to account for — is a complicated chore with big stakes. After opening lines are set, it becomes a battleground between sharp bettors and sports books, with lines moving throughout the week.

Casual bettors walk up to a window or open their app and see the home team as a 3-point favorite. They don’t often think about what goes into setting that line, or how being off by even a half-point can have serious ramifications for the house.

“You have to be careful when you do this type of stuff,” said Davis, who has been in the oddsmaking business for 20 years. “If you put a decimal point in the wrong place in the sports betting industry, you can be in trouble.”

How do oddsmakers set a point spread?

Andrews sits at his desk, in an office just to the left of the sports book that has a huge odds board listing all the point spreads and odds for games that day and week. In the side office there are more than 20 television screens, and Andrews has multiple computer monitors showing point spreads from numerous sites and sports books, as well as how much money has been bet on each team that week at the South Point. It’s important to track it all.

Andrews sets the lines at South Point and it’s common for sports books to have someone personally do it, he said, because algorithms and other programs can’t take into account all of the variables such as the feel over which side bettors will be wagering on.

It’s coming up on Week 15 of the NFL season in mid-December and the game of the week is the Dallas Cowboys at the Buffalo Bills. As Andrews sits in his office, roughly 48 hours before kickoff, the Bills are favored by 2 points. That means Bills bettors need Buffalo to win by 3 or more, Cowboys bettors would win with a 1-point loss, a tie or a Dallas win by any margin, and everyone’s bets would be refunded with a 2-point Bills win. Practically speaking, this point spread gives the Bills a 2-point head start.

The process for setting NFL spreads begins on Sundays before that week’s games are even over. Andrews — who is in charge of NFL and college football lines at South Point — starts working on college football while the NFL’s late set of games is going on. After four hours of that, he’ll take a short break and then shift to the NFL. He’ll have the next week’s NFL lines up by about halftime of the Sunday night game.

For Cowboys-Bills and any football game, setting a line starts with an oddsmaker’s power ratings. Andrews uses a system in which his best team will be at about 100 and everyone goes down from there. Before Week 15, Andrews said he had moved a hot San Francisco 49ers team up to 104. Andrews said others, like famous sports bettor Billy Walters, have their average team at zero and every other team above or below that. The power ratings shift every week and take into account variables like injuries and a significant change in how a team is performing. The difference in opposing teams’ power ratings, once home-field advantage and any other unusual variable is accounted for — weather can be one, though Andrews said he usually worries about only high winds and sometimes extreme conditions like a blizzard or frigid cold — is the baseline for the opening point spread.

The first spread Andrews comes to for an NFL game is simple math, using the power ratings: If Team A is 90, Team B is 91 and at home with a 2.5-point home-field advantage, the line is Team B -3.5. Andrews said he keeps separate home-field advantage numbers for each team, and it ranges from a half-point to 3 points. The standard used to be a 3-point edge for home field but that has dropped to 1.5 or 2 points.

“Power ratings give you a place to start,” Andrews said.

There’s more to consider before that spread goes on the big board and bettors can wager on it. The biggest concern is how bettors will react to a spread. Davis says that so-called recreational money from casual bettors “doesn’t change my opinion one iota” on changing a point spread, he knows of five to 10 sharp customers who he has respect for and their bets can move the line a half-point or even a full point in some cases.

“Even though Joe Public likes to bet the big favorite teams, we have to make sure the number is right so our professional players don’t hammer us and we’re off in that market,” Mitchell said.

Back in December, Andrews’ pure power rating on Cowboys-Bills was Dallas -1. But that didn’t feel right. Andrews has to account for all of the multiple variables, especially how bettors will bet the game. A spread of Dallas -1 would get too much money on the Bills as a home favorite, considering the Bills were coming off a big win at the Kansas City Chiefs.

“I thought, especially the way they’re playing lately, Buffalo deserved to be a small favorite,” Andrews said.

He opened the game at Bills -1.5. But the job of setting a line was just starting.

Why do point spreads change?

In an ideal world for oddsmakers, a point spread wouldn’t move. That would mean they set a perfect, fair line that bettors are 50/50 on. That rarely happens, as point spreads move often during a football week. If oddsmakers didn’t react and change lines based on bets, they’d get too much action on one side. They’d rather be close to an even split.

“If it’s 50/50, we’re going to win either way,” Mitchell said.

Another reason oddsmakers don’t like moving lines much is the fear of being “middled.” If a majority of bettors take the favorite at -2, the line moves and bettors take the underdog at +4, then the game ends up in a 3-point win for the favorite, the sports books lose both ways. That’s a disaster for the house. That’s a reason sports books are very careful moving a game through key numbers like 3, 4, 6, 7 or 10, the most common winning margins in the NFL.

The most famous example of sports books being middled was Super Bowl XIII between the Pittsburgh Steelers and Dallas Cowboys. The Steelers opened as 3.5-point favorites that got bet up to -4.5. Then bettors started taking the Cowboys +4.5. When the game ended with a 35-31 Steelers win, all of the Steelers -3.5 and Cowboys +4.5 tickets cashed, a disaster for sports books.

Bettors took Buffalo right away in Week 15 and Andrews moved the South Point’s line to Bills -2. A bet from a familiar, sharp customer can cause Andrews to immediately move the line. And Andrews said the amount of money being bet doesn’t always matter.

“Once you open a number, it becomes more of an art because you have to see who is betting it. It’s not just pure money,” Andrews said. “There are a lot of guys out there with a lot of money and, in all honesty, don’t know what they’re doing.”

He told of a customer he had back when he was setting lines in Reno, who would bet only $1,000-$2,000 on a game.

“Everyone knew he was really sharp. Everyone paid attention to him,” Andrews said. “He didn’t have a fancy car or anything like that, it’s just how he wanted to live his life. But he was really, really sharp. I followed him.”

When sharp bettors take a side, the line will often move. There are also considerations like injury reports (certain NFL players will have a point value attached to them, with elite quarterbacks usually worth about 6 or 7 points on the line), which are harder to track in college and can leave a sports book exposed.

“In college, there are 130 teams,” Andrews said. “We’ll probably know about Michigan, Ohio State and Alabama. But what is going on with Tulsa? And that’s all week.”

The closing line for Cowboys-Bills was Buffalo -2.5. The Bills won 31-10, easily covering the spread.

Andrews has to also track what other sports books are doing. That’s a reason he has all those computer monitors in his office. For a Denver Broncos-Detroit Lions game in Week 15, he had the Lions rated as 5.5-point favorites. Most other sports books had the game at 4.5, and he knew that setting it at Lions -5.5 would lead to many Broncos bets and he’d have to move the line anyway. He opened it at Detroit -5. Every half-point matters.

“I don’t want to give them that extra half-point if I don’t have to,” Andrews said. “So they took the 5, and win or lose that’s the right move on my part.”

The Lions won, 42-17.

Other games in Week 15 had line movement that Andrews wasn’t expecting. Andrews opened the Tennessee Titans as 2-point favorites over the Houston Texans, whom he anticipated would be without quarterback C.J. Stroud due to a concussion. The game was bet all the way up to Titans -3.5. The line moved due to an unexpected number of bets coming in on Tennessee. Sports books try to avoid having too much money on one side.

“I kind of underestimated what the public would perceive,” Andrews said. “I thought 2 was OK. But we’re up to 3.5, so evidently I wasn’t.”

The Texans won, 19-16.

Even after more than 40 years in the business, Andrews can’t always predict what bettors will do.

Lamenting mistakes setting lines

Andrews lamented what he considered a big mistake on a New York Jets-Miami Dolphins spread for Week 15. Due to the race to get the next week’s spreads up, the lines for Week 15 were on the board before the Dolphins kicked off a Monday night game against the Titans. The Dolphins were 12.5-point favorites vs. the Jets and sharp bettors took the Jets before the line was taken down at kickoff. During the Titans-Dolphins game, Tyreek Hill suffered an ankle injury, the Dolphins eventually lost and when the line went back up, it was Dolphins -9 vs. the Jets. That’s too big of a difference.

“That one, in all honesty, I probably f***ed up,” Andrews said. “I probably put that up too high.”

The Dolphins beat the Jets 30-0 but that wasn’t the most important thing. The process is. When the line changes too much during the week, something went wrong and Andrews will worry he made a mistake.

“Yeah, if there’s a good reason that I missed,” Andrews said. “I’m only speaking for myself, I’m self-critical and I think you need to be that way to be successful. You can’t figure you know everything. You just can’t. Because first of all, you don’t. If there’s something you missed, you missed it. You try to learn from it and not do it again.”

There might be millions at stake in the betting world for a normal NFL game, and that turns into billions for the Super Bowl. If the point spread is off by just a bit, sharp bettors will catch on. That’s why oddsmakers spend so much time getting it right. They can’t afford to be wrong.