A points prop is a minutes projection with a usage rate attached. The book posts a line that looks like a season average, but the season average is the wrong number almost every night, because tonight’s minutes are not the average and tonight’s role is not the average either. The bettors who beat NBA props are the ones who read the injury report at 5 pm and 1 pm and work out who plays how much before the line catches up.
Minutes Come First
Every counting stat is a rate times a number of minutes. In 2024-25 Tyrese Maxey led the league at 37.7 minutes a game, Kevin Durant was second at 36.5, and three players tied at 36.1. Very few players see 36 minutes and most starters land in the low 30s. The difference between a 34 minute night and a 28 minute night is almost 20% of a player’s production, and the prop line rarely moves 20% to match.
What moves minutes is predictable. A blowout cuts them, because starters sit the fourth quarter. A close game adds them. The second night of a back to back trims them for older players and for anyone on a minutes restriction. A teammate’s absence adds them for whoever slides into that slot. None of this is secret. It is just work the book does not do on every player every night.
Usage Is the Second Number
Usage is the share of a team’s possessions a player finishes with a shot, a free throw trip or a turnover while he is on the floor. It is sticky when the lineup is stable and it jumps when the lineup is not. The night the first option sits, the second option’s usage climbs, and so does the third option’s, and the book has to guess how the pie gets split. Those guesses are the softest lines on the board.
The simple version: a player whose usage climbs from 22% to 28% with a teammate out is producing more per minute, and if he is also playing more minutes, the two compound. A points line set off the season average can be short by several points in that spot. The same logic runs the other way when a star returns and everyone else’s touches shrink.
Reading the Injury Report Like a Trader
The NBA requires teams to file their injury report by 5 pm local time the day before a game. For the second night of a back to back the deadline is 1 pm on the day of the game. There is a further game day report between 11 am and 1 pm, and for tipoffs at 5 pm or earlier the game day window is 8 am to 10 am. Reports update through the day as statuses change.
Those deadlines are the schedule you bet to. The 5 pm report is where a player first appears as questionable and the book has to price a lineup it is not sure of. The 1 pm report is where a back to back rest night gets announced and every prop on that roster moves in the same minute. The gap between a report posting and a book repricing every affected prop is where the value is, and it is measured in minutes, not hours.
The statuses are not all the same. A player listed out is out. A player listed questionable is a coin flip the book has to price as a coin flip, which means the props on his teammates are set halfway between two very different nights. If you have a read on which way it goes, from a beat reporter or from how the team has handled that player all season, the teammates’ props are where that read pays.
Why the Season Average Is the Wrong Number
The season average blends the games a player started with the games he came off the bench, the games with the full roster and the games without, the games against a team that fouls and the games against one that does not. The book’s model does better than the raw average, but it still leans on it, and the raw average drags the line toward a role the player may not have anymore.
The better inputs are the last ten games with the current rotation, the minutes the coach has given him in that role, and the matchup. A big who gets his points at the rim against a team that protects the rim is not the same bet as the same big against a team playing small. A guard against a team that switches everything is not the same bet as the same guard against a team that drops its center.
The honest limit is that nobody beats a points line by enough to overcome heavy juice. Props at -115 both ways need 53.5% to break even. A prop at -120 both ways needs 54.5%. Shop the price as hard as the number.
Scratches, Voids and the Prop Restrictions
At nearly every book a prop on a player who is listed inactive and never plays is voided and your money comes back. If he plays at all, even a few minutes before leaving hurt, the bet stands. That rule cuts both ways. Betting an under because you expect a late scratch gets you a refund, not a win.
The market itself has tightened. The NBA and a group of major sportsbooks have agreed to stop offering parlays made of multiple unders on the same player, “first of” bets on who takes the first shot or wins the opening tip, and unders on players expected to average fewer than 12 minutes. That followed limits already in place on 10 day and two way contract players and on turnover, foul and missed free throw props. Those changes came out of regulated US books. The offshore books we cover set their own menus and their own rules, and they are outside US state regulation, so check what each one offers rather than assuming the US list applies.
Where to Bet NBA Props
Depth and speed are what you want. Bovada has the cleanest interface of the five for shopping a prop board quickly after a report posts, with the honest caveat that it limits winning players and prop winners get noticed first. BetOnline posts early and carries a wide market selection, though its interface is dated. Whether either is available to you depends on where you live. See our sportsbook reviews for how each book handles prop rules and limits.