Why the half-point matters more than the number
A projection isn't a prediction of what a player will do. It's the number that splits opinion evenly — and the .5 on the end is doing more work than the digits in front of it.
New players read a projection the way they'd read a weather forecast: someone smart has estimated what will happen, and the number is their best guess. That's not what it is. A projection is the number that splits opinion most evenly — the line where roughly as many people want More as want Less.
That distinction changes how you use it.
The number is a balance point, not a forecast
If a line were a pure prediction, beating it would mean out-forecasting whoever set it. It isn't, so you don't have to. You have to find the cases where the balance is off — where the number reflects what people expect rather than what the matchup actually supports.
Those two things agree most of the time. The edge lives in the gap when they don't: a role change nobody has priced in, a matchup that suppresses one stat while leaving another alone, a player whose usage moved for a reason that hasn't shown up in the box score yet.
Why every line ends in .5
The half-point is there to remove ties. A line of 24.5 points has exactly two outcomes — 25 or more, or 24 or fewer. A line of 24 has three, because the player can land exactly on it.
That third outcome is a Push, and a Push is not a win. It drops out of the lineup and the lineup settles on what's left. So the .5 isn't decoration; it's the difference between a pick that resolves cleanly and one that can vanish from your lineup after you've already committed to it.
The line you took is the line you keep
Projections move. New information arrives, a lineup is announced, someone is ruled out, and the number shifts before tip-off. What matters is that the line attached to your confirmed pick is the one used to settle it. If it moves afterwards, your pick doesn't move with it.
This cuts both ways, and it's the most practical thing on this page. If you like a number, taking it locks it in. If you're waiting for better information, you're also accepting that the number may not be there when you come back.
What to actually look at
- Whether the projection assumes a role the player still has — minutes, batting order, set-piece duty.
- Whether the matchup suppresses this specific stat, or just the team's output generally.
- Whether recent form is a trend or three games of noise.
- How far the number would have to be wrong for you to be right. Two points off is a different bet from six.
The question worth asking
Not "will this player have a good game?" but "is this specific number in the wrong place?" Those sound similar and lead to completely different picks. A star can have an excellent night and still finish under a line set for a star having an excellent night.
The second question is harder and much more useful. It's also the only one the line is actually asking you.
Go and look at a line.
The half-point is easier to think about with a real number in front of you. Tonight's board is up.