NBA alternate lines are the same player or game market as the main line, set at a different number: over 19.5 points at a short price, over 29.5 at a long one. A milestone like 20+ points is simply an alternate over (it wins exactly when over 19.5 wins). To judge one, remove the margin from the main over/under to find the player's fair centre, decide how widely their games spread around it, and read off the chance of reaching the milestone. Close to the main line that gives a solid answer. Far from it, the answer depends mostly on your assumption, and that's where most of the risk sits.
The 2026-27 NBA season tipped off on Tuesday 20 October, so the prop boards are full again. Everything below uses invented numbers and shows a method, not a view on any player or game.
Alternate lines and milestones at a glance
| What you see | What it really is | How it's priced |
|---|---|---|
| Over/under 22.5 points | The main line | Two sides, usually close to even money |
| Over 19.5 points | An alternate over | Shorter, because it wins more often |
| 20+ points | The same bet as over 19.5 | Often listed on its own, with no "under" |
| Over 29.5 / 30+ points | An alternate over a long way from the main line | Long price, and the hardest to judge |
US sites call a run of these "laddering": you take the standard over, then step up to bigger milestones that pay more. Action Network's example from April 2026 pairs over 20.5 points at -123 with 25+ points at +276. In Australian decimal odds that's about 1.81 and 3.76. Each rung is a separate bet with its own price, and each one has to stand up by itself.
Why you can't de-vig a milestone directly
On a two-way line, you can see the bookmaker's margin. Over 22.5 at 1.90 and under 22.5 at 1.90 imply 52.6% each, which adds up to 105.3%. Scale both back to 100% and each side's fair price is 2.00. Our NBA odds guide walks through that step, and the vig explainer covers where the margin comes from.
A milestone usually has no other side. "20+ points at 1.40" doesn't come with "19 or fewer at …", so there's nothing to add up and nothing to scale back. The margin is still in the price; you just can't see it. The only way to judge the price is to build a fair price yourself, and the best place to start is the main line, because that's the market with the most money and attention on it.
Step 1: find the fair centre from the main line
Take the main line and remove the margin. If over and under 22.5 are both 1.90, the fair chance of each side is 50%. That tells you the market's middle for this player is about 22.5 points: half the time more, half the time fewer.
If the two sides aren't equal, the fair chance of the over won't be 50%, and the centre shifts a little above or below the line. The method still works; you just start from that fair percentage instead.
Step 2: choose a shape for the player's output
The main line tells you the middle. It says nothing about how spread out the player's games are. Two players with the same 22.5 line can have very different nights: one scores 20 to 25 almost every game, the other swings between 12 and 35.
For points, a simple choice is a normal curve, the bell curve that is symmetric about its mean and puts about 68% of results within one standard deviation (SD). You pick the SD, which is the size of a typical swing from the middle. A rough guide is the spread of the player's recent games, but treat a 10-game sample with care: our post on NBA player prop hit rates shows how little a short run tells you.
Because points are whole numbers, "20+" means "over 19.5", so you measure from 19.5, not 20.
Worked example: a points ladder
Invented numbers. Main line: over/under 22.5 at 1.90 each, so the fair centre is 22.5. A bookmaker prices three milestones: 20+ at 1.40, 25+ at 2.50 and 30+ at 7.00.
Here's the fair price of each milestone under three different SD assumptions:
| Assumed SD | 20+ fair price | 25+ fair price | 30+ fair price |
|---|---|---|---|
| 5 points | 1.38 (72.6%) | 2.90 (34.5%) | 12.38 (8.1%) |
| 6 points | 1.45 (69.2%) | 2.71 (36.9%) | 8.22 (12.2%) |
| 7 points | 1.50 (66.6%) | 2.58 (38.8%) | 6.30 (15.9%) |
And the expected value of each price on the board (fair chance × price, minus 1):
| Assumed SD | 20+ at 1.40 | 25+ at 2.50 | 30+ at 7.00 |
|---|---|---|---|
| 5 points | +1.6% | -13.9% | -43.5% |
| 6 points | -3.2% | -7.6% | -14.8% |
| 7 points | -6.8% | -3.1% | +11.1% |
What the table shows
- Near the main line, the assumption barely matters. The fair price of 20+ only moves from 1.38 to 1.50 across all three SDs. You can judge it with some confidence.
- Far from the main line, the assumption is everything. The fair price of 30+ runs from 6.30 to 12.38. The same 7.00 looks like a 43% loser or an 11% edge depending on one number you chose.
- The middle rung was below fair in every case. At 2.50, 25+ came out negative whichever SD we used. When a price fails under every reasonable assumption, that's a clearer signal than one that only works under one.
None of this says a bet will win. A positive expected value is an estimate that only shows up, if at all, over many bets, and any single milestone can miss.
Worked example: threes
Made threes are small whole-number counts, so a Poisson model is a common starting point. It has a single number, the average, which is equal to both its expected value and its variance. Once you know the average, the chance of every count follows.
Invented numbers. Main line: over 2.5 threes at 2.00, under at 1.80. That market adds up to 105.6%, so the fair chance of the over is 47.4% (fair price 2.11). The Poisson average that gives exactly 47.4% for 3 or more is about 2.57 threes. From that:
| Milestone | Fair chance | Fair price | Bookmaker price | Expected value |
|---|---|---|---|---|
| 2+ threes | 72.6% | 1.38 | 1.40 | +1.7% |
| 3+ threes (the main over) | 47.4% | 2.11 | 2.00 | -5.3% (the margin) |
| 4+ threes | 25.7% | 3.89 | 3.60 | -7.4% |
| 5+ threes | 11.8% | 8.46 | 7.50 | -11.4% |
The same pattern appears. The rung next to the main line is close to fair. The longer rungs are where the price drifts further below the model.
Where both models are too tidy
Real basketball output is lumpier than these models. Statisticians call it overdispersion: more variation in the results than the simple model expects. A 2023 study modelling NBA scoring by shot type started with Poisson counts and found a negative binomial model, which has a second number for the extra spread, predicted results more accurately because the variance was larger than the mean.
For milestones that matters most in the tails. A model that's too tidy understates how often a player has a huge night, so it makes long milestones look worse than they might be. That's why the 30+ and 5+ rows are the least reliable numbers in this post. Points also can't go below zero, so a symmetric bell curve is a simplification at the low end. Treat both models as a way to see the shape of the problem, not as a price you can trust to the decimal.
Two checks that need no model at all
You don't always need a model to spot a bad price.
- Compare the same milestone across bookmakers. If one book has 20+ at 1.40 and another has it at 1.50, the 1.50 is the better price for the same bet, whatever the true chance is. An odds screen shows these side by side.
- Check the ladder makes sense. Scoring 20+ always happens at least as often as scoring 23+, so at any bookmaker, the 20+ price should be shorter than the over 22.5 price. If a milestone is priced longer than a harder line at the same book, one of the two prices is stale or wrong.
Rules to check before you bet a milestone
- If the player doesn't play. The usual rule is that props on a player who never takes the court are void, while a player injured during the game means the bet stands. Rules differ between bookmakers, so read your bookmaker's terms.
- Minutes matter more than the line. A blowout that rests a player in the fourth quarter hits long milestones hardest, because they need the most time on court.
- Ladders multiply your stakes, not your edge. Three rungs means three bets, and three margins. Spreading money across a ladder doesn't make any rung better value.
- Record the price you took. Comparing it with where the milestone closed, your closing line value, is the honest test of whether your fair prices were any good.
A quick checklist
- Remove the margin from the main over/under to find the fair centre.
- Turn the milestone into an over: 20+ is over 19.5.
- Pick a spread (SD for points, the average for threes) and work out the fair chance.
- Try a smaller and larger spread too. If the verdict flips, the price is too close to call.
- Compare the same milestone across bookmakers before anything else.
- Check the ladder is in order, and check the bookmaker's void rules.
Our guide to prop betting covers the wider market, and the calculators turn any price into an implied chance in seconds.
See NBA props side by side
BetBank's NBA board lines up head to head, line and total prices, plus points, rebounds, assists, threes and combo props, across Australian bookmakers, with hit rates and no-vig fair percentages where available. That gives you the main line to start from and every bookmaker's price on the same rung. For the new season, first-time members on a weekly or monthly Silver or Gold plan pay half price for their first 2 months after the 7-day free trial, if they start by 30 November 2026. See how the NBA season special works.
BetBank is for adults 18+. Betting involves risk, and no model or tool removes it. Set a limit, take breaks, and gamble responsibly. If it stops being fun, our responsible gambling page lists free, confidential support, and BetStop lets you block yourself from all licensed Australian online and phone betting.