AI basketball betting prompts for pace, spreads and totals
Basketball is the sport where a projection actually works: possessions times efficiency gets you close to a real total. It is also the sport where a single injury report published an hour before tip-off can make that projection obsolete.
What a basketball prompt has to get right
Ask a model for "who wins" and you get a vibe. Ask it to project each team's points from pace and efficiency and you get a number — projected total, projected margin — that you can hold against the bookmaker's line. That single change of framing is the difference between an entertaining answer and a testable one, and it is why the v1 prompt below is written as a projection task rather than a prediction task.
The second prompt exists because basketball lines move on news, not on numbers. Load management, a star listed as questionable, the second night of a back-to-back after a flight across time zones — these are the situations where the market is briefly slow and where a prompt told to prioritise availability over season averages earns its keep.
Pace times efficiency
Possessions per 48 minutes for both teams, offensive and defensive rating. Multiply, adjust for home court, and you have a projected score. Compare that to the posted total instead of guessing over or under.
Availability above all
Injury and load-management reports land late and matter enormously. Give the model the current status list and tell it to say the pick is void if a listed star is ruled out after the answer.
Rest, back-to-backs and travel
Second night of a back-to-back, third game in four nights, a long flight and a time-zone change. Fatigue shows up first in defensive rating and in fourth-quarter margin.
Recent rotation, not the season
A trade or a lineup change makes the season averages describe a team that no longer exists. Feed last-10 numbers alongside the season line so the model can see the drift.
Basketball prompts v1 and v2 — and how they differ
The same model, two instruction sets, two different betting personalities. Run both on the same matches; that comparison is the only thing that settles the argument.
| Version | Focus | Style | Best for |
|---|---|---|---|
| v1 | Projected score from possessions and ratings | Model-anchored | Totals, spreads |
| v2 | Availability and schedule spots | Situational | Injury-news and schedule value |
You are a basketball betting analyst. Game: {home} vs {away}, {league}, {date}. Line: {odds}.
Project each team's points as pace (possessions per 48) multiplied by offensive efficiency against the opponent's defensive efficiency. Use both season and last-10 numbers and say which you trusted. Adjust for home court and rest days.
Then compare your projection with the posted total and spread.
Output exactly:
1) Projected score for each team and the implied total
2) Over/Under pick stated against the posted line
3) Spread pick and moneyline win probability
4) Confidence 1-10
5) One-line reasoning
If your projection lands within two points of the total or one point of the spread, answer "no bet".
You are a situational basketball handicapper. For {home} vs {away} ({league}, {date}):
Prioritise availability (star injuries, load management, minutes restrictions), schedule spots (back-to-backs, three games in four nights, travel and time zones) and matchup exploits — above season averages. A late injury can flip the edge, so state what would void your pick.
Check {odds} for lag against the news you have.
Output exactly:
1) Adjusted winner + win probability
2) Spread pick and total lean
3) Confidence 1-10
4) The key situational factor
5) The condition that voids this pick
If the injury report is not final, say so and answer "no bet".
Placeholders in braces are filled automatically when you run a prompt from a match in the AI Lab. Pasting into your own chat window works too — just replace them by hand.
What to feed the model, and what a usable answer looks like
Feed it this
- League, date and home/away — plus playoff or tournament context.
- Pace (possessions per 48) and offensive/defensive rating, season and last 10.
- The current injury and load-management report, with the time it was published.
- Rest days, back-to-back flags and travel for both teams.
- The line: spread, total and moneyline — all three.
- Any known rotation change: trade, new starter, minutes restriction.
Good output has
- A projected score for each team and the implied total, with tabular numbers.
- An Over-Under pick stated against the posted line, not in the abstract.
- A spread pick and a moneyline probability.
- Confidence 1-10, plus the single situational factor that decides the game.
- An explicit condition that voids the pick (for example a questionable starter being ruled out).
- A "no bet" when the projection lands within a point or two of the line.
Where basketball prompts usually go wrong
- Season averages hiding a post-trade rotation.
- Three-point variance mistaken for a shooting trend.
- Blowouts and garbage time distorting pace and totals.
- Answering before the final injury report is out.
Odds, model context and market drift for each fixture are on the basketball games with odds and AI picks board, so most of the input list above can be copied straight from the match page.
Measure both versions before you trust either
Store both versions
Save v1 and v2 as separate prompts in the AI Lab so every run is attributed to a version instead of blurring together.
Run them on the same matches
Pick fixtures from the basketball board and lock both forecasts before start. Same slate, same information, no hindsight.
Judge on ROI, not hit-rate
A value prompt taking underdogs will always look worse on hit-rate and can still be the profitable one. Settlement and scoring are automatic once the match finishes.
The AI Lab starts on the $19 tier with one sport and five stored prompts, which is enough for a full v1-versus-v2 comparison in basketball. Open a free trial to run it on today's card, or read the prompt library overview for the shared structure behind every sport.
Basketball prompt questions
Why does the basketball prompt ask for a projected score?
Because a projected score is falsifiable. "Home team looks strong" cannot be graded, but a projection of 112-107 against a posted total of 223.5 gives you a pick, a margin of error, and a number to review after the game. It also makes two prompt versions genuinely comparable.
How do I handle late injury news?
Build it into the prompt: ask for a pick plus the condition that voids it. In the AI Lab you lock a forecast before tip-off, so a pick that names its own void condition is far more useful than one that silently assumes a full roster.
Does the same prompt work for EuroLeague and college basketball?
The structure does — pace and efficiency are league-agnostic. The reference points are not: totals, pace ranges and home-court advantage differ by league, so tell the model which league it is looking at and feed league-specific numbers rather than NBA habits.
Prompts for the rest of the board
Find out which basketball prompt actually wins
Start the 5-day AI Lab trial without a card. Bring your own AI key, run v1 and v2 on today's basketball card, and let the dashboard settle it on a virtual $10,000 bank.
AI Lab $19/mo after the trial · cancel anytime · Full Access $49/mo for all sports and the official 7-AI consensus