Prompt library
Sports Questions That Get a Better Read


Kam AI
Product and research

Prompt library


Kam AI
Product and research

This post is a list of sports questions you can ask Kam.
The best questions tell Kam what spot to watch: sport, game, market, book, and what you want to check.
Good prompts ask for what moved, what is missing, and one next action.
They should help you avoid bad numbers, not pretend a bet is certain.
The best sports research prompts are not long.
They are specific.
They tell Kam what decision you are trying to make, which market you care about, and what kind of answer would help you move faster.
This post is a strict prompt library. It is not a product essay. It is not a betting strategy guide. It is a list of sports questions you can ask when you want Kam to inspect a board, explain movement, review a ticket, compare markets, or check whether an answer is safe to trust.
Use this shape when you are not sure how to ask:
For [sport/league/date], check [market/question] using [book/source/context].
Show [specific output], name missing data, and give one next action.
Examples:
For today's NBA board, check spreads at FanDuel. Show the three games with the clearest movement, name missing data, and give one next action for each.
For Warriors +5.5, check whether my ticket is covering right now. Use my entry line, current score, and game status. If anything is missing, ask one question.
For this selected game, explain why the line moved. Separate confirmed movement from possible causes, and do not invent injury or news context.
Prompt upgrade
Weak prompt
Who should I bet tonight?
Better prompt
Show me tonight's NBA board. Separate real line movement from stale or missing data. For each game, give me the market, current line, what moved, confidence level, missing info, and one next check.
Prompt upgrade
Weak prompt
Is this trend good?
Better prompt
For this ATS trend, show the games counted, closing spread, final score, cover margin, sample size, and any missing data. Then tell me whether the trend is strong enough to keep researching.
Prompt use cases
Takeaway: Good sports prompts are scoped enough to prevent fake confidence, but short enough to use while scanning a live board.
Use board prompts when you are trying to decide where to spend attention.
What are the most important games on today's NBA board?
Which games have enough spread, total, moneyline, movement, and freshness data to inspect?
Show me the board by sport, but do not rank games where source coverage is missing.
Give me a quick board triage: inspect, watch, skip, or missing data.
Which games changed most since the last board refresh?
Show only games where the current line differs materially from the open.
Show today's NBA board with FanDuel lines only. If FanDuel is missing, say missing instead of substituting another book.
Board prompt intent
Find which games deserve attention without pretending every row has equal signal.
Limit by sport, book, market, favorite, dog, total, watchlist, or freshness state.
Ask Kam to name stale, missing, delayed, or unavailable data before ranking.
Takeaway: A board prompt should help you spend attention, not manufacture a pick.
Use movement prompts when the market moved and you need to know what moved, what is confirmed, and what is still unknown.
Why did this line move?
What moved from open to current line for this selected game?
Did the movement happen before or after the injury news?
Did the spread, total, and moneyline all move in the same direction?
Did this move help or hurt my entry price?
Show observed movement first. Then list possible causes only if there is source context.
Do not say sharp money, injury, or public sentiment unless the source data supports it.
Movement prompts should force Kam to separate facts from interpretation.
Observed movement is a fact.
Cause is a claim.
Do not mix them.
Use ticket prompts when you already have a bet and need the state explained from your entry line.
I took Warriors +5.5. Am I covering right now?
I took over 221.5. Is the current score pace helping or hurting?
I took Knicks moneyline. Are they winning, losing, or final?
Did the market move in my favor after I placed this ticket?
Compare my entry line to the current line and closing line, but do not grade my bet from a later line.
What is the simplest next action for this ticket: hold, watch injury news, check final score, or review later?
If you do not have my entry line, score, or game status, ask one narrow question.
Ticket prompts need strict grading rules:
Use trend prompts when you need sample size, included games, and missing-data caveats.
Is this team covering spreads lately?
Which games counted in that ATS trend?
Show each game in the trend sample with opponent, closing spread, final score, result, and cover margin.
Is this player trend supported by enough recent games?
Is the trend based on current roster context, or is it mixing old context with today's game?
Give me the trend only if the sample size and denominator are loaded.
If the denominator is missing, use counts instead of percentages.
Trend prompts should make Kam prove the sample.
If the answer cannot show what counted, the trend is not ready to trust.
Use prop prompts when player usage, matchup, injury, and line context matter.
Is this player points prop supported by recent usage and matchup context?
Did the player's role change in the last five games?
Is the prop line moving with injury news, matchup news, or no confirmed source?
Compare the current prop line to recent outcomes, but name the sample size.
What would make this prop a no-answer-yet situation?
Show me the strongest reason to keep watching this prop and the strongest reason to pass.
Do not ask for a prop answer without asking for sample size.
Small samples can be interesting, but they should not be written like certainty.
Use injury prompts when you need timing, confirmation, and market reaction.
What injury or lineup news is relevant to this game?
Did the line move before or after the injury update?
Which source confirms this player status, and when was it updated?
Did the total move in a way that matches the lineup change?
What should I wait for before trusting this market?
Separate confirmed injury status from possible market interpretation.
Injury prompts should never invite invention.
The safe answer is often:
I can see movement, but I do not have a confirmed injury source tied to the timing.
That is better than a confident guess.
Use market comparison prompts when you want to compare sportsbooks, prediction markets, or current prices.
Do books and prediction markets agree on this game?
Which sportsbook is furthest from the market consensus?
Is the gap still open, or did books already adjust?
Compare FanDuel, DraftKings, and prediction market price if all are loaded.
Do not call it a gap unless at least two comparable prices are loaded.
If the prediction market data is stale or unavailable, say so before answering.
Market comparison prompts should define the comparison.
One price is not a gap.
Use watchlist prompts when you care about change over time.
What moved since I last checked?
Which watchlist item moved the most?
Which saved read is now weaker because of line, score, or injury update?
Which watchlist item needs a fresh source check?
Show only moves that matter to the original reason I saved the read.
What should I remove from the watchlist because the signal expired?
Watchlist prompts are memory prompts.
The answer should compare current state against why the user saved the read, not only summarize the current board.
Use postgame prompts when you want to learn from a result without rewriting history.
Did this bet win for the reason I expected?
Was the original read right, wrong, stale, or unrelated to the outcome?
What did the closing line say about my entry?
Did my ticket cover, push, or fail to cover by the entry line?
What should I change in my process next time?
Separate bad process from normal variance.
Postgame review should not be emotional.
It should classify:
Use these prompts when the risk is not the market, but the answer itself.
Is any data in this answer stale?
Which source supports the key claim?
What is the as-of time for the line, score, injury, and market data?
Which part of this answer is confirmed, and which part is interpretation?
What data would change your answer?
Is this safe to rank, or should it stay as a watch-only note?
These prompts are useful when an answer sounds too confident.
They force Kam to show the receipt.
Use clarification prompts when Kam should stop instead of guessing.
What exact context do you need before answering?
Ask me one question that would unblock this.
Can you answer from the selected game, or do you need me to choose one?
Do you have enough data to answer, or should this be a no-answer-yet state?
What should I open first: sport, game, market, book, ticket, or source?
Clarification should be narrow.
Bad clarification:
Can you provide more details?
Good clarification:
Which game should I use for the line-movement check?
You can combine prompts when the workflow is clear.
For this selected NBA game, compare spread movement, injury timing, and prediction-market gap. Lead with what moved, name missing data, and give one next action.
For my saved Warriors +5.5 ticket, check current cover margin, market movement since entry, and whether the original read still holds.
For today's NFL board, find games where FanDuel differs from consensus, but only rank games with fresh source data.
For this player prop, compare recent usage, matchup, injury context, and line movement. If the sample is too thin, say no answer yet.
Combined prompts should still be bounded.
If a prompt asks for everything, the answer gets weaker.
Avoid prompts that force fake confidence.
Tell me exactly what to play.
What is certain?
Tell me the outcome.
Find me something with no risk.
Explain the line move even if data is missing.
Use your intuition.
These prompts should be rewritten into research prompts:
Rewrite weak prompts into research prompts
Takeaway: A better prompt turns prediction pressure into research discipline.
Before asking, check whether your prompt names enough context.
You do not need all seven every time.
But the more money or time a decision can cost, the more context the prompt should include.
Here are compact prompts ready to use:
Show today's NBA board. Rank only games with fresh source data and tell me what is missing.
Why did this selected game's spread move? Separate observed movement from possible causes.
I took [team] [line]. Am I covering right now? Use my entry line and current score.
Did the market move in favor of my ticket since entry?
Which games counted in this trend? Show date, opponent, closing spread, final score, result, and margin.
Is this prop supported by recent usage, matchup, and injury context? Name the sample size.
Do books and prediction markets disagree here? Do not call it a gap unless both prices are loaded.
What moved since I saved this watchlist read?
Did the original read still hold after the result?
What exact data is missing before you can answer safely?
Sports prompts should help you research.
They should not pressure an AI system to pretend the future is certain.
The best Kam prompt is not:
What should I bet?
It is:
What is worth researching, what moved, what is sourced, what is missing, and what should I check next?
Prompt library
Good sports research starts with scoped questions: game, market, book, source, freshness, and next action. Kam is strongest when the prompt asks for evidence, not certainty.
Read next
Why Kam is not just chat: it watches your spots, checks what moved, and gives you the read before you chase a bad number.
16 min read
A plain-English look at how Kam checks freshness, missing data, scope, and answer quality before a card or explanation reaches the user.
20 min read
Why Kam helps bettors slow down, see what moved, check what is missing, and avoid chasing bad numbers.
6 min read