
- Each prompt asks a single question: form and roster, patch, format, odds and checklist.
- A new patch makes data stale within weeks, and the AI does not know which version is being played today.
- A prompt organizes your series analysis, it does not pick the bet and it does not guarantee a result.
Save this article: the next time you look at a series, come back, copy the prompt you need and pick up where you left off.
Your progress is saved in this browser.
Which prompts work for betting on esports?
In esports the game itself changes several times a year.
A team that is strong in one version may not be in the next.
If you ask an AI “tell me who wins”, it answers with what it knew months ago.
That is why this article splits the analysis into five prompts.
Each one asks a single question.
You put the answers together and draw your own conclusions.
All the prompts use parameters in [BRACKETS].
Replace each one with your own data before you paste.
If you leave a bracket unfilled, the AI fills it with whatever comes to mind.
Golden rule: every piece of data has to come with its date.
A statistic from before the latest patch may describe a game that no longer exists.
In every prompt we ask for a source and for it to write “I don’t know” when it cannot verify.
This article does not explain the markets from scratch.
For that there is AI for esports betting: how a model reads a series.
And if you are looking for a prompt for another market, there is the index of ChatGPT prompts for every betting market.
Before you paste: three rules so the AI answers better
- Turn on web search: without it, the AI does not know the current patch or today’s rosters.
- Name the game and the tournament: the same organization can have rosters in several titles.
- Paste the odds yourself: you see them at your bookmaker, the AI does not.
Features change: check in the app what your plan includes.
Not every version of an AI searches the internet.
If yours does not, paste the data in yourself and use it only to organize and calculate.
Prompt 1: the team’s recent form and roster changes
This is the starter prompt.
In esports rosters change often, sometimes between one tournament and the next.
A streak built with five other players does not count for today’s series.
Analyze the [GAME] series between [TEAM A] and [TEAM B] on [DATE] in [TOURNAMENT]. For each team, list its last [N] series: date, opponent, result in maps and tournament. Tell me which players make up the roster today and since what date each one has played for the team. Mark which series on the list were played with a roster different from the current one. Cite the source and the date of each piece of data, and if you cannot verify it, write “I don’t know”.
What it returns
A list of series for each team and the date of each roster change.
It helps you see how many series are really worth comparing.
Often only three or four are left.
How to verify it
Open the source and check two series from each list.
Look at the roster on the team’s or the tournament’s official page.
If the AI names a player who has already left, discard that whole answer.

Prompt 2: the patch or game version and what changed
This is the prompt that does not exist in any other sport.
A patch changes characters, weapons, maps or the rules of the game.
What worked last month may not work today.
Look up which patch or version of [GAME] the [TOURNAMENT] tournament is played on as of [DATE]. Tell me when that patch came out and summarize the main changes in five points. State how many official series each team, [TEAM A] and [TEAM B], has played since that patch came out. Clearly separate the results from before and after the patch. Cite the official patch notes and, if you cannot confirm the version, write “I don’t know”.
What it returns
The version being played and a cutoff line on the calendar.
Everything before that date is worth less.
It also tells you how many new series there are, which are usually very few.
How to verify it
Go to the game’s official notes and look at the date of the latest patch.
Check on the tournament page which version it is played on, because sometimes it is not the newest one.
If the AI brings no link, treat the data as unconfirmed.
The AI does not know which version is being played today.
Its knowledge has a cutoff date, and the game kept changing afterwards.
If it did not search the internet, what it says about the “current meta” is old history.
Prompt 3: series format and maps
A single-map series is not the same as a best-of-five.
The more maps are played, the more the difference in level weighs.
In a short series, the weaker team pulls off a surprise more often.
The [GAME] series between [TEAM A] and [TEAM B] is played as a best of [1, 3 OR 5] maps. Explain how maps are picked and banned in the [TOURNAMENT] tournament. For each team, give me its result by map on the current patch: games played and won on each one. Tell me which maps each team usually picks and which it usually bans. State the number of games behind each number, cite the source and write “I don’t know” if you do not have it.
What it returns
The format, the pick-and-ban mechanics and a sheet for each map.
It helps you think about the map handicap and the total maps.
In games with no map picks, the AI should say so.
How to verify it
Look at how many games are behind each percentage.
A “100% win rate” on a map played twice says nothing.
The map pick rules are on the tournament page.
| Format (example) | Maps to win | Wins the series | Wins without losing a map |
|---|---|---|---|
| Best of 1 | 1 | 60.0% | 60.0% |
| Best of 3 | 2 | 64.8% | 36.0% |
| Best of 5 | 3 | 68.3% | 21.6% |
The table is an example calculation, not data from any team.
It assumes a favorite who wins each map 60% of the time, with maps independent of each other.
It shows why the format changes the probability even if the level is the same.
Prompt 4: the odds on the winner and the map handicap
This prompt compares what the market offers with what you estimate.
It needs you to pass it the odds.
It does not look for a bet, it looks for the distance between your number and the market’s.
In the series between [TEAM A] and [TEAM B], best of [N] maps, [TEAM A]’s odds are [ODDS A] and [TEAM B]’s odds are [ODDS B]. [TEAM A]’s map handicap is [HANDICAP] at odds of [HANDICAP ODDS]. Calculate the implied probability of each set of odds, step by step, and how much the two on the winner add up to. Explain which series results win and lose that handicap. Do not estimate the real probability yourself and do not tell me whether it is worth betting: just do the calculations.
What it returns
The probability the market gives to each outcome.
The calculation is 1 divided by the odds.
Example: odds of 1.50 equal 66.7% and odds of 2.50 equal 40%.
In that example the two sets of odds add up to 106.7%.
Whatever goes over 100% is the bookmaker’s margin.
In small markets the margin is usually higher than in big ones.
How to verify it
Redo the calculations with the odds converter.
If you want to review the idea, there is the guide to implied probability.
If your calculation and the AI’s do not match, yours is the one that counts.

Prompt 5: the checklist before betting on a series
This is the closing prompt.
It does not analyze the series: it reviews your analysis.
You ask it to look for gaps before you put money down.
This is my analysis of the [GAME] series between [TEAM A] and [TEAM B] on [DATE]: [PASTE YOUR DATA AND CONCLUSIONS]. Act as a critical reviewer. Tell me: 1) which piece of data is not verified, 2) which data predates the current patch or a roster change, 3) how many series or games support each conclusion, 4) what information is missing (substitutes, format, online or LAN play) and 5) what bias I may have. Do not give me a betting verdict. Just flag risks.
What it returns
A list of gaps and risks.
It is the most useful prompt for stopping an impulsive bet.
If the list is long, maybe the best thing is not to bet.
How to verify it
Tick off each risk you resolved.
The ones you cannot resolve stay as a reason to skip the series.
A good analysis can also end in not betting.
How to use the 5 prompts in order: step by step
What are the most common mistakes with these prompts?
- Using statistics from an earlier patch as if they were from today.
- Counting a streak that was built with another roster.
- Reading a percentage by map with two or three games behind it.
- Analyzing minor leagues with data nobody can check.
- Asking the AI for the “real” probability of the series: it does not have it and makes one up.
No AI bets for you or knows the result.
All of them can make up data, including the ones that search the internet.
In esports, on top of that, what was true may have stopped being true.
With so few comparable series, any number has a wide margin of error.
We explain it in why a small sample misleads.
That is why it pays to bet little, or not at all, when the data is doubtful.
What Gambeta does with esports, no promises
To be clear: Gambeta does not publish esports picks.
These prompts are for organizing your own analysis.
How to look after your money in these markets is in expected value and bankroll in niche sports and esports.
What we do publish every day are predictions for other sports, with our own model.
Each one stays in the public record with its odds, win or lose.
You can see how it works in the guide to AI for sports betting, with free predictions every day.
For adults 18 and over only.
Bet only money you can afford to lose, set a limit before you start and stop if you stop enjoying it.
More at responsible betting.
What to read next
Another field with small samples and last-minute data is combat sports: 5 prompts for UFC and boxing betting.
If you prefer a sport with years of public data, look at the 5 prompts for tennis betting.
If you want a prompt of your own, build it with the prompt builder.
And for the rest of the markets, go back to the index of betting prompts by market.
Frequently asked questions
Which prompt should I use to bet on esports?
Start with the one for recent form and roster changes.
Then add the patch prompt and the series format prompt.
Always finish with the odds prompt and the checklist.
Does ChatGPT know which patch is being played?
Not on its own.
Its knowledge has a cutoff date and games are updated often.
It has to look it up on the internet, and you have to check it in the game’s official patch notes.
Do the same prompts work for any video game?
Yes, the structure is the same.
In the brackets you state the game, the tournament and the series format.
What changes is which statistics exist for each title.
What is a map handicap?
It is a head start or a handicap in maps that is added to the series result.
In a best-of-three series, a −1.5 map handicap requires winning 2-0.
The +1.5 map handicap is met by winning at least one map.
Is it a good idea to bet on minor esports leagues?
That is where there is the least reliable information.
Results and rosters sometimes do not even appear in public sources.
If you cannot verify the data, the reasonable thing is to skip that series.
Does Gambeta publish esports picks?
No.
Gambeta publishes predictions for other sports, with a public record.
This article is about the method, so you can organize your own analysis.