The field before the count

FRAME

A team wins 80% of its home matches. Before you believe that number, ask two questions: what got measured, and which matches were counted? Just as important — what sits outside that count?

Probability is the language built to answer that second question: it names the full field of outcomes — the sample space. Crop the same four wins down to five matches, and the number reads 80%. Zoom out to twenty matches, same four wins, and it reads 20%. Nothing about the wins changed. The frame did.

Name the full field before you count anything inside it. That discipline is what this whole chapter builds.

the same four wins — two very different probabilities
The crop4 wins in 5 matches → 80%
the sample space — the denominator — changed
The zoom-out4 wins in 20 matches → 20%
name the full field of possibilities before counting anything
before you trust any number — ask two questions
What exactly got measured — and which results were counted?
What sits outside the frame — the part that got cropped out?
probability exists to answer the second question with a precise number, not a guess

↑ Back to top

Why probability belongs in AI

CONCEPT

The Math for AI survey names four branches: statistics, linear algebra, probability, and calculus. Of these four, only two are actually worked out in Class 9 — statistics and probability. The other two appear by name only, not in your homework.

Probability earns its own chapter for one reason. Every AI prediction is a probability estimate — a number between 0 and 1, never a flat yes or no. This chapter's payoff is understanding that number — what it means, where it comes from, and when it is trustworthy enough to act on.

why probability earns its own chapter in AI
Every AI prediction is a probability — not yes/no but a number between 0 and 1
Statistics and probability are the two branches calculated at G9 depth
Understanding the number — meaning, source, reliability — is the chapter's payoff
a model that outputs 0.94 is not saying 'spam' — it is saying 'in the long run, 94% of inputs like this are spam'
KEY-TERM

Probability is a number between 0 and 1, inclusive, that measures how likely an event is to occur. The formula: P(event) = favourable outcomes ÷ total equally likely outcomes.

Three values anchor the scale: P = 0 means impossible, P = 0.5 means equally likely either way, P = 1 means certain. Every calculation ahead runs through this formula — favourable outcomes on top, total on the bottom, divide.

one formula — three anchor values
FORMULA — P = favourable outcomes ÷ total equally likely outcomes
P = 0 — the event cannot happen (impossible)
P = 0.5 — equally likely to happen or not
P = 1 — the event must happen (certain)
every calculation in this chapter runs through this formula: write the event set on top, the sample space on the bottom, divide
CONCEPT

Before calculating any probability, name the sample space — the complete set of all possible outcomes. A coin toss has sample space {H, T} — two outcomes. A standard die has {1, 2, 3, 4, 5, 6} — six outcomes. Toss two coins together and the sample space grows: {HH, HT, TH, TT} — four outcomes, one for each combination.

The sample space is the formula's denominator — the field the crop sits inside. A probability claim that does not name its sample space is incomplete: it has cropped the field without showing the frame.

Take the team that wins 80% of its home matches: four wins is the same four wins whether the sample space is five matches or twenty. Only the size of the named denominator changes the answer — not the wins.

One doubt worth flagging: does the sample space shift with the event you're asking about? It does not — section 4's die shows one fixed six-outcome sample space answering four separate questions in turn.

the sample space is the denominator
COIN TOSS — sample space = {H, T} — 2 equally likely outcomes
STANDARD DIE — sample space = {1, 2, 3, 4, 5, 6} — 6 equally likely outcomes
TWO COIN TOSSES — sample space = {HH, HT, TH, TT} — 4 equally likely outcomes
the sample space is fixed by the EXPERIMENT, not by the event you are looking for

ai-probability.sample-space

/50-system/standards/book-look/cell1-a4/adv-dignaga.png

Dignaga

kingfisher

Before you trust any fraction, name its field exactly — the denominator is a definition, and a percentage means nothing until you have said what it was counted against.

CONCEPT

A fair coin has probability 0.5 of landing heads. Toss it ten times and it feels like five heads should come up. Try it — you might get three. You might get eight.

Probability is a long-run frequency, not a guarantee about the next trial. Toss it hundreds of times, and the proportion of heads creeps toward 0.5. But any short run can sit far from it — and still be perfectly consistent with a fair coin.

The same rule holds for an AI model: predicted probability 0.7 is right roughly 70% of the time in the long run — any one prediction can still land in the wrong 30%. A model trained on a small dataset guesses from a narrow sample space, which gives an unreliable long-run estimate. A probability is only as trustworthy as the sample space behind it.

probability is long-run, not per-trial
Mythexactly 7 in the next 10
a long-run frequency; short runs deviate widely
Reality≈70% in the long run
an AI model trained on small data is estimating with a narrow sample space — more data is not optional, it is how the long-run estimate becomes reliable

↑ Back to top

Types of events

KEY-TERM

Every event sorts into one of four types — really, one question: how many outcomes qualify, against how many are possible?

An impossible event has probability 0 — nothing qualifies (rolling a 7). A certain event has probability 1 — every outcome qualifies (rolling 1 to 6). A simple event has exactly one qualifying outcome (rolling a 4, P = 1/6). A compound event groups more than one: rolling an even number, 2, 4, 6 qualify, P = 3/6 = 1/2.

Event type only tells you how to count the numerator — the denominator, the sample space, never moves.

two named cases — one fixed sample space
Type
Definition
Probability
Die example
Impossible
Cannot happen
P = 0
Roll a 7
Certain
Must happen
P = 1
Roll 1–6
Named event (one outcome)
Exactly one outcome qualifies
P = 1/6
Roll a 4
Named event (several outcomes)
More than one outcome qualifies
P = 3/6 = 1/2
Roll an even number (2,4,6)
the denominator never changes — only how many outcomes qualify (the numerator) changes
does a bigger group win by definition?
Trapmore outcomes is always more likely
three impossible outcomes together is still P = 0
Realityonly the count against the fixed sample space matters

"A compound event is always more likely — it covers more outcomes."

Compare the count, not the label: even numbers (2/4/6) give P = 1/2, a 4 alone gives P = 1/6 — compound wins only because three beats one. The group's size alone decides nothing — only the count against the fixed sample space does.

↑ Back to top

Calculating probability

CONCEPT

Calculate a probability in three fixed steps.

Two coin tosses: {HH, HT, TH, TT} — four outcomes. "At least one head" excludes TT: three outcomes qualify. P = 3/4 = 0.75.

That order keeps the denominator honest.

three steps — always in this order
NAME THE SAMPLE SPACE — list all equally likely outcomes (the denominator)NAME THE EVENT SET — list the favourable outcomes (the numerator)CALCULATE — P = favourable ÷ total
two coin tosses: sample space = {HH,HT,TH,TT}, event 'at least one head' = {HH,HT,TH}, P = 3/4 = 0.75
How to calculate a probability — the three-step discipline
  1. Name the sample space. Every outcome first: {HH, HT, TH, TT} — the denominator.
  2. Name the event set. "At least one head" excludes TT — {HH, HT, TH}, the numerator.
  3. Calculate. P = 3/4 = 0.75. Skip step 1 and the denominator is a guess, not a count.
CONCEPT

Four different questions, one die, one sample space: {1, 2, 3, 4, 5, 6}, six outcomes, unchanged throughout.

P(rolling a 3) = 1/6 — one outcome qualifies. P(rolling an odd number) = 3/6 = 1/2 — event set {1, 3, 5}. P(rolling a number greater than 6) = 0/6 = 0 — impossible, nothing qualifies. P(rolling a number from 1 to 6) = 6/6 = 1 — certain, every outcome qualifies.

The denominator is fixed by the experiment, not the question.

one die — four different events, one fixed denominator
P(roll a 3) = 1/6 — one outcome in the event set
P(roll odd) = 3/6 = 1/2 — event set = {1,3,5}
P(roll > 6) = 0/6 = 0 — impossible; nothing qualifies
P(roll 1–6) = 6/6 = 1 — certain; every outcome qualifies
the sample space = {1,2,3,4,5,6} is written ONCE and never changed; only the event set varies

↑ Back to top

Probability in three real-life domains

CONCEPT

Teach a machine three guesses: win chance, rain chance, traffic-jam chance. Feed it a record of what actually happened.

In sports: win probability is wins ÷ total matches. For weather: fraction of similar days it rained. For traffic: fraction of similar hours it was congested.

Three domains, one structure: history supplies the sample space, and a bigger history gives a more reliable number.

three domains, one structure — historical record = sample space
Domain
Sample space
Event
Probability
Sports
All past matches in the season
Team wins a match
Wins ÷ total matches
Weather
Past days with similar conditions
It rains on that day
Rainy days ÷ total past days
Traffic
All recorded hours at that time
Road is congested
Congested hours ÷ total recorded hours
more historical data = wider sample space = more reliable probability estimate — this is why AI needs large training sets

↑ Back to top

How AI uses probability — and its limits

CONCEPT

A spam filter hands you P(spam) = 0.94. A medical AI hands you P(pneumonia) = 0.78 — numbers that close to 1 feel almost certain.

A high probability is not certainty. P = 0.94 means the model is right roughly 94% of the time in the long run. Any single prediction can still land in the remaining 6%. Only P = 1 means certain.

The model outputs the probability; a human or a rule turns it into an action. An honest AI report always states the probability, not just the decision.

the model outputs a probability — a human converts it to an action
AI outputsP(spam) = 0.94
probability expresses uncertainty; only P = 1 is certainty
NOT certaintyroughly 6% still wrong
in medical diagnosis or credit decisions, the error rate is the number that matters most — a 94% accuracy model still fails about 6% of the time
what an honest AI report states
Decision onlyprobability stays hidden
states the probability, not just the decision
Honest reportprobability is stated

"If an AI model gives a high probability — like 0.95 — the answer is definitely correct."

P = 0.95 means right ~95% of the time — not guaranteed correct. A 95%-accurate spam filter still misses 1 in 20 emails — the number that matters most. Only P = 1 is certain, and real AI systems rarely reach it.

CONCEPT

Take a coin you suspect might be biased. Toss it 20 times: 13 heads, so P(heads) = 0.65 — far from 0.5. Biased coin, or a short, unlucky-looking run?

Toss 60: 0.53. Toss 120: 0.50 — the early 0.65 was illusion. A spurious pattern works the same way: if the effect is not real, a large enough sample converges its long-run frequency back to the true value, and the deviation disappears.

This is why more training data is the first cure for a spurious pattern. A pattern that survives a large sample is more likely genuine. One that shrinks as the sample grows is more likely noise.

the first cure for a suspected spurious pattern
Suspect a pattern is spurious? Increase the training-data sizeWatch where the frequency settles as data grows
long-run frequency converges as data grows
Small sample (20 tosses) 0.65
P(heads) estimated at 0.65 — deviates from true 0.5
Medium sample (60 tosses) 0.53
estimate tightens toward 0.5
Large sample (120+ tosses) 0.5
long-run frequency converges to true probability
a spurious pattern in small data shrinks toward the true long-run value as data grows — convergence is the test
what convergence actually tests
Noisetrue probability is zero — frequency converges to zero as data grows
run the sample forward and see where the frequency settles
Real effecta genuine non-zero probability — frequency converges to that value
Probability for AI (Grade 9 · 417) · projected from the LATTICE via prism_html.py · register: school-g9

↑ Back to top