Skip to content
NoviLearn
Formula Library

Every formula, explained

What it means in plain English, what each symbol is, when to use it, and the mistakes to avoid.

Symbols & notation— what ∑, μ, σ, P(A | B), ≥ and the rest mean
Sigma (summation)

Add up all the values that follow it.

Example: For the values 1,4,61, 4, 6: ∑x=1+4+6=11\sum x = 1 + 4 + 6 = 11.
Remember: Sigma and SUM both start with S.
Greater than or equal to

The value on the left is bigger than, or the same as, the value on the right.

Example: X≥3X \ge 3 includes 3,4,5,…3, 4, 5, \dots — the 3 itself counts.
In words: “At least”, “no fewer than”, “minimum of”.
Remember: The line under the symbol is the “or equal” part.
Less than or equal to

The value on the left is smaller than, or the same as, the value on the right.

Example: X≤3X \le 3 includes 0,1,2,30, 1, 2, 3 — the 3 itself counts.
In words: “At most”, “no more than”, “maximum of”.
Remember: The line under the symbol is the “or equal” part.
Not equal to

The two sides are different values.

Example: 2≠32 \ne 3.
Approximately equal to

The two sides are close, usually because one was rounded.

Example: 13≈0.333\frac{1}{3} \approx 0.333.
Mu (population mean)

The mean (average) of a whole population or of a probability model.

Example: If a model says the average wait is 4 minutes, μ=4\mu = 4.
Remember: μ is the Greek “m” — for mean.
Sigma, lowercase (standard deviation)

How far values typically fall from the mean. Bigger σ means more spread out.

Example: Heights with σ=2\sigma = 2 cm cluster tightly; with σ=10\sigma = 10 cm they spread widely.
Remember: Lowercase σ is the Greek “s” — for standard deviation (spread).
Sigma squared (variance)

The standard deviation squared. It measures spread in squared units.

Example: If σ=3\sigma = 3, then σ2=9\sigma^2 = 9. Going back: σ=9=3\sigma = \sqrt{9} = 3.
Remember: Variance is “σ before the square root”.
Probability of an event

How likely event A is, as a number from 0 (impossible) to 1 (certain).

Example: For a fair coin, P(heads)=0.5P(\text{heads}) = 0.5.
Remember: Read it out loud as “the probability of A”.
Conditional probability (“given”)

The probability of A when you already know B happened. The bar means “given”.

Example: P(umbrella∣rain)P(\text{umbrella} \mid \text{rain}): of the rainy days, the share with an umbrella.
Remember: Whatever comes after the bar is what you already know — it becomes your new “everything”.
AND (intersection, ∩)

Both A and B happen — the outcomes the two events share.

Example: Rolling a die, A = even and B = more than 3: A AND B = {4,6}\{4, 6\}.
Remember: ∩ looks like an arch over what's shared — and “n” for iNtersection.
OR (union, ∪)

A happens, B happens, or both — every outcome in either event, counted once.

Example: Rolling a die, A = even and B = more than 3: A OR B = {2,4,5,6}\{2, 4, 5, 6\}.
Remember: ∪ is a cup that holds everything from both.
Complement (“not A”)

Every outcome where A does not happen. $P(A') = 1 - P(A)$.

Example: If P(rain)=0.3P(\text{rain}) = 0.3, then P(no rain)=0.7P(\text{no rain}) = 0.7.
Remember: The little tick (′) means “everything else”.
Factorial

Multiply a whole number by every whole number below it, down to 1.

Example: 4!=4×3×2×1=244! = 4 \times 3 \times 2 \times 1 = 24. By definition 0!=10! = 1.
Remember: The ! says “keep multiplying down”.
Combinations (“n choose r”)

The number of ways to choose r items from n when order doesn't matter. Also written nCr.

Example: Choosing 2 snacks from 4: (42)=6\binom{4}{2} = 6 ways.
Remember: Read it as “n choose r”.
Tilde (“is distributed as”)

Names the probability model a random variable follows, with its parameters in brackets.

Example: X∼N(10,2)X \sim N(10, 2): X is normal with mean 10 and standard deviation 2.
Remember: Read ~ as “follows”.
Expected value

The long-run average of X if you repeated the situation many times. It equals μ.

Example: A game that pays $1\$1 half the time and $0\$0 otherwise has E(X)=$0.50E(X) = \$0.50.
Remember: E for “expect on average”.

Probability with equally likely outcomes

P(A)=number of outcomes in Anumber of outcomes in SP(A) = \frac{\text{number of outcomes in } A}{\text{number of outcomes in } S}

When every outcome is equally likely, a probability is a count of favorable outcomes over a count of all outcomes.

SS
Sample space (all possible outcomes)
AA
The event of interest
When to use: Fair coins, fair dice, cards, random draws — any experiment where each outcome has the same chance.
Example: Fair die, AA = at least 5: A={5,6}A = \{5, 6\}, so P(A)=26≈0.3333P(A) = \frac{2}{6} \approx 0.3333.
See it in the Lab: Probability Playground
Watch out: Using it when outcomes aren't equally likely (e.g. the sums of two dice). · Mishandling boundaries such as “greater than” vs “at least”.
P(A′)=1−P(A)P(A') = 1 - P(A)

An event and its complement cover the whole sample space with no overlap, so their probabilities add to 1.

A′A'
“A prime”: all outcomes not in AA
When to use: “Not”, “fails”, “misses”, or whenever the opposite event is easier to find.
Example: A guard makes 80% of free throws: P(miss)=1−0.80=0.20P(\text{miss}) = 1 - 0.80 = 0.20.
See it in the Lab: Probability Playground
Watch out: Reporting P(A)P(A) when the question asks for “not AA”. · Writing the answer as a percent number like 20 instead of 0.20.
P(A∣B)=P(A AND B)P(B)P(A|B) = \frac{P(A \text{ AND } B)}{P(B)}

The probability of AA once you know BB happened: BB becomes the new sample space.

P(A∣B)P(A|B)
Probability of AA given BB
P(A AND B)P(A \text{ AND } B)
Probability both happen
P(B)P(B)
Probability of the condition (must be > 0)
When to use: “Given”, “if we know”, “among those who”, or a table question restricted to one row or column.
Example: Fair die, AA = 2 or 3, BB = even: P(A∣B)=1/63/6=13≈0.3333P(A|B) = \frac{1/6}{3/6} = \frac{1}{3} \approx 0.3333.
See it in the Lab: Probability Playground
Watch out: Dividing by the whole sample space instead of by BB. · Swapping the roles: P(A∣B)P(A|B) is not P(B∣A)P(B|A).

Test for independent events

P(A AND B)=P(A)P(B)P(A \text{ AND } B) = P(A)P(B)

Events are independent when the overlap equals the product of their probabilities. Equivalently P(A∣B)=P(A)P(A|B) = P(A) or P(B∣A)=P(B)P(B|A) = P(B).

P(A AND B)P(A \text{ AND } B)
Probability both happen
P(A)P(B)P(A)P(B)
Product of the separate probabilities
When to use: Deciding whether two events are independent; or, once told they are, finding P(A AND B)P(A \text{ AND } B) by multiplying.
Example: P(C)=0.3P(C) = 0.3, P(L)=0.4P(L) = 0.4, P(C AND L)=0.12P(C \text{ AND } L) = 0.12: since (0.3)(0.4)=0.12(0.3)(0.4) = 0.12, CC and LL are independent.
See it in the Lab: Who's in the Room?
Watch out: Multiplying P(A)P(B)P(A)P(B) for events that aren't known to be independent. · Confusing independence with mutual exclusivity.

Mutually exclusive events

P(A AND B)=0P(A \text{ AND } B) = 0

Two events are mutually exclusive when they share no outcomes, so they can never happen together.

P(A AND B)P(A \text{ AND } B)
Probability both happen — zero for mutually exclusive events
When to use: Checking whether two events can occur at the same time; then P(A OR B)=P(A)+P(B)P(A \text{ OR } B) = P(A) + P(B).
Example: One roll of a die: A={1,2}A = \{1, 2\}, B={5,6}B = \{5, 6\}. No shared outcomes, so P(A AND B)=0P(A \text{ AND } B) = 0.
See it in the Lab: Who's in the Room?
Watch out: Thinking mutually exclusive events are independent. · Requiring the two events to cover the whole sample space (that's the complement).

Multiplication rule

P(A AND B)=P(A∣B) P(B)P(A \text{ AND } B) = P(A|B)\,P(B)

The chance both events happen is the chance of one times the chance of the other given the first. For independent events this becomes P(A)P(B)P(A)P(B).

P(A∣B)P(A|B)
Probability of AA given BB
P(B)P(B)
Probability of the condition BB
When to use: “And”, “both”, or a sequence (“first … then …”). Use P(B∣A)P(A)P(B|A)P(A) if that's the conditional you have.
Example: P(G)=0.6P(G) = 0.6, P(Q∣G)=0.9P(Q|G) = 0.9: P(G AND Q)=(0.9)(0.6)=0.54P(G \text{ AND } Q) = (0.9)(0.6) = 0.54.
See it in the Lab: Probability Rules Explorer
Watch out: Using P(A)P(B)P(A)P(B) for dependent events. · Pairing a conditional with the wrong probability, e.g. P(Q∣G)P(Q)P(Q|G)P(Q).

Addition rule

P(A OR B)=P(A)+P(B)−P(A AND B)P(A \text{ OR } B) = P(A) + P(B) - P(A \text{ AND } B)

The chance at least one event happens: add both, then subtract the overlap so it isn't counted twice. For mutually exclusive events, just add.

P(A OR B)P(A \text{ OR } B)
Probability AA happens, BB happens, or both
P(A AND B)P(A \text{ AND } B)
The overlap, subtracted once
When to use: “Or”, “either”, “at least one of” two events. Rearranged, it finds a missing overlap.
Example: P(G)=0.6P(G) = 0.6, P(L)=0.45P(L) = 0.45, P(G AND L)=0.25P(G \text{ AND } L) = 0.25: P(G OR L)=0.6+0.45−0.25=0.80P(G \text{ OR } L) = 0.6 + 0.45 - 0.25 = 0.80.
See it in the Lab: Probability Rules Explorer
Watch out: Forgetting to subtract the overlap. · Multiplying instead of adding.

Contingency table: single event and AND

P(A)=total for Agrand total,P(A AND B)=cellgrand totalP(A) = \frac{\text{total for } A}{\text{grand total}},\qquad P(A \text{ AND } B) = \frac{\text{cell}}{\text{grand total}}

With nobody excluded, the denominator is everyone. A single event uses a whole row or column; AND uses one cell.

total for A\text{total for } A
Row or column total for AA
cell\text{cell}
Count where the AA row meets the BB column
grand total\text{grand total}
Everyone in the table
When to use: Any “probability that a randomly chosen person is …” question from a two-way table with no “given.”
Example: Coffee table: P(Daily coffee)=120250=0.48P(\text{Daily coffee}) = \frac{120}{250} = 0.48; P(Daily coffee AND Under 7)=72250=0.288P(\text{Daily coffee AND Under 7}) = \frac{72}{250} = 0.288.
See it in the Lab: Who's in the Room?
Watch out: Using one cell for a single-event question. · Dividing an AND cell by a row or column total (that's a conditional).

Contingency table: OR

P(A OR B)=totalA+totalB−cellA AND Bgrand totalP(A \text{ OR } B) = \frac{\text{total}_A + \text{total}_B - \text{cell}_{A \text{ AND } B}}{\text{grand total}}

Count everyone in AA's row or BB's column exactly once by subtracting the shared cell.

totalA, totalB\text{total}_A,\ \text{total}_B
Row total for AA, column total for BB
cellA AND B\text{cell}_{A \text{ AND } B}
The overlap, counted in both totals
When to use: “AA OR BB”, “either … or …”, “at least one of these” in a two-way table.
Example: Coffee table: P(Daily coffee OR Under 7)=120+110−72250=0.632P(\text{Daily coffee OR Under 7}) = \frac{120 + 110 - 72}{250} = 0.632.
See it in the Lab: Who's in the Room?
Watch out: Forgetting to subtract the overlap. · Subtracting the overlap twice (that gives “exactly one”).

Contingency table: conditional probability

P(A∣B)=cellA AND BtotalBP(A \mid B) = \frac{\text{cell}_{A \text{ AND } B}}{\text{total}_B}

Knowing BB happened restricts you to BB's row or column; its total becomes the denominator.

totalB\text{total}_B
Row or column total of the given event BB
cellA AND B\text{cell}_{A \text{ AND } B}
How many in the BB group also have AA
When to use: “Given”, “of those who”, “among”, “if we know” — any time the question names a group first.
Example: Coffee table: P(Under 7∣Daily coffee)=72120=0.60P(\text{Under 7} \mid \text{Daily coffee}) = \frac{72}{120} = 0.60.
See it in the Lab: Who's in the Room?
Watch out: Dividing by the grand total. · Dividing by the wrong total (that answers the reverse question P(B∣A)P(B \mid A)).

Independence check from a table

P(A∣B)=P(A)  ⟺  A,B independentP(A \mid B) = P(A) \iff A, B \text{ independent}

If restricting to BB leaves the chance of AA unchanged, BB carries no information about AA.

P(A∣B)P(A \mid B)
Cell ÷ total for BB
P(A)P(A)
Total for AA ÷ grand total
When to use: Asked whether two traits in a table are independent (or related / dependent).
Example: Coffee table: P(Under 7∣Daily coffee)=0.60P(\text{Under 7} \mid \text{Daily coffee}) = 0.60 but P(Under 7)=0.44P(\text{Under 7}) = 0.44, so dependent.
See it in the Lab: Who's in the Room?
Watch out: Confusing independent with mutually exclusive. · Deciding by intuition instead of computing both probabilities.

Tree diagram: multiply along a path

P(A AND B)=P(A)⋅P(B∣A)P(A \text{ AND } B) = P(A)\cdot P(B \mid A)

The probability of a path is the product of its branches; each later branch is conditional on the earlier ones.

P(A)P(A)
First-stage branch
P(B∣A)P(B \mid A)
Second-stage branch, given the first outcome
When to use: A specific sequence of outcomes across stages (draws, games, days).
Example: 5 of 8 did the reading; two different people called: P(RR)=58⋅47≈0.3571P(RR) = \frac{5}{8}\cdot\frac{4}{7} \approx 0.3571.
See it in the Lab: Probability Rules Explorer
Watch out: Not reducing the counts without replacement. · Adding instead of multiplying along a path.

Tree diagram: add across paths

P(event)=∑P(paths giving the event)P(\text{event}) = \sum P(\text{paths giving the event})

Different paths are mutually exclusive, so the probability of an event is the sum of the path products that produce it.

∑\sum
Sum over every path where the event happens
When to use: “Either order”, “exactly one”, or the overall probability of a second-stage result (e.g. P(B)=P(A)P(B∣A)+P(A′)P(B∣A′)P(B) = P(A)P(B \mid A) + P(A')P(B \mid A')).
Example: Exactly one did the reading: 58⋅37+38⋅57=3056≈0.5357\frac{5}{8}\cdot\frac{3}{7} + \frac{3}{8}\cdot\frac{5}{7} = \frac{30}{56} \approx 0.5357.
See it in the Lab: Probability Rules Explorer
Watch out: Counting only one order. · Averaging second-stage rates instead of weighting them by the first-stage branches.

Venn diagram: OR (addition rule)

P(A OR B)=P(A)+P(B)−P(A AND B)P(A \text{ OR } B) = P(A) + P(B) - P(A \text{ AND } B)

Everything in either circle, with the overlap counted once.

P(A AND B)P(A \text{ AND } B)
The overlap of the two circles
When to use: Overlapping groups and the word OR / either / at least one of the two.
Example: 66 of 120 have a morning class, 48 drive, 30 both: P(M OR D)=66+48−30120=0.70P(M \text{ OR } D) = \frac{66+48-30}{120} = 0.70.
See it in the Lab: Probability Rules Explorer
Watch out: Forgetting to subtract the overlap.

Venn diagram regions: only and neither

P(A AND B′)=P(A)−P(A AND B),P(A′ AND B′)=1−P(A OR B)P(A \text{ AND } B') = P(A) - P(A \text{ AND } B),\qquad P(A' \text{ AND } B') = 1 - P(A \text{ OR } B)

“Only AA” is the circle minus the overlap; “neither” is everything outside both circles.

B′B'
Not BB (the complement of BB)
A′ AND B′A' \text{ AND } B'
Outside both circles
When to use: “AA but not BB”, “only AA”, “neither”, “exactly one.”
Example: Morning class/drive: MM only =66−30120=0.30= \frac{66-30}{120} = 0.30; neither =1−0.70=0.30= 1 - 0.70 = 0.30.
See it in the Lab: Probability Rules Explorer
Watch out: Writing the circle total in the “only” region. · Computing neither as 1−P(A)−P(B)1 - P(A) - P(B) (subtracts the overlap twice).

Multiplication (counting) principle

total=n1×n2×⋯×nk\text{total} = n_1 \times n_2 \times \cdots \times n_k

If a task happens in stages with n1,n2,…,nkn_1, n_2, \ldots, n_k choices, the number of complete outcomes is the product. With rr slots of nn options and repeats allowed, that's nrn^r.

nin_i
Number of choices at stage ii
nrn^r
rr slots, nn options each, repeats allowed
When to use: Outfits, menus, codes and PINs with repeats, answer sheets.
Example: 4-digit PIN with repeats: 104=10,00010^4 = 10{,}000. 5 shirts and 3 pants: 5×3=155 \times 3 = 15 outfits.
See it in the Lab: Permutations & Combinations
Watch out: Adding instead of multiplying. · Swapping base and exponent: it's (options)^(slots).

Factorial

n!=n(n−1)(n−2)⋯2⋅1,0!=1n! = n(n-1)(n-2)\cdots 2\cdot 1,\qquad 0! = 1

The number of ways to arrange nn different items in order.

nn
Number of different items
When to use: Arranging all of a set of different items; also the building block of nPr{}_nP_r and nCr{}_nC_r.
Example: 6 friends in a row for a photo: 6!=7206! = 720 orders.
See it in the Lab: Permutations & Combinations
Watch out: Thinking 0!=00! = 0. · Cancelling “!” like a number: 9!6!≠3!\frac{9!}{6!} \ne 3!.

Permutations

nPr=n!(n−r)!{}_nP_r = \frac{n!}{(n-r)!}

Ordered selections of rr items from nn different items, with no repeats.

nn
Items available
rr
Items chosen and ordered
When to use: Order matters: places, ranks, offices, positions, different prizes.
Example: President, VP, secretary from 10 members: 10P3=10⋅9⋅8=720{}_{10}P_3 = 10\cdot9\cdot8 = 720.
See it in the Lab: Permutations & Combinations
Watch out: Using it for unordered groups (overcounts by r!r!). · Dividing by r!r! instead of (n−r)!(n-r)!.

Combinations

nCr=(nr)=n!r! (n−r)!{}_nC_r = \binom{n}{r} = \frac{n!}{r!\,(n-r)!}

Unordered groups of rr items from nn different items. Equals nPr/r!{}_nP_r / r!.

nn
Items available
rr
Items in each group
When to use: Order doesn't matter: committees, hands, toppings, lottery picks.
Example: 3-person committee from 10 members: 10C3=7206=120{}_{10}C_3 = \frac{720}{6} = 120.
See it in the Lab: Permutations & Combinations
Watch out: Using nPr{}_nP_r when order doesn't matter. · Forgetting one of the two factorials in the denominator.

Probability by counting

P(E)=# outcomes in E# equally likely outcomesP(E) = \frac{\#\text{ outcomes in } E}{\#\text{ equally likely outcomes}}

With equally likely outcomes, a probability is a ratio of two counts — count both the same way (both ordered or both unordered).

# outcomes in E\#\text{ outcomes in } E
Favorable count (often a product of combinations)
# equally likely outcomes\#\text{ equally likely outcomes}
Total count
When to use: Random committees, lottery tickets, random codes, random line-ups.
Example: Exactly 3 first-years on a random 5-person team from 7 first-years and 5 sophomores: 7C3⋅5C212C5=350792≈0.4419\frac{{}_7C_3\cdot{}_5C_2}{{}_{12}C_5} = \frac{350}{792} \approx 0.4419.
See it in the Lab: Permutations & Combinations
Watch out: Ordered count on top, unordered on the bottom. · Adding the counts for the two groups instead of multiplying.

Rules of a discrete probability distribution

0≤P(x)≤1,∑P(x)=10 \le P(x) \le 1, \qquad \sum P(x) = 1

Every probability in the table is between 0 and 1, and all of them together add up to exactly 1.

xx
A possible value of the random variable XX
P(x)P(x)
The probability that X=xX = x
When to use: Checking whether a table is a valid PDF, or finding a missing probability.
Example: P(0)=0.22P(0) = 0.22, P(1)=0.35P(1) = 0.35, P(3)=0.12P(3) = 0.12, P(4)=0.06P(4) = 0.06 → P(2)=1−0.75=0.25P(2) = 1 - 0.75 = 0.25.
See it in the Lab: Discrete Distribution Explorer
Watch out: Checking only one of the two rules. · Assuming the values are equally likely.

Cumulative and complement probabilities from a table

P(X≤c)=∑x≤cP(x),P(X≥c)=1−P(X≤c−1)P(X \le c) = \sum_{x \le c} P(x), \qquad P(X \ge c) = 1 - P(X \le c-1)

To find the probability of an event, add P(x)P(x) for every value it includes — or subtract the values it excludes from 1.

cc
The cutoff value named in the question
P(X≤c)P(X \le c)
Cumulative probability: all values up to and including cc
When to use: “At most”, “at least”, “more than”, “fewer than”, “between” questions about a discrete table.
Example: Fewer than 3: P(X<3)=P(X≤2)=P(0)+P(1)+P(2)P(X < 3) = P(X \le 2) = P(0) + P(1) + P(2).
See it in the Lab: Discrete Distribution Explorer
Watch out: Including the endpoint for “fewer than” / “more than”. · Leaving it out for “at least” / “at most”.

Expected value of a discrete random variable

μ=E(X)=∑x P(x)\mu = E(X) = \sum x\,P(x)

Weight each value by its probability and add: the long-run average result.

xx
A possible value (for games, the net gain)
P(x)P(x)
Its probability
μ\mu
Mean / expected value
When to use: Mean of a probability table; expected winnings of a game or raffle; expected profit of an insurance policy.
Example: P(0)=0.2P(0)=0.2, P(1)=0.45P(1)=0.45, P(2)=0.25P(2)=0.25, P(3)=0.1P(3)=0.1: μ=0.00+0.45+0.50+0.30=1.25\mu = 0.00 + 0.45 + 0.50 + 0.30 = 1.25.
See it in the Lab: Is This Game Worth Playing?
Watch out: Averaging the xx values without weights. · Forgetting to subtract the cost to play (use net gain).

Standard deviation of a discrete random variable

σ=∑(x−μ)2 P(x)\sigma = \sqrt{\sum (x-\mu)^2\,P(x)}

The typical distance of a value from the mean, with each squared deviation weighted by its probability.

μ\mu
The mean ∑x P(x)\sum x\,P(x) (find it first)
σ2\sigma^2
Variance — the sum under the square root
When to use: Asked for the variance or standard deviation of a probability table.
Example: Same table: σ2=0.7875\sigma^2 = 0.7875, so σ=0.7875≈0.8874\sigma = \sqrt{0.7875} \approx 0.8874.
See it in the Lab: Is This Game Worth Playing?
Watch out: Reporting the variance as the standard deviation. · Not squaring the deviations (their weighted sum is always 0).

Binomial probability

P(X=x)=(nx) px q n−xP(X = x) = \binom{n}{x}\,p^{x}\,q^{\,n-x}

The probability of exactly xx successes in nn independent trials, each with success probability pp.

nn
Number of trials
xx
Number of successes
pp
Probability of success on one trial
q=1−pq = 1 - p
Probability of failure on one trial
(nx)\binom{n}{x}
Number of orders the successes can appear in (nCx_nC_x)
When to use: A fixed number of independent success/failure trials with the same pp, asking for exactly xx successes. Excel: =BINOM.DIST(x, n, p, FALSE).
Example: Free throws, n=8n = 8, p=0.75p = 0.75: P(X=6)=(86)(0.75)6(0.25)2≈0.3115P(X = 6) = \binom{8}{6}(0.75)^{6}(0.25)^{2} \approx 0.3115.
See it in the Lab: Binomial Explorer
Watch out: Leaving out (nx)\binom{n}{x}. · Swapping the exponents on pp and qq. · Using TRUE (cumulative) in Excel for an “exactly” question.

Binomial “at least / more than / fewer than” translations

P(X<k)=P(X≤k−1),P(X≥k)=1−P(X≤k−1),P(X>k)=1−P(X≤k)P(X < k) = P(X \le k-1),\quad P(X \ge k) = 1 - P(X \le k-1),\quad P(X > k) = 1 - P(X \le k)

Every binomial event can be written using cumulative probabilities P(X≤c)P(X \le c) and complements.

kk
The number named in the question
P(X≤c)P(X \le c)
Cumulative probability, Excel =BINOM.DIST(c, n, p, TRUE)
When to use: Any binomial question worded with at least, at most, more than, fewer than, or between.
Example: Guessing on 12 questions with p=0.25p = 0.25: P(X≥4)=1−P(X≤3)≈0.3512P(X \ge 4) = 1 - P(X \le 3) \approx 0.3512.
See it in the Lab: Binomial Explorer
Watch out: Using P(X≤k)P(X \le k) for “fewer than kk”. · Using 1−P(X≤k)1 - P(X \le k) for “at least kk”.

Binomial mean and standard deviation

μ=np,σ=npq\mu = np,\qquad \sigma = \sqrt{npq}

The expected number of successes is npnp; the typical spread around it is npq\sqrt{npq}.

nn
Number of trials
p, qp,\ q
Success and failure probabilities (q=1−pq = 1 - p)
When to use: Mean, expected number, variance, or standard deviation of a binomial count.
Example: B(12,0.25)B(12, 0.25): μ=3\mu = 3, σ=12(0.25)(0.75)=1.5\sigma = \sqrt{12(0.25)(0.75)} = 1.5.
See it in the Lab: Binomial Explorer
Watch out: Reporting npqnpq (the variance) as σ\sigma. · Using qq instead of pp in the mean.

Continuous probability = area

P(c<X<d)=area under f(x) from c to dP(c < X < d) = \text{area under } f(x) \text{ from } c \text{ to } d

For a continuous random variable, the probability of an interval is the area under the density curve over that interval. The total area is 1, and P(X=c)=0P(X = c) = 0.

f(x)f(x)
Probability density function (a height, never negative)
c,dc, d
Ends of the interval
When to use: Any probability question about a measured (continuous) quantity.
Example: f(x)=120f(x) = \frac{1}{20} on [0,20][0, 20]: P(5<X<12)=7×120=0.35P(5 < X < 12) = 7 \times \frac{1}{20} = 0.35.
See it in the Lab: Area = Probability
Watch out: Reading the height f(x)f(x) as a probability. · Thinking P(X≤c)≠P(X<c)P(X \le c) \ne P(X < c) for continuous variables.

Cumulative probability and complement

P(X>x)=1−P(X≤x)P(X > x) = 1 - P(X \le x)

The cumulative probability P(X≤x)P(X \le x) is the area to the left of xx; the area to the right is what remains of the total 1.

P(X≤x)P(X \le x)
Cumulative (left) area
P(X>x)P(X > x)
Right-tail area
When to use: “More than / at least” questions, or “between” questions built from left areas: P(c<X<d)=P(X<d)−P(X<c)P(c < X < d) = P(X < d) - P(X < c).
Example: If P(X≤8)=0.7P(X \le 8) = 0.7, then P(X>8)=0.3P(X > 8) = 0.3 and P(X≥8)=0.3P(X \ge 8) = 0.3.
See it in the Lab: Area = Probability
Watch out: Reporting the left area for a “more than” question.

Rectangle and triangle areas under a density

rectangle=b×h,triangle=12 b×h\text{rectangle} = b \times h, \qquad \text{triangle} = \tfrac12\, b \times h

Simple density shapes turn probability into geometry: flat pieces are rectangles, straight sloped pieces are triangles.

bb
Base (width of the interval)
hh
Height of the density (tall end, for a triangle)
When to use: The density is drawn or given as straight-line pieces.
Example: f(x)=10−x50f(x) = \frac{10-x}{50} on [0,10][0,10]: P(X>6)=12⋅4⋅0.08=0.16P(X > 6) = \tfrac12 \cdot 4 \cdot 0.08 = 0.16.
See it in the Lab: Area = Probability
Watch out: Forgetting the ½ for triangles. · Using width ÷ total width when the height is not constant.

Uniform probability (area of a rectangle)

P(c<X<d)=(d−c)⋅1b−aP(c < X < d) = (d-c)\cdot\frac{1}{b-a}

For a uniform distribution, a probability is the shaded width times the constant height 1b−a\frac{1}{b-a}.

a,ba, b
Minimum and maximum possible values
c,dc, d
The boundaries of the region you want
1b−a\frac{1}{b-a}
Height of the density, so total area is 1
When to use: Any “less than / more than / between” question about a variable that is equally likely across an interval.
Example: Wait ∼U(0,15)\sim U(0,15): P(X>11)=(15−11)⋅115=415≈0.2667P(X>11) = (15-11)\cdot\frac{1}{15} = \frac{4}{15} \approx 0.2667.
See it in the Lab: Uniform Distribution Explorer
Watch out: Using 1b\frac{1}{b} instead of 1b−a\frac{1}{b-a} when a≠0a \ne 0. · Giving the height as the probability.

Uniform mean and standard deviation

μ=a+b2,σ=(b−a)212\mu = \frac{a+b}{2},\qquad \sigma = \sqrt{\frac{(b-a)^2}{12}}

The mean is the midpoint of the range; the spread grows with the width of the range.

μ\mu
Mean (expected value)
σ\sigma
Standard deviation
When to use: Asked for the average, expected value, or standard deviation of a uniform variable.
Example: U(0,15)U(0, 15): μ=7.5\mu = 7.5, σ=225/12≈4.3301\sigma = \sqrt{225/12} \approx 4.3301.
See it in the Lab: Uniform Distribution Explorer
Watch out: Forgetting the square root (that gives the variance). · Subtracting instead of adding in the mean.

Uniform percentile

k=a+p (b−a)k = a + p\,(b-a)

The value with area pp to its left — work the area formula backwards.

pp
Area to the left, as a decimal (90th percentile → 0.90)
kk
The percentile value
When to use: You're given a percent/area and asked for the value.
Example: U(0,15)U(0,15), 90th percentile: k=0+0.9(15)=13.5k = 0 + 0.9(15) = 13.5.
See it in the Lab: Uniform Distribution Explorer
Watch out: Multiplying pp by bb instead of by the width b−ab - a.

z-score (standardizing)

z=x−μσz = \frac{x - \mu}{\sigma}

How many standard deviations a value lies above (positive) or below (negative) the mean. Standardized values follow Z∼N(0,1)Z \sim N(0, 1) when XX is normal.

xx
The raw value
μ\mu
Mean of the distribution
σ\sigma
Standard deviation (not the variance)
When to use: Measuring how unusual a value is, comparing values from different distributions, or before using a z-table.
Example: X∼N(72,8)X \sim N(72, 8): x=60x = 60 gives z=60−728=−1.5z = \frac{60 - 72}{8} = -1.5.
See it in the Lab: Build a Bell Curve
Watch out: Dividing by σ2\sigma^2. · Computing μ−x\mu - x, which flips the sign.

Value from a z-score

x=μ+zσx = \mu + z\sigma

Start at the mean and move zz standard deviations (left if z<0z < 0).

zz
Number of SDs from the mean
μ,σ\mu, \sigma
Mean and standard deviation
When to use: Given a z-score (or “k SDs above/below the mean”) and asked for the value; also used for normal percentiles.
Example: X∼N(72,8)X \sim N(72, 8), z=2z = 2: x=72+2(8)=88x = 72 + 2(8) = 88.
See it in the Lab: Build a Bell Curve
Watch out: Adding zz without multiplying by σ\sigma. · Going the wrong direction for negative zz.

Empirical rule (68-95-99.7)

P(μ−kσ<X<μ+kσ)≈68%, 95%, 99.7% for k=1,2,3P(\mu - k\sigma < X < \mu + k\sigma) \approx 68\%,\ 95\%,\ 99.7\% \text{ for } k = 1, 2, 3

For a normal distribution, about 68% of values lie within 1 SD of the mean, 95% within 2 SDs, and 99.7% within 3 SDs.

kk
Whole number of standard deviations (1, 2, or 3)
When to use: Quick estimates when values are exactly 1, 2, or 3 SDs from the mean.
Example: X∼N(50,5)X \sim N(50, 5): P(X>60)≈100%−95%2=2.5%P(X > 60) \approx \frac{100\% - 95\%}{2} = 2.5\%.
See it in the Lab: Build a Bell Curve
Watch out: Using it for non-whole z-scores like 1.3. · Forgetting the percent covers both sides of the mean.

Normal probability (left and right tails)

P(X<x)=P ⁣(Z<x−μσ),P(X>x)=1−P(X<x)P(X < x) = P\!\left(Z < \tfrac{x - \mu}{\sigma}\right), \quad P(X > x) = 1 - P(X < x)

Standardize the value, then find the area to the left; the right tail is the complement. Excel: =NORM.DIST(x, μ, σ, TRUE) gives the left area directly.

xx
The value of interest
μ,σ\mu, \sigma
Mean and standard deviation of XX
ZZ
Standard normal, N(0,1)N(0, 1)
When to use: Given a value and asked for a probability, proportion, or percent.
Example: X∼N(30,5)X \sim N(30, 5): P(X>38)=1−Φ(1.6)≈1−0.9452=0.0548P(X > 38) = 1 - \Phi(1.6) \approx 1 - 0.9452 = 0.0548.
See it in the Lab: Normal Distribution Explorer
Watch out: Reporting the left area for a “more than” question. · Dividing by σ2\sigma^2.

Normal probability between two values

P(a<X<b)=P(X<b)−P(X<a)P(a < X < b) = P(X < b) - P(X < a)

Subtract the left area at the smaller value from the left area at the larger value.

a<ba < b
The two boundaries
When to use: “Between”, “from … to”, or (via 1 − between) “outside” questions.
Example: X∼N(30,5)X \sim N(30, 5): P(25<X<38)≈0.9452−0.1587=0.7865P(25 < X < 38) \approx 0.9452 - 0.1587 = 0.7865.
See it in the Lab: Normal Distribution Explorer
Watch out: Forgetting to subtract the lower left area. · Subtracting z-scores instead of areas.

Normal percentile / cut-off (inverse normal)

x=μ+zσ,z=invNorm(left area)x = \mu + z\sigma, \quad z = \text{invNorm}(\text{left area})

Turn a left area into a z-score, then into a value. Excel: =NORM.INV(left area, μ, σ).

left area\text{left area}
Share of values below the cut-off (top 10% → 0.90)
xx
The percentile / cut-off value
When to use: Given a percent and asked for a value: percentiles, “top/bottom p%”, “middle C%”.
Example: X∼N(15,2)X \sim N(15, 2): top 10% cut-off =15+1.2816(2)≈17.56= 15 + 1.2816(2) \approx 17.56.
See it in the Lab: Normal Distribution Explorer
Watch out: Using the right-tail area in NORM.INV (top 10% needs 0.90). · Treating the percentile (a value) as a probability.