Odds and probability: two ways to describe uncertainty
Probability and odds both measure how likely an event is, but they express the idea in different forms.
If an event \(A\) occurs with probability \(p=P(A)\), then \(1-p\) is the probability that it does not occur.
Probability is often the most intuitive scale because it lives between 0 and 1 (or 0% and 100%).
Odds, on the other hand, compare the “chance it happens” to the “chance it doesn’t.”
Odds in favor vs odds against
The most common mathematical definition is odds in favor:
\[
\text{odds in favor}=\frac{p}{1-p}.
\]
If the odds in favor equal 3, that means the event is three times as likely to happen as not.
A related ratio form is written as \(a:b\), meaning “\(a\) favorable outcomes for every \(b\) unfavorable outcomes.”
In that case the implied probability is
\[
p=\frac{a}{a+b}.
\]
For example, odds \(3:1\) in favor give \(p=\frac{3}{3+1}=0.75\).
Some contexts use odds against, defined as
\[
\text{odds against}=\frac{1-p}{p}.
\]
If the odds against are \(a:b\), interpreted as “\(a\) unfavorable for every \(b\) favorable,” then
\[
p=\frac{b}{a+b}.
\]
The tool supports both interpretations to avoid the most common source of confusion: whether the ratio is “for” or “against.”
Decimal odds (betting) and implied probability
In sports betting and racing, odds are often shown as decimal odds \(D\).
Under a simple “fair odds” interpretation, \(D\) corresponds to an implied probability
\[
p=\frac{1}{D}.
\]
This is useful for quick conversions, but real bookmakers typically include a margin (sometimes called an overround),
so \(1/D\) may be slightly higher than the true probability. Still, it is a standard way to compare odds across events.
Log-odds (logit): the university-level extension
A powerful transformation is the log-odds, also called the logit:
\[
\operatorname{logit}(p)=\ln\!\left(\frac{p}{1-p}\right).
\]
Log-odds map probabilities \((0,1)\) onto the entire real line \((-\infty,\infty)\).
This is why logistic regression models the log-odds as a linear function of predictors:
it turns a bounded probability problem into an unbounded scale where linear methods behave well.
When \(p\) is close to 0, the logit is very negative; when \(p\) is close to 1, the logit is very positive.
How to use this tool
Choose whether you want to convert from odds to probability or probability to odds.
In ratio mode, enter \(a\) and \(b\) and pick whether the ratio is “in favor” or “against.”
In decimal mode, enter \(D\) to compute the implied probability \(1/D\).
For probability inputs, the calculator returns odds in favor, odds against, decimal odds, and (optionally) log-odds.
The bar visualization helps you see the split between favorable and unfavorable portions, making the conversion feel concrete.
