Omega Ratio — Return

Ratio of positive to negative daily returns on the Return curve (low-to-low). >1 = net positive expectation.

Computed from: Equity curve
Scope: Single report
Range: ≥ 0
Direction: Higher is better

Basis

Computed on the low-to-low Return Series (money-weighted) — measured day-worst-point to day-worst-point, not close-to-close.

The Omega ratio weighs all the gains an account produced against all its losses, measured against a breakeven line. Where the Sharpe ratio — the popular risk-adjusted score — boils a record down to an average return and how bumpy it was, Omega looks at every single result and adds them up on each side of the line. That lets it see the lopsidedness that an average hides.

Key takeaways

What it is — the total of all the up-results divided by the total of all the down-results, measured against a threshold (the line you call breakeven — here, zero). Above 1 means the ups outweigh the downs.
Read it as — Omega = 1 means gains and losses exactly balance; above 1 is favorable, higher is better. On daily data the numbers sit close to 1, so a reading of 1.3 is already strong — don't read it like a trade-level profit factor.
Why it matters — Sharpe assumes returns are roughly symmetric (wins and losses mirror each other) and rewards a smooth average. Omega adds up the actual results, so it sees skew (when one side has rarer-but-bigger moves) and fat tails (the occasional outsized day) that Sharpe averages away.
In EquityTruth — computed at a zero threshold on the daily low-to-low returns (one return per trading day), the same series that feeds Sharpe and Sortino.

How it's calculated

In plain terms: add up every day the account finished ahead, add up every day it finished behind, and divide the first total by the second. The chart below shows the idea — gains pile up on the right of the breakeven line, losses on the left, and Omega is one pile divided by the other.

Omega = (sum of the up bars) ÷ (sum of the down bars). A right-leaning curve like this — many small losses, the occasional big win — can post a strong Omega even when its Sharpe is only middling.

Omega(θ) = Σ max(r − θ, 0) / Σ max(θ − r, 0)

r: each day's return in the series
θ (theta): the threshold — the return you treat as breakeven. EquityTruth uses θ = 0.
Σ max(r − θ, 0): add up how far every winning day finished above the line
Σ max(θ − r, 0): add up how far every losing day finished below the line

In this product: EquityTruth fixes the threshold at θ = 0, so the numerator is the sum of all positive returns and the denominator the absolute sum of all negative returns. (The textbook Omega is a ratio of areas under the return distribution's curve above and below θ; for a real sample that works out to exactly this ratio of summed gains to summed losses.) It runs on the low-to-low series — one return per trading day, each from a day's lowest equity point to the next day's lowest — the same array used for Sharpe and Sortino. If a record has no losing days the denominator is zero and Omega is infinite.

On the Return curve the underlying returns are money-weighted — deposits and withdrawals move the equity, so this Omega reflects the gain-vs-loss balance you personally experienced, cashflows and all.

What it tells you

Because this is Omega at a zero threshold on daily returns, the numbers cluster near 1 — daily returns straddle breakeven fairly evenly, so even a strong record rarely climbs far above it. Read the bands on that scale, not like a trade-level profit factor:

Value	Reading	Notes
< 1	Losing	The down-total exceeds the up-total — net negative at breakeven.
1.0 – 1.1	Marginal	Gains barely outweigh losses. A thin, fragile edge.
1.1 – 1.25	Decent	A genuine, repeatable surplus of gains over losses.
1.25 – 1.5	Strong	Gains dominate losses by a wide margin for daily data.
> 1.5	Exceptional	Rare on a long record — confirm it isn't a short sample or a not-yet-realized tail.

When to reach for it. Most of the time Sharpe, Sortino, and Calmar carry the load. Omega earns its place when the return distribution is skewed or fat-tailed, when you're judging an option-like / asymmetric payoff (grid, martingale, premium-selling, breakout), or as a tiebreaker between two accounts with near-identical Sharpe — because it respects the shape of the tail that Sharpe collapses into a single number.

Worked example

Suppose a record's daily returns sum to +1.8% across all the days that finished positive and −1.5% across all the days that finished negative:

Omega = 1.8% / 1.5% = 1.2

The up-total is 1.2× the down-total — a decent reading on daily data. Two accounts with the same Sharpe can land here differently: if one earns its surplus through a few large winning days (positive skew), its Omega rises above the other's, because Omega counts the full size of those wins rather than averaging them into a standard deviation.

Pitfalls

Pitfalls & caveats

It depends on the threshold. Omega is defined relative to θ, and EquityTruth fixes θ = 0 — so this is a breakeven Omega (the equity-curve cousin of profit factor), asking only whether gains outweigh losses with no hurdle. A version set to a minimum acceptable return would say something stricter. Never compare two Omega numbers measured at different thresholds.
The tail it loves is the tail that bites — "picking up pennies." The same positive-skew sensitivity that makes Omega attractive is what makes it gameable. Grid, martingale, and premium-selling strategies post a beautiful Omega that is smooth until it isn't — not because their negative skew inflates the ratio (a realized large loss actually lowers Omega, by enlarging the down-total), but because the rare large loss simply hasn't printed yet. A high Omega on a short record is a warning, not a green light; confirm the tail has been tested (a long record that survived a real drawdown). Always pair it with max drawdown.
It needs enough data. Built from the whole distribution, Omega is noisier on short records than a simple mean-based metric, and a record with no losing days returns an infinite Omega — a short-sample artifact, not an unbeatable strategy, not a quality signal. Read it next to the track-record length.

Omega vs Sharpe

Sharpe compresses the return record into two numbers — its average and how bumpy it was — and implicitly assumes the distribution is roughly symmetric. That makes Sharpe blind to skew and fat tails, and it penalizes a big winning day exactly as much as a big losing one.

Omega keeps the whole distribution. It weighs every gain above the line against every loss below it, so the lopsidedness is preserved: a strategy with many small losses and the occasional large win can show a mediocre Sharpe but a strong Omega. For normally distributed returns of similar volatility the two rank accounts nearly identically — they diverge precisely when the returns are skewed or fat-tailed, which is the entire reason to look at Omega.

Omega is the distribution-aware cousin of the Sharpe ratio, and at a zero threshold it is the equity-curve analogue of profit factor (gains-sum over losses-sum, but on daily returns rather than on trades). For the middle ground — return per unit of downside risk, which also stops penalizing upside volatility — see the Sortino ratio.

Compare withSharpe Ratio