What Is Value at Risk (VaR) in Trading?

In short. Value at Risk (VaR) is an estimate of the loss that a position or account should not exceed over a set period at a chosen confidence level. A one-day 95% VaR of $650 means that on roughly 95 days out of 100 the loss should stay below $650, and on about 1 day in 20 it is expected to be larger. VaR describes a threshold, not the worst case, so it is most useful next to a measure of how bad the tail days get.

Banks and funds have used VaR for decades to put the market risk of a whole portfolio into one dollar figure. The same idea works on a much smaller scale: a trader with a few months of daily results in a journal can calculate a personal VaR and compare it with a daily loss limit, an account size, or a prop firm rule. The calculation is simple. Reading it correctly is where most of the value lies.

How VaR is defined

Every VaR figure has three parts, and a number without all three is incomplete:

• An amount: the loss threshold, in money or as a percentage of the account.
• A time horizon: one day, one week, or another holding period.
• A confidence level: usually 95% or 99%.

A research paper from the Federal Reserve Bank of New York describes a value-at-risk model as a measure of how much a portfolio could decline over a given period with a given probability. At the 99% level, VaR is the loss the portfolio is expected to exceed only 1% of the time. The same paper notes that the holding period is a choice, that VaR models assume the portfolio does not change during that period, and that the 95th to 99th percentile range is the one used most often.

For an active trader, the one-day horizon is usually the practical one. Positions change during the week, so a weekly figure built from one fixed set of positions describes a portfolio that rarely exists.

Three ways to calculate VaR

There is no single VaR formula. The method changes the result, sometimes by a lot, so it should always be stated next to the number.

The New York Fed paper explains the historical approach with a concrete case: with 500 days of data, the 99th percentile historical VaR is the sixth largest loss, because 1% of the sample, five days, is expected to fall beyond it. The parametric approach is simpler but leans on the assumption of a normal distribution, which daily trading results often do not follow.

A worked example from a trading journal

Take an illustrative $25,000 account with 100 trading days logged. The average day is a profit of about $57, and the standard deviation of daily results is about $390. The ten worst days in the sample, from the worst, are −$1,240, −$980, −$860, −$790, −$720, −$650, −$610, −$580, −$560, and −$540.

With 100 days and a 95% confidence level, five days are expected to fall beyond the threshold. Historical VaR is therefore the sixth worst day: $650.

Two points stand out. First, the parametric figure is smaller than the historical one, which is the usual direction of the error when a strategy has occasional large losing days. Second, the days beyond VaR are not just slightly worse. The average of those five days is about 41% above the VaR figure, and the worst one is almost double it.

To extend a one-day figure to several days, the parametric approach multiplies by the square root of the number of days. Five-day VaR in this example is about $585 × √5, roughly $1,310. The New York Fed paper points out that this shortcut relies on the assumptions of normality and of independence between days, and that historical simulation does not convert between horizons in such a fixed way.

What VaR does not tell you

VaR answers one question: how large a loss should be rare at this confidence level. It says nothing about how large the rare losses are. That gap is the reason regulators moved on. In the Basel Committee's revised market risk framework, expected shortfall replaces VaR in the internal models approach. The Committee's summary defines expected shortfall as the average of all potential losses beyond VaR at a given confidence level, and lists insufficient measurement of tail risk as one of the issues with the older VaR-based approach.

For an individual trader, the practical limits of VaR are:

• It looks backwards: a quiet three months produces a low VaR right before volatility returns.
• It needs enough data: with 30 days, a 95% figure rests on one or two observations.
• It ignores gaps and leverage changes: a weekend gap or a larger position size is not in the historical sample.
• It is not a drawdown measure: VaR covers one period, while drawdown tracks how losses accumulate from a peak over many periods.

How traders use VaR in practice

The most direct use is a comparison with a hard limit. If an account has a daily loss limit of $1,000 and the 95% one-day VaR is $650, the limit sits above the threshold, but the sample in this example still breached it once in 100 days. If VaR were close to the limit, a breach roughly once every 20 trading days would be the expected result of the strategy as it is traded, not bad luck.

VaR also helps check whether position sizing has drifted. If risk per trade is supposed to be stable, a VaR that keeps climbing over several months signals that size or exposure has grown without a deliberate decision. Recalculating it on a rolling window, for example the last 100 trading days, makes that drift visible.

Finally, VaR fits into a broader set of risk metrics rather than replacing them. The Sharpe ratio relates return to overall volatility, VaR focuses on the bad end of the distribution, and expected shortfall describes what happens past that point. Together they give a far more honest picture than any one of them alone, which is also the core idea behind risk management rules that hold.

This article is for educational purposes only and is not financial or investment advice. All figures above are an illustrative example, not a benchmark or a recommended setting. Trading, especially with leverage, carries a high risk of losing some or all of your capital, and no risk measure can predict or prevent losses.

A performance dashboard built on daily results from your journal makes it easier to see the size of your worst days, how often they happen, and whether they are growing over time.