Loading your account

How to read a trader's stats: what a 30-day whale record actually contains

In shortTo read a trader's stats, start with realized dollars, then check what each percentage divides by and how many closes stand behind each rate. On 2026-10-02, one white whale on our board realized $66,719 over 30 days, an 8.5% return on the $781,563 its account held when the window opened. Every field has a basis, and every field has a limit.

The quickest way to learn how to read a trader's stats is to walk one real card in order: the realized dollars first, then what each percentage divides by, then how many closes stand behind each rate, and only then the colour and the habits. Every field on a whale card has a window and a basis, every rate has a denominator, and every field leaves something out.

The card here belongs to a white whale on our board, 0xf0394ffb…, and every value is as our live API served it on 2026-10-02. The figures will have moved by the time you read this; the definitions will not. Every return on the card is the whale's own, realized, on its own whole account, spot and perps together, and never a follower's.

How to read a trader's stats, field by field

Reading a trader's stats means reading each field with its own window and its own denominator, because no two fields on a whale card measure the same thing. The table lists the fields behind this whale's card on 2026-10-02, what each one measures, and what it does not tell you.

FieldValue on 2026-10-02What it measuresWhat it does not tell you
Realized profit, 30 days+$66,719The whale's net realized dollars on closed positions, rolling 30 daysAnything still open; any follower's result
30-day return8.5% of $781,563Realized dollars divided by the account at the start of the windowWhat the account holds today
Account value now$679,615The whole account today, spot and perps togetherWhy the account moved
Win rate, 30 days78.3% of 46 closesThe share of eligible closes that made moneyHow big the wins and losses were
Win rate, engine's ninety-day figure63.3% of 109Our engine's admission win rate: a win only after fees and funding, over ninety daysHow it compares with the 30-day rate, which counts a win differently
Max drawdown, 30 days0.0%The deepest dip in the realized return curveLosses still open on the account
ColourWhite, at 4.5%The worst single position on whole capital, as a bandWhether the next month makes money
Down from high0.0%Distance below the high of the closed-trade P&L curveMoves in account value that were not trades
Median hold, engine's ninety-day figure40.3 hoursHow long the middle position stays openHow fast the quickest positions close
Signals a day0.9Signals per day from this whale, 30-day averageWhether they arrive in clusters
Trades and eligible closes, 30 days53 trades; 46 eligible closesHow busy the whale was, and what the rates stand onWhich trades the rates leave out
Verified days132How far back our own measured record of the whale goesAnything before that

Realized dollars: what is the headline number?

The headline is the whale's net realized profit in dollars over a stated window: +$66,719 in the 30 days to 2026-10-02 for this whale. Realized means closed, money the whale's positions actually booked, not what they showed while they were still open.

Dollars lead because they need the least interpretation. A closed position's result is a fact about the whale's own account, not a ratio that depends on a denominator, and it tells you the whale trades at real size. The dollars do not tell you how big the account is, which is why the return sits beside them, and they say nothing about a follower, who enters later, at a different size, and pays different fees. Why followers lose even when the trader wins covers that gap.

A peak is the other number a record can lead with, and on a whale's record it never leads. A peak is the most a position was ever up before it closed; nobody captures it, so where a peak profit figure appears it is labelled peak, sits beside realized, and is never added to it. On 2026-10-02 our API served no 30-day peak figure for any whale, so there is none to quote here.

Our board shows why peak never leads. Across 14,114 positions our live whales closed in the 30 days to 2026-10-02, the reconciled, untruncated ones with a recorded peak and trough, measured as the price move from the whale's own entry to its own exit, the median peak was 1.4% and the median realized move was 0.2%. That gap is exactly what a record that leads with peaks hides.

Why does the return divide by the account at the start of the window?

The 30-day return divides the realized dollars by the account as it stood when the window opened: $66,719 on $781,563, or 8.5%. The start of the window is the honest denominator because it is the capital the whale was trading with while it earned those dollars.

The example whale's account read $679,615 on 2026-10-02, less than at the start of a window in which its closed trades made money. Realized profit cannot explain that fall; money taken out, or open positions and holdings marked lower, can, and the card cannot tell you which. Dividing the same dollars by today's smaller account would print a bigger return for no trading reason at all, so every return over a window divides by the account at the start of it.

The same logic is why a withdrawal is not a drawdown. An account-value curve falls when the owner takes money out, so a drawdown measured on equity would mark a whale down for withdrawing its own profit. Our "down from high" figure is measured on the closed-trade P&L curve instead, never on equity, and for this whale it read 0.0% on 2026-10-02: its realized P&L stood at its high. The limit runs the other way as well, because a loss still open on the account shows up in none of these realized fields until the position closes.

Account size is part of the basis. A return on almost nothing is not a return, so a whale needs an account of $100,000 or more on whole capital, re-checked every night; one that stays under it three nights running is paused.

Why does a win rate need its n, and why do two win rates disagree?

A win rate is a fraction, and the card prints both halves: this whale won 78.3% of 46 eligible closes in the 30 days to 2026-10-02. Without the 46, the 78.3% could stand on a handful of trades or on hundreds, and those are different claims.

A rate's n tells you what was left out. The record counts 53 trades in the window, but the rates are computed on the 46 eligible closes, so the trade count and the rate do not stand on the same population. Wherever both appear, the rate's own n is the one to read.

Behind the card, our engine keeps a ninety-day win rate for the same whale: 63.3% of 109 positions on 2026-10-02. Both figures are true, but they do not count a win the same way. The 30-day rate counts a close as a win when the whale's realized P&L on it is above zero; the engine's figure, the one admission uses, counts a win only when the price move, net of funding, clears an allowance for fees. A small winner passes the first test and fails the second, so the gap between 78.3% and 63.3% is largely a difference of definition, not evidence that the last month beat the longer run.

For scale, the median live whale's ninety-day win rate on 2026-10-02 was 60.5%, and the median count behind those rates was 82 positions. Across the 187 live whales with both a 30-day win rate and a 30-day return, the Spearman rank correlation between 30-day win rate and 30-day return was 0.24, a weak link. Win rate counts trades, not dollars, and our piece on risk-to-reward versus win rate works through why a trader can win often and still make little.

Two drawdowns on one card: what do the colour and the curve measure?

The card carries two drawdowns that measure different things. The colour is a band for the deepest drawdown a single position put on the whole account, and the 30-day max drawdown is the deepest dip in the whale's realized return curve; on 2026-10-02 this whale read 4.5% on the first and 0.0% on the second.

The colour comes from the engine's ninety-day figure. At 4.5% of whole capital, this whale's worst single position falls in the white band, which covers anything under 5%, and a colour needs at least 50 measured positions behind it. The bands are published, and the board on 2026-10-02 split across them like this:

ColourWorst single position, % of whole capitalLive whales on 2026-10-02
Whiteunder 5%60
Green5–10%32
Yellow10–15%25
Orange15–20%22
Red20% or more60

The 30-day max drawdown measures something else: the largest fall in the daily realized return curve over the window. For this whale it read 0.0%, so at one decimal the realized curve never dipped in the 30 days to 2026-10-02. The whale did not win every close, as its 30-day win rate shows; the closes that did not win simply never pulled the curve down by enough to show at one decimal.

Neither drawdown is a forecast. A colour is a statement about risk taken, not about returns to come: a white whale risks less of its account per position than a red one, and that is the entire claim. Our methodology note explains how the colour is measured and what it does not promise.

Hold time and signals a day: can you follow this whale at all?

Median hold and signals a day answer a practical question: is this whale's trading slow enough, and sparse enough, for a person to act on? On 2026-10-02 its median hold over the engine's ninety days was 40.3 hours, and it produced 0.9 signals a day over the 30-day window.

This whale's hold and signal rate both sit well above the middle of our board. On 2026-10-02 the median live whale's median hold was 17.2 hours, and the median live whale with a 30-day signal rate produced 0.4 signals a day. Our admission rules require a median hold of at least two hours, a copyability rule rather than a quality one: a whale whose positions close before a follower can act can look excellent on paper and still be impossible to follow.

Hold time and signals a day have limits of their own. A median hides the spread, so half of this whale's positions closed sooner than 40.3 hours and the figure does not say how much sooner; 0.9 a day is an average, and signals can arrive in clusters. Acting on every signal is a choice, not an obligation, and our guide on using whale alerts without overtrading explains why most alerts should produce no trade at all.

Verified days and the '—' rule: when is a number withheld?

Verified days says how far back our own measured record of a whale goes, and a "—" says a rate was withheld because too few closes stand behind it. This whale had 132 verified days and 46 eligible closes on 2026-10-02, so none of its 30-day rates was withheld.

Verified days limits what the headline can honestly claim. Under 30 verified days, the headline falls back to the verified period and says so, rather than presenting a short record as a full 30-day trading record.

The '—' rule handles thin samples. With fewer than 5 eligible closes in 30 days, the 30-day rates are withheld and print "—", never 0: a zero would be a false claim, and a perfect rate on a single close is a coincidence printed as a statistic. On 2026-10-02, 11 live whales had their 30-day rates withheld, on a board of 199 live whales, and a dash on any of them is the card declining to compute a rate it cannot support.

What does a whale card not tell you?

A whale card tells you what one trader did with its own money; it does not tell you what you would make. Every return on it is the whale's own, realized, on its own whole account, and we never state, estimate or imply a follower's result, because a follower enters later, sizes differently, pays its own fees and chooses its own exit.

We never place an order and never hold funds, so what you do with a signal, and when you exit, is yours. Read in order, the fields of any crypto trader's track record stop competing: the dollars say what happened, the denominators say what it was measured against, the counts say how much weight it bears, and the colour and habits say what following it would ask of you.


Every figure above is a whale's own realized result on its own account, over closed trades only. Peak is the most a trade was up before it closed — nobody captures it, and it is never added to anything. A colour is a measured drawdown band, not a promise. How we measure.

Common questions

What is the most important number on a trader's stats card?

Realized profit over a stated window comes first, because it is money the trader's closed positions actually booked rather than a ratio or a peak. On 2026-10-02, the white whale we walked through realized $66,719 over 30 days. Read the return beside it next, and check what it divides by: here, 8.5% of the $781,563 the account held when the window opened.

Why does a whale's return divide by the account at the start of the window?

The start of the window is the capital the whale was trading with while it earned the dollars. Account value also moves for reasons other than closed trades: money taken out, or open positions marked lower, can shrink an account while its closed trades make money. Our example whale's account read $679,615 on 2026-10-02, below the $781,563 it started the window with, and dividing by the smaller figure would inflate the return.

What does "78.3% of 46" mean on a whale card?

A win rate written as "78.3% of 46" means the whale won 78.3% of the 46 eligible closes in the 30 days to 2026-10-02; the number after "of" is the count the rate stands on. Without it, a win rate could rest on a handful of trades or on hundreds. The same whale's ninety-day win rate in our engine, which counts a win only after fees and funding, was 63.3% of 109 on 2026-10-02.

What does a "—" mean instead of a number on a whale card?

A dash means the rate was withheld because too few closes stand behind it. With fewer than 5 eligible closes in 30 days, the 30-day rates on our whale cards print "—", never 0, because a rate on so few closes is closer to a coincidence than a statistic. On 2026-10-02, 11 live whales on our board had their 30-day rates withheld.

What does a whale's colour mean?

A colour is a measured drawdown band: the deepest drawdown any single position put on the whale's whole account. White is under 5%, green 5–10%, yellow 10–15%, orange 15–20%, and red 20% or more. The whale in this walk-through was white on 2026-10-02, at 4.5%. A colour describes risk taken, not returns to come, and it is never a promise about next month.

Do a whale's stats show a follower's result?

No. Every dollar figure and return on a whale card is the whale's own, never a follower's. A follower enters later, at a different size, pays its own fees and chooses its own exit, so we never state or imply a follower's result. DarkTrade is intelligence, not execution: we never place an order and never hold funds, and the exit is always yours.

Everything above is measured from wallets you can check yourself on the Whales tab.

Whale data is public on-chain · not financial advice