Analytics
Breakdowns by coin, side, weekday, hour and duration. Filters recalculate instantly, without a reload.
No trades for this filter
Pick another period or reset the filters.
ResetMethodology — how we calculate the numbers
- P&L is what the exchange reports. For futures we take the closed P&L calculated by the exchange itself and don’t recalculate it. Fees and funding are shown separately; “Gross” is P&L on price alone, without them.
- A trade’s month and day are set by its close. P&L is realized when you exit the position. Days and months are counted in your time zone (Settings).
- Win rate keeps breakevens in the denominator. Breakeven trades (absolute P&L no larger than the fee) are not thrown out: counting only wins against losses would look better and be dishonest. The tile color follows the sign of expectancy, not a 50% threshold: with a 35% win rate and a good win/loss ratio, trading is profitable.
- Drawdown is in dollars, from the peak of cumulative P&L. We show a percentage only relative to equity (deposit plus P&L), when the account balance is known: a percentage from the peak of a P&L curve without the deposit is misleading.
- Money means dollar stablecoins only. USDT, USDC and similar are summed as dollars. Inverse contracts with P&L in coin (BTC, ETH) appear in the list but are excluded from totals.
- Weekday, hour and session are by entry. The decision is made at entry, so time breakdowns are based on it.
- Change on tiles. With a period selected, it is compared with an equally long stretch right before it (30 days vs the previous 30): money and R:R as a difference, win rate in percentage points, trade count in percent. For “all time”, it is the change of the same metric over the last 30 days (“30d”): the value now minus the value for trades closed before then. No earlier trades — a dash.
- Avg R:R. Average win divided by average loss (payoff): “1 : 1.5” means a winning trade brings on average one and a half times what a losing one takes. This is not the planned R-multiple — the “Average R” tile shows that.
- Return and balance. Return percentage and the balance curve need a deposit. We don’t store exchange balances yet, so they show an honest dash, and P&L is shown in dollars.
- P&L by exchange. The ring is split by the absolute P&L of each exchange: a losing exchange also takes its share. The sign of P&L is in the legend.
- Risk breakdown. The loss for the period (sum of losing trades) is split into three parts by the trade plan: “at stop” — lost no more than the planned risk (R ≥ −1.1), “beyond stop” — lost more, “no stop” — there was no plan. Below, separately: the share of the loss in trades with leverage above 20× and the share of the loss from the single worst coin — these overlap with the ring.
- Best hours. P&L by entry hour: in UTC and in your time zone. The decision is made at entry, so we count by it.
- Analytics breakdowns use the filtered trades. Bars: P&L is the sum for the bucket; "Win rate" is the share of winners, colored by the sign of the bucket's expectancy (like the win rate tile); "Trades" is the count. The "best" win rate is picked among buckets with 5+ trades. Hours are in UTC or your time zone (toggle), weekdays in your time zone, sessions in UTC; all by entry time.
- P&L distribution uses 14 equal ranges. The scale ends are the 2nd and 98th percentiles of P&L: one big trade does not stretch the scale, it lands in the edge bar. So the edge labels are not the minimum and maximum.