> For the complete documentation index, see [llms.txt](https://help.hydravision.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.hydravision.ai/products/live-analytics-and-hydra-agent/live-analytics/analytics-page.md). # Analytics page The Analytics page looks at the *shape* of your trading over time: when grids close, how long they run, how each symbol behaves, and — if a backtest report has been added for this account — how live trading compares with that backtest over the same dates. #### Backtest bar If a backtest report has been added for this account in your portal, a bar at the top describes it: | Item | Meaning | | -------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Compared against** | The name of the backtest report. | | **Symbol** | The symbol and timeframe the backtest ran on. | | **Period** | The first and last day of the backtest. | | **Strategy Print** | The backtest's Strategy Print: a short code that fingerprints the settings that decide what the strategy does. The same strategy settings always give the same code. | The **Backtest** button opens the full backtest report. {% hint style="info" %} With no backtest report, the bar is replaced by a note: "Add a backtest report for this account in your portal to compare it with your live trading over the same dates." The Backtest vs Live panel below only appears once a report has been added. {% endhint %} #### Backtest vs Live This panel puts the backtest next to your live trading on the **same symbol and the same dates**. The line in the panel header names the symbol and the dates used. Only live grids on that symbol, inside the backtest's period, are counted. Broker suffixes are ignored, so GBPUSD, GBPUSD.r and GBPUSD-ECN all count as GBPUSD. **Summary** — one row per measure, with **Backtest**, **Live** and **Live − Backtest** columns: | Measure | Meaning | | ----------------------------- | ------------------------------------------------------------------------------------------------------------------- | | **Realised net profit** | Profit after commission and swap from everything closed in the period. | | **Costs (commission + swap)** | Commission and swap together. These are usually negative, because they are a cost. | | **Grids closed** | Number of grids that fully closed, with the buy / sell split shown as **B** / **S**. | | **Grid win rate** | Share of closed grids that ended in net profit. The difference column shows the gap in percentage points (**pts**). | | **Avg net per grid** | Average net profit of one closed grid. | | **Max depth (positions)** | The most positions any single grid held at once. | | **Worst grid drawdown** | The deepest open loss any single grid reached. | | **Avg grid duration** | Average time from a grid opening to fully closing. | **Net by engine** — each engine's net profit in money, for the backtest and for live, with the difference. Engines that lost money are included. Engine names are the same as on the other pages (for example **PCL Engines**, **Position Hunter**, **Anchor SL**). **not reported live** means that type of close is not recorded on a live account, so there is no live figure to compare. Notes under the tables explain what the figures leave out: * Positions still open when the backtest ended are not counted on either side, and their grid is left out of the grid figures. Closes that grid made before the end are still in realised net profit. * Backtest times are broker server time and live times are UTC. A close within a few hours of the start or end of the period may be counted on one side only. * On an account with a long history, the earliest part of the period may be missing from the live figures. A note says so when this happens. How to read it: the backtest and live columns cover the same market over the same dates, so a large gap shows where live trading behaved differently from the test. Small gaps are normal, because live prices, spreads and costs never match the tester exactly. The Net by engine table shows which engines the gap comes from. {% hint style="warning" %} Reports made by older versions of Hydra Vision can't be compared. The panel then shows: "This backtest report is from an earlier version of Hydra Vision and can't be compared with live trading." Running the backtest again on the current version and adding the new report makes the comparison available. If there was no live trading on that symbol between the backtest's dates, the panel says so and the Live column stays empty. {% endhint %} #### Grid Stats (all assets) | Row | Meaning | | ---------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Grids Completed** | Total number of grids that have opened and fully closed. | | **Grid Win Rate** | Percentage of those that closed in net profit. | | **Grid Profit Factor** | Total profit from winning grids ÷ total loss from losing grids. Above 1.0 means winners outweigh losers; 2.0 means you made twice what you lost. | | **Avg P\&L / Grid** | Average net profit per grid. | | **Return / Lot** | Net profit per lot traded. | | **Anchor-Only Rate** | Percentage of grids that closed with only their first position — they never needed to add a second. A high rate means the market is usually moving your way straight away. | | **Avg Duration** | Average time from a grid opening to fully closing. | #### Grid Risk (all assets) | Row | Meaning | | ----------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------- | | **Worst Drawdown** | Largest floating loss any single grid reached. | | **Avg Drawdown** | Average of each grid's worst floating loss. | | **Widest Grid** | The largest price range (in pips) any grid spanned from its first position to its furthest. | | **Avg Depth** / **Max Depth** | Average and maximum number of positions grids have held at once. | | **Peak Density** | Average of each grid's peak density — how crowded the grid's positions became at its busiest point. Higher means positions were packed more tightly. | #### Streaks (all assets) Runs of consecutive winning or losing grids. **Max Win Streak** and **Max Loss Streak** are the longest runs; **Avg Win Streak** is the typical length of a winning run, rounded to a whole number of grids; **Best** and **Worst Streak P\&L** are the most made or lost during a single run. A long loss streak with a small Worst Streak P\&L means the losses were small individually. #### PCL Engine Breakdown The same engine donut as on the Overview, with hover details including average positions closed per fire and total volume closed. #### Grid Duration Distribution Bars showing how many grids fell into each duration band: **< 1h, 1–4h, 4–12h, 12–24h, 1–3d, 3–7d, 7d+**. Hover a bar for that band's net P\&L, average P\&L and average depth. This tells you whether your profit comes from quick grids or long-running ones — and whether the long-running ones are worth their tied-up margin. #### Grid P\&L by Hour (UTC) and by Day of Week Net P\&L from grids grouped by the **hour** or **weekday** they closed. Hover for net P\&L, average P\&L and the number of grids closed in that slot. It shows which sessions or days have added to or taken from the account. Note that the hourly chart is in **UTC**, not your local time. #### Per-Symbol Grid Performance One row per symbol traded: number of grids, win rate, profit factor (**PF**), average P\&L, average and max depth, return per lot and average drawdown. Click any heading to sort — for example by PF to see which symbols are pulling their weight. #### GridSense Intelligence — Buy vs Sell The same statistics split into **Buy** grids and **Sell** grids side by side: profit factor, grid return, win rate, net profit, average drawdown, depth, duration and peak density. A clear imbalance (say, sells consistently deeper and slower than buys) is often the first sign that one direction is fighting the trend. #### Explore Explore works on **every closed grid on the account, across all symbols**. It groups the grids one way, then shows the chosen measures for each group as a chart and a table. The panel header shows how many grids are loaded and the dates they cover, in UTC. | Control | What it does | | ------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Group by** | How the grids are split: **All grids**, **Month opened**, **Day opened**, **Hour opened**, **Side**, **Symbol**, **How it closed**, **Deepest point (positions)** or **Time open**. Month, day and hour use UTC. | | **Show as** | **Bar**, **Line** or **Table only**. | | **Symbol**, **Side** | Limit the grids to one symbol, or to buy or sell grids only. | | **Opened between (UTC)** | Limit the grids to those opened between two dates. | | **Measures** | The figures to show, for example **Grids**, **Net profit**, **Net per grid**, **Winning grids**, **Profit factor**, **Peak drawdown per grid**, **Worst grid drawdown**, **Depth per grid**, **Deepest grid**, **Time open per grid**, **Longest grid**, **Volume traded** and **Commission and swap**. Hover a measure for its description. | | **Closed by** | Limit the grids to those that closed a certain way (for example **PCL Engines** or **Anchor SL**). With none selected, all close types are included. | | **Saved views** | Name the current set-up and **Save view** to reuse it later; **Delete** removes a saved view and **Reset** goes back to the default. Saved views are stored with your account, so they are available on any device, and the page reopens on the last view you used. | | **Download CSV** | Downloads the table as a CSV file, which opens in a spreadsheet. | How to read it: the line above the table shows how many grids match the filters (for example "120 of 450 grids"). The chart shows the first two measures; the table shows all of them, with a **Total** row when the grids are split into groups. Some measures are greyed out because live trading does not record them (for example **Commission** and **Swap** separately, or **Best open profit per grid**); hovering one gives the reason. Only the most recent 5,000 grids can be loaded, and a note says so when older grids are left out.