> 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/concepts/concepts/risk-management.md).
# Risk Management
Hydra Vision was designed from the ground up as a risk first platform.
Good risk management is foundational in auto trading. It is insufficient to rely purely on drawdown levels as an example, and so Hydra surfaces a variety of configuration options, that when combined can provide a solid risk framework suitable for auto trading.
{% hint style="danger" %}
Nothing can replace good human oversight, when it comes to risk management. We advocate using solid judgement and manual over-rides when any auto trading system is unable to sufficiently deal with unfavorable market conditions.
{% endhint %}
## Position sizing (Volume)
The most obvious yet often overlooked risk control is exposure, which starts with position sizing. It is strongly recommended to use starting and incremental volumes that are suitable for your account capitalization, leverage, risk tolerance and asset being traded. We do not make any recommendations on volume as this is a personal decision which needs to be made, and what works for one, may not work for another. Use comprehensive testing (or demo trading), to provide you with enough data on typical exposure which your strategy can generate.
Look carefully at how positions scale as the grid extends, and importantly if this scaling is supportive or unnecessary. As an example, many grid based system use Martingale as an approach to provide fast grid closures with low retrace requirements, however the aggressive scaling nature of Martingale is often contradictory to efficient and safe risk controls.
## Grid strategy and position scaling
Choosing a grid strategy which is complimentary to volume choice is very necessary. We do not randomly choose a grid strategy in the hope that it will work. If our volume choice is for slower gradual scaling, we need a grid strategy that can work effectively with the volume of the re-entry positions. Hydra offers 3 fixed grid variants and dynamic grids, which all have their place in the risk portfolio.
The difference between fixed and dynamic position entries can be quite significant. Permissive grid setups such as fixed, provide more fire power for PCL, however when the market extends fast, these positions can quickly generated unwanted drawdown. Dynamic grids on the other hand provide less opportunity for PCL, and conversely in an adverse market generate comparably lower drawdown than fixed grids. The tradeoff may be how long grids remain open (a key consideration for high swap rate assets).
## [Risk Zones](/ea-configuration/ea-settings/grid.md#risk-zone-setup)
Zones play a crucial role in dynamically adjusting both PCL and SmartEntry depending on the Risk Zone the EA is executing in. Zones represent the potential risk involved.
As a loose definition, we could define the zones and behaviors as follows:
### :traffic-light-go: Green Zone
Permissive re-entry positions, with less aggressive PCL closures allowing for more profit taking opportunity setups.
### :traffic-light-slow: Yellow Zone
Tighten up on re-entry positions, be more selective and assured of potential retrace potential. Provide PCL with high confidence positions to compact the grid.
### :traffic-light-stop: Red Zone
Open new positions very cautiously, look for high quality entries as volume is higher. PCL needs to transition into a de-risk mode, to remove potentially high risk mid-grid positions that will attract high drawdown with an adverse market extension.
Visualization of pre-set risk zones based on 20 pip step size
## Scaling Profit
Seeking profit is a double edged sword. When grids and markets are behaving its time to profit, however when we need to de-risk our positions, scaling back on profit is the right strategy.
Lowering profit hunting when market conditions are extending, is a quick remedy to provide PCL with optimal quick closeout opportunities, thereby compacting runaway grids, and reducing exposure.
Sometimes, even a reduction of 10 pips in the retrace requirement can make the difference between a safe grid and an out of control grid. Hydra Vision provides [3 scaling options](/ea-configuration/ea-settings/pcl.md#profit-mode) to manage PCL profit hunting.
## PCL De-Risking
PCL maintains its own internal de-risking rules, which can be enabled/disabled separately with **Enable PCL De-Risk Engine** (on by default).
PCL runs to a configured strategy, curating the grids into profit hunting setups. De-Risk overrides these setups, and switches to closing high risk positions, usually those clustered around the grid VWAP point.
The settings below work together to trigger de-risking, they do not each trigger it on their own. A grid is de-risked when:
* there is a **Volume Blowout**. This is enough on its own, in any risk zone and on any grid type, or
* the grid is not recovering, it is in the Red risk zone, and:
* **Bounce:** nothing more is needed, the Red zone is enough.
* **Hunter and Multi Trail:** the grid also holds at least **Min positions for DeRisk** positions, its density is above **Grid Max Density % (0-100)**, and it is a dynamic grid. A fixed grid running Hunter or Multi Trail only de-risks on a Volume Blowout.
A grid counts as recovering once its position count has come down from its peak by at least **Risk Recovery - Min count reduction from peak (as %)**.
### [Volume Blowout](/ea-configuration/ea-settings/pcl.md#de-risk-qualifier-volume-blowout-x-start-volume)
When the Frontier position volume is greater than Starting Volume x **Volume blowout (x startVol)**.
For example, Starting Volume is 0.02 and your multiplier is 5. So, when the Frontier volume is greater than 0.1, PCL will de-risk. A Frontier of exactly 0.1 does not trigger it.
### [Min Positions](/ea-configuration/ea-settings/pcl.md#de-risk-qualifier-min-positions-for-de-risk)
A simpler qualifier: a Hunter or Multi Trail grid in the Red Zone must hold at least **Min positions for DeRisk** positions before it can de-risk. On its own it does not trigger de-risking, and Bounce does not use it.
### [Grid Max Density](/ea-configuration/ea-settings/pcl.md#de-risk-qualifier-density-floor-dynamic-grids)
In dynamic grids density plays a key role in grid health, and works to avoid position clustering. When the Density of the grid positions exceeds **Grid Max Density % (0-100)**, and the other conditions above are met, a Hunter or Multi Trail grid will de-risk. Density is only checked on dynamic grids, and Bounce does not use it.
This is a different setting from **Max Market Density % (0-100)** in the Anchor Regime settings, which is an entry check on new positions.
## General Grid Risk configurations
We have introduced some additional risk configurations, which would generally over-ride other EA functions.
### No Profit during de-risk
Curtailing on profit taking during de-risk operations is a quick win to bring grids back under control.
### Max drawdown before auto-close
{% hint style="danger" %}
To be used with extreme caution
{% endhint %}
This only works when **Enable Hard Stop (force close on max loss)** is switched on (it is off by default). **Max Drawdown before auto close (negative)** is a money amount, and it is checked separately for the buy grid and the sell grid: when one grid's floating loss goes beyond this value, that grid is closed. The breach has to hold for a few ticks in a row first, and the EA does not close while the spread or trading conditions are poor; it waits and checks again.
After the close, the EA goes into a graceful stop on that pair: no new grids open, and a grid still open on the other side keeps being managed until it closes. Once the pair has no open positions left, the EA removes itself from the chart.
**Warn at % of max loss (0 = no warning)** can send a warning notification before the limit is reached. From that warning on, no new grids open on the pair, while open grids are still managed.
### Extreme DD (new grids only)
**Extreme DD Veto** is a SmartEntry check on new grids (off by default). While the account's open drawdown (what the positions open right now are losing, on every symbol, as a % of balance) is above **Extreme DD Threshold %**, no new grid opens. Closed losses do not count. Grids that are already open keep running and adding positions, and nothing is closed. Once the open trades recover or close, new grids can open again.
Extreme DD is not the hard stop. Extreme DD looks at the whole account and only holds back new grids. The hard stop above looks at one grid's floating loss on this pair, closes that grid and stops the EA on the pair.
### Max Positions
When grid count reaches this value, no new positions will be opened. Ensure a complimentary strategy setup when using this risk configuration. A lower value here can leave grids 'stranded' for a long period of time.
### Maximum Missing Steps Gap
A slightly more sophisticated risk option. When running with dynamic grids or Cooldown Timing, we can end up with larger than expected gaps in our grid (the difference in price between the last position, the Frontier, and the market price, with no new position opened).
When we experience these gaps, it can be the market silently informing the EA, that it is high risk and may be safer to exit than to try recover. When price moves more than this many steps beyond the Frontier, all positions in that grid will close. A step here is the spacing the grid is using at that point: the Minimum Step Size, or the ATR spacing when an ATR **Re-Entry Spacing Mode** is in use. For example, a setting of 5 with a 20 pip step closes the grid once price is more than 100 pips beyond the Frontier.
Only that grid is closed. The EA stays on the chart, there is no graceful stop, no injection position is opened after the close, and a new grid can open when the normal entry checks pass.
