How do you decide when to pause or stop an algo strategy?

Risk & Drawdowns
1

Hi everyone,

I wanted to understand how other users manage start/stop rules for algo strategies, especially for option buying or naked buying strategies.

I have subscribed to multiple algos through DhanHQ/Stratzy. Among them, a couple of strategies are option buying based, where drawdowns can become uncomfortable if there is no clear rule for when to pause or stop trading.

My main question is:

How should a user decide when to stop trading a strategy?

For example, do you follow any of the below rules?

  1. Stop after a fixed percentage drawdown from deployed capital

  2. Stop after the algo hits its historical maximum drawdown

  3. Stop after drawdown crosses a multiple of historical DD, like 1.5x or 2x

  4. Stop if the strategy does not recover within a certain MTTR period

  5. Stop after a fixed number of consecutive losing trades

  6. Continue only if the current drawdown is within the backtested risk range

  7. Pause and restart only after the equity curve comes back above a moving average

I am not looking to blame any strategy or platform. I just want to understand the correct risk-management framework that users should follow before continuing or pausing an algo.

It would be helpful if experienced users or the Stratzy team can share:

  • Whether there are predefined stop-trading rules for each algo

  • How much drawdown is considered normal vs dangerous

  • Whether users should compare live drawdown with backtested max drawdown

  • What is a practical restart rule after pausing an algo

  • How to manage naked buying algos during bad market regimes

Would appreciate guidance from the community.

2

@MGL_Data2 I keep it simple and focus on the algo’s max dd.

My approach:

  • I run at max capital allocation, but cut it by 50% if the historical max DD is breached.

  • If it drops another 10% on that reduced capital, I pause it completely.

  • I monitor the Sharpe ratio and resume only when it improves.

  • I avoid algos with >35% max DD and always run at least 3 algos combined.

Curious to know how others manage this.

2 Likes
3

I also follow something similar if dd crosses 1.5x of max historical dd reduced to half if crosses 2x pause and watch till ratios improves , I keep algo corealtion below .3 to each others