To hold a trade when the trade is open and it immediately starts going against you once it's open, even as it nears your stop loss without exiting, requires a lot of conviction in the thesis of the trade - depends on various factors. When it works, it works wonders.
But this happens very rarely. And to hold the trade requires objective, tangible inputs to your data and not just gut feeling. The data can be anything you have tested - volume, S&D, imbalance between buyers/sellers, the balance shifting, anything for that matter.
My theory is as follows. Usually, if you are right, observe when the trade starts working. If you got the entry right and trade right, it will start working immediately within first 2-3 candles. If it doesn't, and it's hovering around your trade open price, you can hold.
But if you open a trade and price starts going against favor, exit within 2-3 trades, re-evaluate and re-enter later. There maybe whipsaws, but few points lost here and there by exiting trades that don't seem to work within first few candles, will save you lots of points in loss.
These things are of course nice to say/observe. But you have to implement a hardcoded and thoroughly backtested objective way to this approach. Any strategy you pick up, see how it performs, and when it performs and when it's successful.
You can't see a lot of things in backtest - like open interest change, market dynamics, buyers-sellers balance/imbalance, etc. So, this is a skill that you partially learn over time of observing charts as they unfold.
Do this at least early on in your trading career. Even if you sacrifice % points in returns, you will have a relatively less stressful time building your equity. Especially, done the right way, this leads to a smoother equity curve.
Approach the market objectively as much as possible. But initially when you're trading with a very small capital, you have to learn to cut your losses very very quickly and not let losses balloon. And you have to find a systematic approach to let profits run to exhaustion.
