If this ever happens, my psychic powers have failed and its no longer appropriate to remain exposed. I just dump it with an urgent sell order, at market. Yelling into the phone, if necessary!
But do you refer to it as a stop loss? Phone??? I suppose you got your quotes from the newspaper. What century are you in?
In the quant space you see both people using well defined TP/SL combinations and there are people who prefer to manage risk using sizing and hedges. Both approaches can make money and both can lose money if done incorrectly. Most of the quant gripe with SLs or TPs is that it naturally makes a strategy path dependent and thus much harder to manage. It’s such a tricky problem that people created the whole synthetic data approach to deal with it.
I manage positions by establishing clear risk limits at the portfolio level, e.g. no single position greater than 8% on a beta adjusted basis, needs a certain ADV threshold, total portfolio net exposure limits, etc. If I'm betting on an outcome and am wrong, then I realize a loss - yes. Most stocks aren't liquid enough for more than $15-20k sweeping screens at any given moment, so if my position is worth ~$5mn, bad execution will be very expensive (can cost another 200-500bp). Furthermore, price is not a thesis -- but it does create the "opportunity". If a stock is trading at $15 and I think it's worth $20 then I'll continue buying as long as I'm within my risk limits (which means sized appropriately given its beta). If for some reason the stock drops to $8 then I would need to understand why, and then make a decision about buying/selling. If the thesis broke then I exit, but if not, I may add. Sizing and hedging are much better ways to manage pnl IMO than a simple stop loss. When a trader is right, they should make money, and when they're wrong, they should expect to lose money. That's the consequence of "skill". If stocks were random then your pnl would not be correlated to skill. Note: I'm a discretionary manager... listen to @Slow Learning Elf for the systematic side (plus he's infinitely wiser than me).
Mainly the underlying, at about 80/20 split. But stop loss on options makes even less sense IMO. When I look at my option book, I calculate the beta adjusted delta values and see that as part of my total portfolio risk (with position limits etc.). When I look at single position risk I combine both the options and underlying.