You have a view. The optimizer turns it into a ranked list of structures that express it, scored on probability of profit, return on risk, and what each one is actually worth if you turn out to be right.
Most tools make you pick a strategy first and find out afterwards whether it suited your thesis. The optimizer inverts that. Give it an underlying, a price target, a target date and the maximum you are willing to risk, and it searches structures against those constraints instead of asking you to guess which one fits.
The target is a view, not a prediction you have to defend. Raise it, lower it, move the date, and watch the ranking reorder. That reordering is itself informative: a view that only pays through one exotic structure is usually a view the market has already priced.
Each candidate is scored on probability of profit, return on risk, value at target and maximum loss, side by side. Ranking on premium collected alone is how traders end up selling precisely the structures that pay the most because they are the most likely to hurt.
Because maximum risk is an input rather than an afterthought, every structure in the list is one you have already decided you could survive. The comparison is then about which expression of the view is efficient, not about which is affordable.
Figure: Probability of profit comes from the distribution the market is currently pricing, which widens with time. It is the best available estimate, and still an estimate.
A ranking is a starting point, not an answer. Any candidate opens directly in the Risk Profile modeler, where you can drag its strikes, slice it by date, stress it against a volatility shift, and see it merged with the book you already hold before committing anything.
Nothing here places an order. The optimizer proposes, the modeler tests, and the decision stays yours.
No, and it is not investment advice. It ranks structures against constraints you supply, which is arithmetic rather than a recommendation. The view is yours, the risk limit is yours, and so is the decision.
From the same pricing engine that prices the rest of the app, using live quotes and implied volatilities. It describes the distribution the market is currently pricing, which is the best available estimate and still only an estimate.
Yes. Open a candidate in the modeler and load your held positions alongside it: the payoff curve then shows the merged book rather than the structure in isolation, which is what you would actually be holding.