I'm not sure I would. This is one of those occasions when the distinction between a vector and a range-variable shows up most sharply. The range variable is an iterator and will (as you alreadly know 🙂 ) return the value of max for each value of y(x) ... which will be y(current value of x in the iteration). The problem with changing the behaviour is that Mathcad would have to second-guess whether y would only return a single value or whether it would return multiple values for each x (it's single in this case, but might not be in general) and then act in accordance with whether it thought that's what the user wanted (could be right-click selectable, I suppose). I definitely do not want to see an error message, because there is no error.-1,-0;
If that's what you want, then I think the safest thing to do is use a function equivalent to my vec to convert x (or possibly y(x)) to a vector before applying max.
