Even in a sandbox IMO the DMs job is to tempt, challenge and yes even manipulate the decision making process- you want player agency and you throw out fishing lines and wait for a bite but have lots of options at your disposal (eg using NPCs) to encourage/guide the decision making process when the party decide which rumour (hook) they want to follow up.
IMO you need a hook to start the adventure. Simply saying you are in town what do you want to do? Is not always enough, give each player a clue or two at the start of the game as to possible dangers/adventures in the area. Even setting them up with a little quest to kick the campaign off. In my own Greyhawk PoTA campaign, I started the players off as members of the local militia to avoid 1-2 sessions of wandering around aimlessly “looking” for an adventure. They were tasked with patrolling a road where bandits had attacked caravans recently and this ended up with them being strafed by some vulture flying cultists who they decided to follow back to their lair and it was game on from there.
The smart party will soon realise that NPCs can be a useful way to minimise/mitigate danger by informing them of possible dangers in different areas- hill giants are said to live in the Lortmil mountains over there. An orc warlord is rumoured gathering troops in the Dreadwood for who knows what purpose. A dragon was spotted circling over the Good Hills a few weeks back. If players want to follow up on one of those rumours then sweet we’ll roll with that.
IMO the problem with sandboxes can be decision paralysis- give players too many options and they will freeze, unsure of which way they “should” go or in my groups case decide half way through that “the adventure is too tough let’s bug out” and they wander off to do something else instead (leaving me with a half finished adventure and lots of consequences to sort out- which can be a good thing).
These days I tend to limit to a couple of possible hooks/rumours etc in town when they are looking for their next job but if the party really wants to go and slay that dragon that was spotted in the hills then that’s cool, we’ll roll with that- in fact a “random” encounter with a dragon lead the party to the Moathouse and a show down with it (I’m using the original Hommlet as a base and used the 3E version of the ruined Moathouse of RttToEE).
I’ve enjoyed the aspects of PoTA I’ve run but the campaign has a possible flaw in that it is a sandbox yet still expects in some ways level appropriate dungeons- in my experience if players bust into one of the upper temples then they will want to explore all of it including the links to the lower levels- that is where the adventure can run into problems and may need DM (divine) intervention to rescue. Sly Flourish as some good ideas for minimising the chances of pcs wandering into the wrong temples by having each of the lower temples need a key to access which can leave you with a good lower level sandbox with the upper temples before the party strikes out for the lower ones:
http://slyflourish.com/tying_the_threads_of_princes.html
Me though, I’ve only used a few temples as a backdrop vs larger campaign events and not run the whole thing because of player choices/ wanting to follow up on other leads/ adventures. The party had just finished the upper air temple and decided that they wanted/needed more magic weapons (as stuff had gone pear shaped and the cultists had allied with some cloud giants- SKT) and since I don’t have any magic shops they dug round and researched and learned that some weapons were said to be hidden somewhere within White Plume Mountain far to the north… so off they have gone way up north on a road trip to find some weapons (my own weapons not the original WPM ones). We may get back to Princes eventually but who knows when?
Stormdale
