Reply to thread

Message: <blockquote data-quote="pemerton" data-source="post: 8599925" data-attributes="member: 42582">On the issue of PCs running away.Gygax, in his PHB and to a lesser extent his DMG, presents a fairly clear model of play:* Before play starts, the GM builds a dungeon, which has opportunities to obtain treasure, most of which will require overcoming a guardian monster; * The dungeon is a series of rooms linked in (sometimes complicated) ways by corridors, and the room contents - both creatures and treasures - are largely static; * The initial task for the players is to collect rumours, to map the dungeon, to draw on other available information, and to work out where the treasures can be found; * Equipped with information, the players then enter the dungeon with particular goals in mind, and they do their best to achieve those goals despite harassment from the GM in the form of wandering monsters, tricks and traps; * Thus, while it is the GM who establishes the context for play, it is the players who are taking the lead in establishing pacing and which encounters take place.I have no idea how widespread this model of play ever was. As [USER=7026617]@Thomas Shey[/USER] has posted, there were significant departures from it right from the get-go. And to me it seems like a hard model to implement for level 1 PCs, who don't have the robustness to reliably undertake the information-gathering phase. But clearly some people were doing it, including some of Gygax's players - eg when Robilar entered the GH dungeon to free the demigods, to take the weapons that freed Fraz-Urb'luu, etc.In this style of play, the players have a good reason to retreat from powerful wandering monsters, as tackling them is not part of the plan. They also have a good reason to retreat from an overly-powerful room guardian, with the goal of regrouping and returning better-equipped.There are rules for adjudicating retreats in a dungeon, which establish relatively high chances of creatures letting the PCs go even if, in mechanical terms, they would have a good chance of killing them - especially if the players have their PCs drop food or treasure to distract pursuers.When I read accounts of how 5e is played, it seems to depart from this model at just about every point:* The dungeon is not a series of optional but largely static encounters linked by complicated pathways of rooms - rather, it is a set of events that the GM intends to have occur, often in a pre-configured sequence; * Following from the above, it is the GM who is primarily in charge of pacing and of choosing which encounters take place; * The situation is not static, so the players can't have their PCs retreat and regroup under the assumption that they can then return to the site of their retreat with a better plan, or better equipment, and have another go (the departure from the "video game-y" assumptions of Gygax's PHB begins a year later in his DMG, where he gives advice on "reactive, living" dungeons which, if followed by a GM, will make it very hard for players to apply the advice he gives to them in his PHB).Within the context of what seems to me to be the predominant 5e model of play, retreating seems like a last resort - it's effectively a failure to successfully engage with the content the GM has put in front of the players. (This is what I take [USER=82524]@Vaalingrade[/USER] to be saying.)And that's before we get to the issue of whether the mechanics do or don't support it - I agree with [USER=6877472]@James Gasik[/USER] that the mechanical change from "side" initiative to stop-motion turn-by-turn initiative is a big deal in this respect.The best rules for retreating that I know of, in the context of a modern D&D-style game, are in 13th Age: the players can declare a retreat at any time, and if they do so then they get away; but in exchange the GM is entitled to narrate a "campaign loss", that is, to introduce some significant new fiction that is a significant set-back to the PCs (and thus the players) in terms of their overall goals or circumstances in the ongoing game.</blockquote>

Verification