Party optimisation philosophy

[MaxWilson]

5E has enough exploits normally held back by other party members that I think tier-zero optimal parties are going to be 100% gimmick builds.

1 Araakockra Spell Sniper Sorlock's extreme kiting ability doesn't matter much, because every enemy who can't target him can target the PC that didn't invest in the gimmick.

A 100% Araakockra Spell Sniper Sorlock Party is basically guaranteed no-risk victory against the vast majority of encounters, and because of that can safely nova through all other encounters without worrying about their depleted resources becoming relevant later.

Similarly, stealth is often constrained by the least stealthy party member. But a party that goes all-in on stealth, such as 3 AT's and one Trickery Cleric 3 / AT 17 doesn't even really need to let the vast majority of encounters happen. And those few monsters that pose anything like a serious challenge (e.g. Aboleths with their Detect legendary action) can, again, be safely nova'd to smithereens without worrying about needing those resources later.

The goal of the original thread gets into the mathematical complication of Shapley Values. In "fair" parties playing "normal" DnD I'd argue most of the differences in value comes from two places:

• Reliance on other PC's to not die (-)
• Ability to assist the party in recovering from or averting TPKs (+)

This is because generally most encounters and adventures are highly winnable, with very little upside for "winning more", and the biggest risk IME is cascading character failure.

Which is why I kind-of agree with a lot of the controversial choices in the original thread.

A single PHB paladin on his lonesome will do alright. If he's added to the party later, lay on hands is enough to bring someone back from zero (recovering) and occasionally his nova will bring a threat down (averting).

A single non-bladesinger wizard isn't going to make it very far on their own. This means that a lot of their apparent contribution to the group is substantially attributable to the other party members that keep them alive. While they do have some ability to avert disaster (e.g. force cage the baddy away), there's much less value add if they're redundant (e.g. because there's already a Lore Bard with Force Cage) and they have basically no ability to directly recover the party from disaster via healing. Wizards are extremely powerful, but that doesn't actually mean that that much of party success is due to them.

The main areas where I then have to disagree with the original thread (outside of the issue of having too much in-class variance) is the low ranking for clerics.

[Googles 'Shapley values', stares at formal definition for a while, gives up for now due to lack of time]

Very interesting thoughts, thanks for sharing. That's an interesting way of thinking about things, and I agree that if you have no ability to control the narrative and pursue greater challenges (it's not a sandbox), optimizing to minimize the chances of catastrophe is a perfectly reasonable approach.

A related way to think about party optimization could be "how robust is this party against player error?" Are there certain PC classes which rely less on player skill, perhaps to the point where an incompetent or malicious player still cannot lose?

The Holy Grail of one approach to party optimization (Punching Above Your Weight Class) might be "a party build which is so strong that, when played by skilled players, it can fight its way through an army of 5000 orcs and mind flayers and then kill Tiamat at 1st level at least 50% of the time, without even taking a short rest."

The Holy Grail of another approach (Avoiding Catastrophe) might be "a party build which is so reliable that you can run 10,000 Adventurer's League tables with this build through the same WotC adventure without a single TPK."

The Holy Grail of a third approach (Minimizing Incompetence) might be "a party build which is so reliable that one semi-competent newbie and three covertly-malicious saboteurs who are actively trying to TPK by displaying abject cowardice (Dashing away instead of fighting) and mistakes like Fireballs which 'accidentally' harm more PCs than bad guys will still successfully complete a typical WotC adventure at least half 90% of the time."

I conjecture that these three types of party optimization lead to very, very different party builds.

Furthermore, builds are only optimal w/rt a given set of rules and a style of DMing. E.g. Eldariel mentions Healing Word as an important mitigation strategy, which reminds me that Healing Word has very low importance at my table because HP can go below zero, and you can't just take a Fire Giant critical to the chest and pop back up on your feet with a single Healing Word (unless the Fire Giant crit left you at just-barely-negative HP). Another example: some DMs will actively attempt countermeasures against PC strategies that get used repeatedly, e.g. the Aarakocra party is now facing only monsters with spells and ranged attacks, 80'+ flying speed, or both. At these tables, the "best" strategies aren't those that work the best, they're the ones which are best at deceiving the DM into thinking you're on the verge of TPK when you actually still have lots of hidden defensive depth. (Having lots of Healing Word might be excellent under this type of DM, although I wouldn't really know because I avoid these types of DMs--but if you've got 2 PCs hitting zero HP in every encounter and everybody hovering around 10% health, the DM may not think hard about the fact that the party collectively still has 20 spell slots that could go towards Healing Word, full HD for short-resting, a Rope Trick to enable short rests, a Brazier of Summoning Fire Elementals, and a Horn of Valhalla for emergencies. The trick is to look weak without being weak.)

The Adventurer's League Punching Above Your Weight Class build might not even work at my table, or vice-versa.