Problem: Too many people still think that playing Rules As Written is somehow the only way to play... but then get all bent out of shape when the books that are printed do not give them the rules they want.
Problem: Dungeon Master has trouble remembering to give Inspiration; but, when he does remember, the Players already have Inspiration because they haven't used it from last time. The Players don't use their Inspiration, because they get it so rarely and they want to save it for a pivotal roll later.
Solution: Action Points. Each Player gets a number of points at the start of each gaming session called Action Points that can be used to change the outcome of a die roll. Action Points can be used to 'add pips to the die' so to speak, and Players should use them during the current game session because they are non-transferable from session to session. At the beginning of each session each Player's Action Points refresh anew.
The advantage over Inspiration is that Action Points can be awarded to Players even if they already have points. This means Players can hoard points throughout a session, but because Action Points are non-transferable between sessions, they need to use them or they are wasted.
I start each Player with 3 points per hour, so a 2-hour session starts with 6 per player and a 4-hour session with 12. However, a Dungeon Master could start with more or less points, or none at all.
I have found Players use Action Points more often. This is perhaps because it is not an all-or-nothing deal. You can use 2 or 3 points and still have some for later. Furthermore, since Players use them more often I am reminded of them and tend award them more often.
Finally, Action Points need not replace Inspiration; the two can be used in tandem.
ProblemHit Points in 5e are completely and totally dependent on your class and Constitution score . . . which makes absolutely no sense given the explanation of hit points in the PHB as being a "combination of physical and mental durability, the will to live, and luck".
Solution Part 1 (fixing the class dependence aspect of this issue)Grant players an extra hit dice based off of their race/lineage. Small races (halflings, gnomes, goblins, kobolds) would get an extra 1d6, most "medium" races would get an extra d8 (Humans, Elves, Half-Elves, Tieflings, Genasi, Satyrs, etc), the tougher Medium races would get an extra 1d10 (Dwarves, Warforged, Tortles, Half-Orcs, Firbolg, Hobgoblins, etc), and the really big ones would get an extra 1d12 (Orcs, Goliaths, Centaurs, Bugbears, etc).
Solution Part 2 (fixing the Constitution dependence aspect of this issue) Allow Hit Points to be based off of two different ability scores dependent on their class, alternating the ability score every level. This would keep people from being as SAD as can be, while having the definition of HP fit how they are calculated better.
Hybrid solution: use the constitution score (1-20) to determine base hp, and then add an amount based on class (possibly adding the class's key ability score) per level.
Pros: still includes meat points (meat points are hit points, although hit points are not meat points), but makes the difference more about luck/skill/divine favor, which is what high-level hit points are all about. Give level 1 characters a lot more hp, making them tougher (or at least less likely to be dropped by an unlucky crit). Con is still useful, but you no longer feel the need for at least a 12 or whatever, since the relevance of con score will drop over time. Makes all classes more SAD (allowing a bigger variety within classes)
Cons: 1st-level pc are a lot tougher, making old-school gritty play (where you need to be careful or you'll get dropped by an unlucky crit) much less doable. Makes all classes more SAD (causing every pc to be better at both their main job and secondary stuff because you can load ability scores more).
The attunement mechanic means that PCs end up carrying around magic items they can’t use. With the difficulty of selling items (and the difficulty of buying them) they effectively become bumpf as characters progress. This cheapens the overall value of magic items rather making them precious.
What use can the PCs have for old magic items they got at levels 3 or 4 when they are now obsolete?
For the record there is zero to nil chance of the players giving items away. They would drag it around for years unless they thought they could get value out of it.
I generally offer upgrades to major items (weapons, staves, etc) rather newer/better versions. So you don't get rid of the +1 longsword and start using a flame tongue - you upgrade the +1 longsword to do extra fire damage.
This requires planning items ahead - ie random items can't be major items. I still do random potions and luxury items.
[O]ne thing in 5E that I really find to be a problem as a GM is how poorly the action economy is balanced for "solo" creatures. A PC party of 4 or 5 characters punches WAY above its weight class against solo monsters, even in Lairs and with legendary actions. A good part of this has to do with the 5E math -- solo monsters don't hit especially hard and so they aren't terrifying in that "stay away from it or you're dead!" feeling that helps keep the PCs at bay. On top of it, PCs can really pump out a lot of damage when they want to and solos, which are usually just big bags of hit points, don't last long. All that said, the fight against one massive foe is a fantasy staple and I want it to work -- and not just for epic boss battles. There's no reason a random encounter with a giant or whatever shouldn't be viable, too.
One thought I have had to fix this is to treat a big creature like a group of creatures that all stay close together. Like, if the dragon were it's head, it's tail and its torso/claw routine. So the head not only gets to act independently on its own initiative, it has its own list of abilities, its own reach and range, and its own hit point pool. The same for the other parts. But while I think it is a neat idea for a dragon, I don't know how it would translate well to a giant or other creature without lots of "interesting parts."
A intentional design point in 5e with bounded accuracy was that the primary knob for how long a creature lasts is HPs, no longer with defenses as part of the mix. However, what that means is that (a) solo creatures don't last long enough because they are also designed to be non-solo creatures vs. higher level PCs. Especially with classes that can nova or still do much of their damaeg with fewer hits (SA, paladin smite, etc.)
(Note: I don't think this is a bad idea. Bounded accuracy - which might be more accurately described as bounded targets - not only keeps things viable across longer level ranges, but also lowers what a +1 means - if you need a 11 or higher, a +1 will succeed 10% more often (50% to 55% -> 5% of 50% is 1/10th). If you need an 18, that +1 is means succeeding 50% more often. So maximizing every +1 is mroe important when you can be going against foes that need extremes to hit.
So in addition to the action economy they try to solve with legendary and lair options, they just don't have the staying power.
This is a difficult problem to resolve if the party is supposed to be able to meet these creatures later in multiples without mechanical changes.