One of my first impressions of 4E could be summed up by the phrase “change for change’s sake.” There were several places where the new rules felt more like a designer stamping their name on the game rather than following a natural path of evolution. Hit points was one of those places.

I agree with the reasons behind the decision to change how hit points were determined. Many starting characters were too fragile, and it really sucked to roll a 1 for hit points when you leveled up. I don’t agree with the solution the 4E designers chose because it shifted a distinguishing characteristic of each class to favor roles, it lessened the impact of Constitution upon level progression, and it introduced an unnecessary resource mechanic.

Starting Hit Points
The obvious resolution to first level fragility is to give characters more hit points to begin with. I love the use of the Constitution score in the formula for starting hit points. I don’t like that hit points are determined by role rather than class in 4E. Two of the roles share the same base. That’s a rather narrow field of flavors, compared to previous editions, where classes ranged from d4s to d12s.

I’ll admit that flavor is the primary reason why I don’t like 4E’s abandonment of the class-based hit die, but hit point bloat is also a factor. Hit point bloat is a complaint I’ve heard more than once, and usually in reference to the resulting grind of encounters. The common solution is halving monster hit points, but part of the reason why monster hit points ballooned in the first place was to match the increase in character hit points.

An alternate formula for starting hit points that bridges third and fourth edition nicely is to take the maximum value for the hit die from the 3.5 version of the classes and add the Constitution score. Not only does this provide the significant boost in starting hit points that the 4E designers (rightly) sought, but it maintains one of the distinctive traits of the classes.

That said, I’d probably tweak the values for each class based on their role:

Class Starting HP
Avenger 8 + CON
Barbarian 12 + CON
Bard 6 + CON
Cleric 8 + CON
Druid 6 + CON
Fighter 10 + CON
Invoker 4 + CON
Monk 8 + CON
Paladin 10 + CON
Ranger 8 + CON
Rogue 6 + CON
Shaman 6 + CON
Sorcerer 6 + CON
Warden 10 + CON
Warlock 6 + CON
Warlord 10 + CON
Wizard 4 + CON

Hit Points Gained per Level
I was torn at first about the loss of the random die roll for hit point gain per level, but I’m over that now. Setting it at half–plus or minus one–of the base starting hit point value was a good move, but they forgot something. Why not include the modifier from Constitution? It stinks of being different just to be different. It also lessens the importance of the ability.

Healing Surges
I think healing surges are largely unnecessary. I don’t understand why healing powers don’t just stipulate a random value or a flat rate (e.g., 1/4 of target’s maximum hit points). Most healing powers already specify the number of times they can be used, making the expenditure of healing surges a redundancy. And giving everyone the ability to use healing surges after combat diminishes the leader role.

The concept behind second wind is a good one, however. It’s one way to capture the hero getting back up after a beating and taking down the villain. I would turn it into a feat that is triggered by the use of an action point instead of making it an encounter power that uses a healing surge, though.

SECOND WIND
Benefit: When you spend an action point, you regain hit points equal to 1/4 of your maximum hit points. You cannot gain more hit points than your maximum from this effect.

Coupling second wind with another standard action to take down the villain–now, that just screams heroic pulp action to me.