Death saving throws: a quick house rule for 5e and all editions

This is what D&D Basic (5e) has to say on the subject:

Falling Unconscious
If damage reduces you to 0 hit points and fails to kill you, you fall unconscious (see appendix A). This unconsciousness ends if you regain any hit points.

Death Saving Throws
Whenever you start your turn with 0 hit points, you must make a special saving throw, called a death saving throw, to determine whether you creep closer to death or hang onto life. Unlike other saving throws, this one isn’t tied to any ability score. [...]
Roll a d20. If the roll is 10 or higher, you succeed. Otherwise, you fail. A success or failure has no effect by itself. On your third success, you become stable (see below). On your third failure, you die. The successes and failures don’t need to be consecutive; keep track of both until you collect three of a kind. The number of both is reset to zero when you regain any hit points or become stable.
Rolling 1 or 20.When you make a death saving throw and roll a 1 on the d20, it counts as two failures. If you roll a 20 on the d20, you regain 1 hit point.
Damage at 0 Hit Points.If you take any damage while you have 0 hit points, you suffer a death saving throw failure. If the damage is from a critical hit, you suffer two failures instead. If the damage equals or exceeds your hit point maximum, you suffer instant death.

Well, it is not a bad rule, but not one I am very fond of either. I don't like is the idea that "a success or failure has no effect by itself". which apparently means they don't represent anything other than extra chances to get well. Also, I avoid the whole concept of "roll many times for the same purpose".

Instead, try this idea that I have adapted from my own RPG, Days of the Damned (although I'm sure someone, somewhere, must have thought of this before). If you don't play 5e, "disadvantage" means basically rolling 2d20 and picking the lowest (you can use a -4 penalty instead, or whatever works in your game).

If you are reduced to 0 hit points, make a death saving throw. Failure means you must tick one of three boxes: unconscious, disabled (which means incapacitated, disadvantage to everything, something from 3e, or whatever you like) or dying. When you tick all three, you're dead. Whatever rules you used for immediate death still apply (-10 HP, -CON HP, massive damage, etc).

You make no further death saving throws, unless: a) you are dying, b) you're incapacitated and decide to take any action, or c) you take more damage.

Success in this saving throw has no immediate effect (there are no "success boxes"), failure makes you tick another box. There is no stabilization or gaining 1 HP without healing, although you can heal yourself if conscious, or wake up unconscious (either one taking 1d4 rounds with help). You still take a while to be ok: for 1d4 hours, you have disadvantage in all rolls, including death saving throws.

Alternatively, just leave the ticks until you get decent healing, but they are only in effect when you're at 0 HP.

But what box should I tick first? Choose whatever fits your tastes:
* Choose randomly (1-2: unconscious, 3-4: disabled, 5-6: dying).
* The player chooses.
* Depends on where you were hurt (head: unconscious, limbs: disabled, torso: dying).
* Depends on how you where hurt (concussion/bludgeoning: unconscious, cutting/burning: disabled, poison/piercing: dying).
* Combine two or more (choose randomly, but roll 2d6 and pick lowest if the head is hit, etc).

What's the point?
The idea is providing more variety, giving more options to the characters (that can consciously choose to fight to the death), and, specially, allowing for dramatic "conscious but dying" scenes.

Andre Norton and the 9th Level Fighter

When a Fighter reaches level 9th in some of the "classic" editions of D&D, he may become a leader of men. Here is how Daniel Proctor's Labyrinth Lord puts it:

Reaching 9th Level: At level 9 a fighter may become a great leader of men, taking control of a parcel of land and a leadership rank in his society. A fighter will, assuming money is at hand, build a castle. He may ultimately control several villages and towns, but must be a good, strong leader and provide protection.

That isn't an aspect of D&D I haven't explored much in my games. I think most of my players prefer adventuring over managing parcels of land unless we're playing a "noble families" campaign, and building castles never sounded as much fun as taking them.

Andre Norton's short story The Toads of Grimmerdale, one of the most "D&Dish" things I have read recently (even compared to some other books in appendix N), has this interesting bit:

“It is in my mind that Nordendale needs a lord, one to give the people heart, rebuild what man and time have wasted. I have come north seeking a chance to be not just my own man, but to have a holding. I am not like Urre who was born to a hall, and drinks and wenches now to forget what ill tricks fortune plays.

“Who my father was—” he shrugged, “I never heard my mother say. That he was of no common blood, that I knew, though in later years she drudged in a merchant’s house before the coming of the invaders for bread to our mouths and clothing for our backs. When I was yet a boy I knew that the only way I might rise was through this—” he touched the hilt of his sword. “The merchant guild welcome no nameless man, but for a sword and a bow there is always a ready market. So I set about learning the skills of war as thoroughly as any man might. Then came the invasion and I went from Lord to Lord, becoming at last Marshal of Archers. Yet always before me hung the thought that in such a time of upheaval, with the old

families being killed out, this was my chance.

“Now there are masterless men in plenty, too restless after years of killing to settle back behind any plough. Some will turn outlaw readily, but with a half dozen of such at my back, I can take a dale which lies vacant of rule, such as this Nordendale. The people there need a leader, I am depriving none of lawful inheritance, but will keep the peace and defend it against outlaws—for there will be many such now. There are men here, passing through Grimmerdale, willing to be hired for such a purpose. Enough so I can pick and choose at will.”

By the way, if you haven't read Andre Norton yet, you probably should. Her stories have everything: old ruins, ancient alien gods, clerics of lawful religions, dark pacts (with nefarious consequences), saving throws, war, tragedy, adventure, multidimensional characters... and many things that might make you eager to play a 9th level fighter, or any number of different characters.

Of Rules and Halberds

There are many reasons to like old school D&D, including the oft-cited “simplicity”. Even though it isn't one of my first criteria to find out if a game caters to my tastes, I do appreciate some of the perceived simplicity of older editions.

For example, I have a hard time wrapping my head around big lists of skills, specially in D&D. I like 3rd edition, but every time I read “Use Rope” on a character sheet I remember why I probably will never play it RAW again. GURPS and Burning Wheel, two games I appreciate, have the exact same problem for me.

Feats are even worse - like the concept, hate the sheer number of options with all the complexity they bring. I get bored before I can read a dozen of them. Repetitive ones with little difference from each others are the worse. Even spells, which are meant to be awesome, can become dull lists of numbers, specially in in some editions (Wiz 3, 400 ft. + 40 ft./level, 20-ft.-radius, 1d6 per level, just to mention a famous one).

But the original game is not that simple, either. For example, I am not a fan of calculating XP, specially if I must add 10% for prime ability and then multiply it by 5/8 because my character is an 8th level wizard fighting a 5th level monster. I understand the idea behind traditional XP rules, but I think it could be made simpler without losing much utility.

And earlier editions could be astonishing complex, sometimes even more than later ones. The first supplement to D&D, Greyhawk, differentiates weapons by efficiency against armor, damage, damage against large opponents, space required, and so on. Because of Gary Gygax's well-known appreciation for polearms, first edition would have about a dozen of slightly different ones.

First edition weapons table.

So maybe early D&D isn't the place to go looking for simplicity. But that's ok; my own tastes are not that minimalist as well.

For example, I do like monsters. Lots of them. Reading about new, well-written monsters is always a pleasure to me. And unique characters, bizarre laws, strange customs, forgotten deities, mysterious religions and crazy factions.

I also happen to love medieval weapons. And armor. And how weapons interact with armor. And how fast, strong, and versatile each weapon is (even though 1e gets too complex for my tastes). Specially swords - I'm not a fan of having too many polearms, to be honest.

That's why I think that “rules light” and “rules heavy” aren't the best terms to describe my favorite playstyle. The games that I like the most are often both too simple and too complex at the same time, so I have to change them until I make them my own.

Detail should be where the heart is; the rest is often uninteresting.

That's also why I think I'll never be able to stop reading (and writing) new games, new house rules, new blogs and so on. I'm just too interested in new ideas for roleplaying games. And if you love polearms, there is no such thing as too many of them.

Brainstorm in a Jar: Dynamic d10 Initiative

Here's is my new post, it's just like the old post… but using d10s instead of playing cards. I've decide to reproduce most of the material to to make reading it and referencing it easier. To save you the trouble of reading it all again, I've put the most important of the “new stuff” in italics. So, if you want to see the whole picture, read it all. If you've read the previous article, you might want to read the italics only.

The idea behind this system is that every combatant is trying to find a good moment to attack (or act), and faster characters get better opportunities. The goal is making things unpredictable, dynamic, and tactical.

In the begging of each round, each player rolls a d10. Rounds last for about 10-15 and are divided in 10 segments, each one lasting about a second. Characters who roll a 1 act on segment 1, and so on. Every action has a duration in segments, usually 4.

Immediately after taking your action, roll another d10. If the new number is equal or higher than your number card plus the duration of your action, you can take an extra action in this round (when the segment corresponding the new number comes up). If it isn't, you can keep the new number for next round, if you wish. For example, if you played a 4, and took an action with a duration of 5 segments, you can act again if you draw a 9 or 10. 

If you roll the exact same number of the segment you're in (another 4, in our example), you can choose the result you want. Manually set the new dice to the chosen segment - segment 9, for example.

No more than three actions can be taken each round, which can be managed by looking on the number of dice in front of each player (so, everyone should set apart three d10s for this sole purpose) .

There is no problem in finishing an action on segments greater than 10 (this "extra segments" are somewhat abstract), and it doesn't affect next round, but your last action cannot finish in a segment greater than your first dice plus 12. Thus, if the first number you rolled is a 3, you cannot start an action in segment 10 that takes more than 5 segments.

When a new round begins, everybody rolls a new dice (the ones who kept a dice from the last round may chose which one to keep), and so on.

Although this looks like a "second-by-second" breakdown of the encounter, it involves quite a bit of abstraction to make things run smoothly. Because of that, you resolve each action on the segment it begins, although it can be interrupted in the middle.

And… that is it, basically. You can use or ignore the optional rules below, depending on the system you're using or the effects you're looking for.

ACTION SPEED

Movement: 2 segments for a 5-feet step, 4 segments for a half move, 8 for a full move, 12 for a double move.

Weapons: 4 segments for a "medium speed" weapon, 2-6 for other weapons (2 if unarmed, or if you have a longer weapon than your opponent and hasn't been hit by him yet). Bows twice the time to draw and shoot, crossbows take half the time to shoot and twice to draw.

Spells: equals to spell level+2.

The GM will decide on the duration of other actions.

You can add a simple, quick action, such as taking a single move or drawing a blade, to other, more complex, actions. Just add the durations. The GM has the final word, but the idea is that if no dice are rolled because of your action you don't need to use an additional card. For example, you can move and attack, although you have a good chance of being interrupted in the process (see below).

DELAYING AND CHANGING ACTIONS

To delay an action, when your segment comes up say you'll wait for something to happen - an attack on yourself or an ally, a move by an specific opponent, etc. If that happens, you can act on any given segment after that. If you rolled a 2, and act on segment 7, for example, you action may end on segment 11. If you don't want to declare your action right now, you can roll and extra die, as if you had just taken a 1-segment action.

Likewise, actions can be changed at any time before they are finished, but the change takes one segment (so, if you change your mind on segment 6, you can act on segment 7, without rolling an extra die).

INTERRUPTING

You can interrupt an enemy if you can finish an action before him. You must declare this the moment the opponent declares his action. For example, if he attacks you on segment 4 with a two-handed axe (speed 6), he will hit you on segment 10. If you act on segment 5, you can hit him with a knife before that. Even if he survives the attack, enough damage (let's say, 10% or more) can cause an ill effect. Some suggestions: spell interruption, -2 penalty to a roll, movement is halted, etc.

You can use you interruption to defend yourself or an ally. This depends on the system you're using. One example: a parry grants +2 AC.

Interrupting is about disrupting your enemy's action. You cannot simply walk away in the middle of an attack.

Optional rule: if you want to interrupt an opponent but your dice doesn't allow it (and its segment is yet to come), you can trade your number, rolling again, but the new die MUST be used to interrupt. If the new die doesn't allow it, you cannot use it.

TIES

If it matters at all, give them to whoever started the action first (in the event of interruptions), or flip a coin.

SURPRISE

Surprised characters roll no dice in the first round.

DEX MODS

DEX mods apply only to the first die you play. If you have DEX +2 and your first roll is a 5, manually adjust it to a 3. If you roll a 1, the bonus is useless unless you use the option below.

Another option is to let players roll one extra die for each +1 in the beginning of the round, choosing the best and discarding the rest.

MULTIPLE ATTACKS

If a character has multiple attacks, he may take all his attacks in his turn. Depending on the edition you are using, you can use special cards or specific suits to represent additional attacks instead, although exact balance becomes difficult.

SPECIAL NUMBERS

When using playing cards, we suggested adding some special effects to specific cards. If you like the idea, you can assign special effects when two identical number are rolled. The first effect, as we've said before, is allowing you to act in any segment. Other effects that can be chosen randomly or by the player are:

1 - Allow you to act on segment 0 (only applies in the beginning of the round).

2 - Any hit is a critical hit.

3 - Take an action using two segments less than it usually takes.

4 - Get a +2 bonus to your action.

5 - Immediately get an extra attack.

6 - Get a +2 bonus to AC if you defend.

And so on.

IN CONCLUSION... WHAT'S THE POINT?

Because the best course of actions depends on the d10, circumstances change from round to round. This forces you to be a bit creative instead of using the same attack round after round. Some of the effects I expect to achieve:

- With the dice on the table, initiative is easy to track.

- Everybody can act every round, and some people will act more than once in some rounds, but not too often.

- No additional rules needed for fighting with two weapons, opportunity attacks, etc.

- Things such as kicking an opponent before striking with your sword, throwing a dagger at a wizard in order to stop his mighty spell (because otherwise you cannot reach him in time), moving cautiously toward an enemy, using low-levels spell at high levels, waiting for your adversary to act first, and trading defense for offense become viable tactics.

COMPARING THE TWO SYSTEMS

There is no big difference between the systems. The d10 makes it easier to physically adjust the segments, but I think playing cards make things easier to track and avoids accidents (such as one die bouncing on another). Since many people would object to adding playing cards to D&D, I've adapted that system to a more traditional d10 system, reminiscent of AD&D - that is where I got my inspiration in the first place.

Brainstorm in a Jar: Card Initiative

After some brainstorming on the subject, here is a complete initiative system using playing cards and segments. It is intended for D&D, but can probably be used with other systems with some tweaking.

The idea behind it is that every combatant is trying to find a good moment to attack (or act), and faster characters get better opportunities. The goal is making things unpredictable, dynamic, and tactical.

In the begging of each round, each player draws a card. Rounds last for about 10-15 seconds and are divided in 10 segments, each one lasting about a second. Characters with an ace act on segment 1, deuces on segment 2, and so on. Every action has a duration in segments, usually 4.

Immediately after taking your action, draw another card. If the card is equal or higher than your current card plus the duration of your action, you can take an extra action in this round (when the segment corresponding the new card comes up). If it isn't, you can keep the card in your hand for next round. For example, if you played a 4, and took an action with a duration of 5 segments, you can act again if you draw a 9 or 10. No more than three actions can be taken each round, which can be managed by looking on the number of cards in front of each player.

There is no problem in finishing an action on segments greater than 10 (this "extra segments" are somewhat abstract), and it doesn't affect next round, but your last action cannot finish in a segment greater than your first card plus 12. Thus, if the first card you played is a 3, you cannot start an action in segment 10 that takes more than 5 segments.

When a new round begins,the cards on the table are shuffled, everybody gets an extra card, chooses what card to play (if they have more than one), and discards any remaining cards.

Although this looks like a "second-by-second" breakdown of the encounter, it involves quite a bit of abstraction to make things run smoothly. Because of that, you resolve each action on the segment it begins, although it can be interrupted in the middle.

And… that is it, basically. You can use or ignore the optional rules below, depending on the system you're using or the effects you're looking for.

ACTION SPEED

Movement: 2 segments for a 5-feet step, 4 segments for a half move, 8 for a full move, 12 for a double move.

Weapons: 4 segments for a "medium speed" weapon, 2-6 for other weapons (2 if unarmed, or if you have a longer weapon than your opponent and hasn't been hit by him yet). Bows twice the time to draw and shoot, crossbows take half the time to shoot and twice to draw.

Spells: equals to spell level+2.

The GM will decide on the duration of other actions.

You can add a simple, quick action, such as taking a single move or drawing a blade, to other, more complex, actions. Just add the durations. The GM has the final word, but the idea is that if no dice are rolled because of your action you don't need to use an additional card. For example, you can move and attack, although you have a good chance of being interrupted in the process (see below).

DELAYING

To delay an action, put your card on the table and say you'll wait for something to happen - an attack on yourself or an ally, a move by an specific opponent, etc. If that happens, you can act on any given segment after that. If you play a deuce, and act on segment 7, for example, you action ends on segment 12. If you don't want to declare your action right now, you can draw an extra card, as if you had just taken a 1-segment action.

Delayed actions can be changed at any time before the round is finished, but the change takes one segment (so, if you change your mind on segment 6, you can act on segment 7, without using an extra card).

INTERRUPTING

You can interrupt an enemy if you can finish an action before him. You must declare this the moment the opponent declares his action. For example, if he attacks you on segment 4 with a two-handed axe (speed 6), he will hit you on segment 10. If you act on segment 5, you can hit him with a knife before that. Even if he survives the attack, enough damage (let's say, 10% of Max HP or more) can cause an ill effect. Some suggestions: spell interruption, -2 penalty to a roll, movement is halted, etc.

You can use you interruption to defend yourself or an ally. This depends on the system you're using. One example: a parry grants +2 AC.

Interrupting is about disrupting your enemy's action. You cannot simply walk away in the middle of an attack.

Optional rule: if you want to interrupt an opponent but your card doesn't allow it (and its segment is yet to come), you can trade your card, drawing another, but the new card MUST be used to interrupt. If the new card doesn't allow it, you cannot use it.

TIES

If it matters at all, give them to whoever started the action first (in the event of interruptions), or flip a coin.

SURPRISE

Surprised characters get no cards in the first round.

DEX MODS

DEX mods apply only to the first card you play. If you have DEX +2 and your first card is a 5, you can act on segment 3. This may allow you to act on segment 0 but not negative segments.

Another option is to let players draw an extra card for each +1 in the beginning of the round, choosing the best and discarding the rest (because of special cards, this is quite powerful).

MULTIPLE ATTACKS

If a character has multiple attacks, he may take all his attacks in his turn. Depending on the edition you are using, you can use special cards or specific suits to represent additional attacks instead, although exact balance becomes difficult.

SPECIAL CARDS

Since you are using playing cards, add special effects to taste. Make them cool and easy to remember. Here are some ideas.

* FACE CARDS can be used to act in any segment you choose, but cannot be used twice in a row.

* When you attack using THE ACE OF SPADES, any hit is a critical hit. If you want to get fiddly, assign special effects to each suit.

* AND DON'T FORGET THE JOKER! It allows you to act in any segment, and choose a special effect: act on segment 0, take one action 2 segments faster than normal (minimum one segment), get a bonus on your action, etc.

The only card you need.

IN CONCLUSION... WHAT'S THE POINT?

Because the best course of actions depends on the cards, circumstances change from round to round. This forces you to be a bit creative instead of using the same attack round after round. Some of the effects I expect to achieve:

- With the cards on the table, initiative is easy to track.
- Everybody can act every round, and some people will act more than once in some rounds, but not too often.
- No additional rules needed for fighting with two weapons, opportunity attacks, etc.
- Things such as kicking an opponent before striking with your sword, throwing a dagger at a wizard in order to stop his mighty spell (because otherwise you cannot reach him in time), moving cautiously toward an enemy, using low-levels spell at high levels, waiting for your adversary to act first, and trading defense for offense become viable tactics.

ADDENDUM: BUT I DON'T LIKE USING CARDS FOR INITIATIVE!

If that's your case and you've read this far, congratulations! And don't worry, I got you covered.

Someone stop the wizard!

The wizard sees the intruders coming from the front door. With a malicious grim, he raises his hands and utters strange words from a long-forgotten language, summoning the dark spirits of the Abyss, as black clouds appear beside him. There is no time to run, nowhere to hide, so the adventures quickly move forward, with weapons drawn, as they try to avoid the evil minions and reach the sorcerer while they can still stop him. Their fate - and the fate of the whole realm - hinges on the very few seconds that the wizard takes to complete the unholy spell.

After doing some quick brainstorming on initiative, I couldn't stop thinking about the subject. It took me a while to understand why. As I've said before, I don't usually appreciate heavily detailed initiative rules - giving initiative based on the situation or in order of DEX is good enough for me.

Still, there is some ideas I like about “segment” initiative that are hard to find (or not as interesting) in simpler systems. The main one, as you may have guessed, is stopping a wizard in the middle of a terrible spell.

The trope is quite old and justifies a few other things I like. For example, you can have earth-shattering spells that do not make the fight one-sided, and that are very risky for the wizard himself (because he can be stopped mid-spell, and lose  the spell - not to mention his life - in the process). Spells get to be amazing being too "unbalanced".

Most important, this forces characters (PCs or NPCs) to work together. For wizard PCs, spell-casting is no longer a solo game, but a tactical group effort, with the mighty fighters protecting the frail wizard to allow him to finish the spell. For wizard foes, the PCs must find creative ways to stop their spells - maybe throwing weapons (improvised or not) at him, or just getting out of the way.

... and sometimes you just have to play dirty.

The whole concept of interrupting a foe action is interesting to me, not only in fiction and tabletop gaming, but also in video games (from Street Fighter to Dark Souls), real MMA fights and so on. Stopping a charge with a spear, stabbing a strong foe the moment he raises his two-handed axe, tripping someone as he runs towards an ally, shooting an arrow before being smashed by a mace (and vice-versa) and threatening an enemy with a blade to the throat are some of the things that I feel are missing from many of the simpler initiative systems.

This, in a nutshell, is why I want a detailed initiative system for my game, and what I'll try to do next.

On alignment, part I: Origins

Alignment, as everyone who has ever discussed it knows, is a very controversial topic in D&D and RPGs in general. One of the reasons for that is how it has changed from its inception to the present days, to a point where its whole meaning became very different than it originally was.

Alignment is described in Moldvay as a "way of life" or "belief". The d20 SRD describes it as "general moral and personal attitudes", which is very similar to 5e Basic. In the original game, alignment isn't described at all, so one might look to this more recent definitions to understand it. But OD&D tells a different story.

Although OD&D doesn't define alignment, it has a feature that is not repeated in other editions: a table with all creatures divided by alignment. Not even the monster's manuals, as far as I remember, index creatures by alignment. The table is on page 9 of Men & Magic:

Individual monster descriptions didn't necessarily mention alignments, nor did the monster reference table. On the contrary, as the table above shows, different alignments were separated in factions or "teams".

Although the table above doesn't appear in later editions, a similar one did appear in Chainmail. Since Chainmail is a wargame, the purpose seems to be to allow different armies to use some special units and not others.

This means that, in OD&D, choosing alignment was apparently akin to choosing a side in an ongoing struggle between the opposing forces of Law and Chaos. "Taking a stance", as OD&D describes.

There are other bits in OD&D that points to "alignment as faction" concept. For example, the fact that there are alignment languages (people in the same team must have a way to communicate) and that chaotic creatures will apparently recognize lawful language and attack its speakers. If "chaotic" was an evil, selfish or reckless behavior, you would expect chaotic creatures to attack anyone (specially other chaotic creatures), but they seem to be taking sides.

This would quickly change, with subsequent editions as early as Holmes describing alignment as behaviors and adding new options such as Good and Evil in addition to Law and Chaos - while on the earlier editions, Chaos was synonymous to Evil and Law to Good, as you can clearly see in Chainmail (3rd edition, page 39):

It is impossible to draw a distanct line between "good" and "evil" fantastic
figures. Three categories are listed below as a general guide for the wargamer
designing orders of battle involving fantastic creatures:
LAW NEUTRAL CHAOS [...]

After OD&D, alignment is described in every monster's description in all editions, making it look more like an individual behavior than a collective trait. There is no explicit reason for this change, nor do I know how voluntary it is. I would guess one of the main causes is a natural change of perspective from wargames focusing on large armies to roleplaying games focusing on individual characters and the interactions between them.

This change, although early, would never be complete. Most editions of D&D would use alignment both as personal morals/attitudes and as cosmic factions/teams. Take 5e basic, for example, that says:

For many thinking creatures, alignment is a moral choice. Humans, dwarves, elves, and other humanoid races can choose whether to follow the paths of good or evil, law or chaos. According to myth, the good-aligned gods who created these races gave them free will to choose their moral paths, knowing that good without free will is slavery. 
The evil deities who created other races, though, made those races to serve them. Those races have strong inborn tendencies that match the nature of their gods. Most orcs share the violent, savage nature of the orc god, Gruumsh, and are thus inclined toward evil. Even if an orc chooses a good alignment, it struggles against its innate tendencies for its entire life. (Even half-orcs feel the lingering pull of the orc god’s influence.) 
Alignment is an essential part of the nature of celestials and fiends. A devil does not choose to be lawful evil, and it doesn’t tend toward lawful evil, but rather it is lawful evil in its essence. If it somehow ceased to be lawful evil, it would cease to be a devil.

The confusion between the concepts of "morals" and "teams" is the source of many arguments on alignment. For example, should a good-aligned creature slay an evil-aligned one in cold blood, just because it is evil? If alignment is a moral choice, probably not, but if alignment is a side in a cosmic war, it depends on what is best for your team.

Another example are "lawful" characters. Are they honorable, or defenders of Law? The two will often lead to different actions. Moorcock readers will remember the scene where Elric and his companions face the Sad Giant to get his Chaos Shield in order to help the forces of Law to defeat the Lords of Chaos. Expecting to fight a fearful monster, they find instead a lonely, broken creature that offers to give away the shield without a fight, asking only to be spared. Elric decides to let him go. But his friend Moonglum (a loyal and benevolent character) immediately goes after the Giant. This is how the (awesome) comic book version by P. Craig Russell portrays what happens next:

In my opinion, alignment-as-morals doesn't work too well, because it is too restrictive to express personal attitudes (although there are some good ideas online on how to use it this way). Alignment-as-factions, on the other hand, can work well sometimes, if the literary foundations of the idea are known, and depending on the setting. But using both at the same time is a quick path to argument and confusion, which might be the reason why the latest editions have little mechanic effects related to alignment.

One problem that using alignment both ways causes is that it might make you think that people with the same "morals" are on the same "team", which is strange in real life (where you would expect even good, honest people to bump heads and maybe even go to war from time to time), and only makes sense in very black-and-white settings, where Good and Evil are easier to define and recognize. Another problem is that it doesn't make much sense if we think about D&D's literary inspirations, as David Paul explains in this great post:

The Law vs. Chaos concept first shows up in Poul Anderson's fantasy (Three Hearts and Three Lions and The Broken Sword.) In Anderson, Law is pretty much synonymous with Good and Chaos with Evil. Even the more sympathetic Chaos creatures, like elves, are pretty much always bad news for humanity.

Michael Moorcock adopted the labels of "Law" and "Chaos," but radically changed them by introducing the notion of the Cosmic Balance. In this version, the extremes of Law and Chaos are both equally bad, or at least inimical to life as we know it. Only when the two are in balance is life possible. The metaphysical forces in charge of keeping the Balance are thus, broadly speaking, good; but they can be extremely ruthless and manipulative in pursuit of their duties.

The earliest version of D&D used Lawful, Neutral, and Chaotic as the only alignments, and leaned more towards Anderson than Moorcock in how it interpreted them: Lawful characters were usually heroic, Chaotic characters were usually villainous, and Neutral characters were indifferent rather than serving any kind of Cosmic Balance. [...]

AD&D added the Good vs. Evil axis to Law vs. Chaos, creating the nine-point alignment system we all know and... know. The thing is, nine-point alignment has never made a lot of sense in terms of D&D's literary inspiration -- from the Andersonian perspective it's redundant, and from the Moorcockian perspective it tries to add objective morality to a cosmology that was deliberately designed to be morally ambiguous. Making things even more confusing, True Neutral became the alignment of both the morally indifferent and servants of the Balance -- a Balance which now included the totally unprecedented and rather nonsensical idea of parity between Good and Evil. I honestly have no clue what led anyone to think this was a good idea. [...]

So this are some of the reasons why most "modern" versions of alignment don't work for me. Granted, they may work for many people, and there are many ways to change and interpret it as to make the best use of it in any particular game (EDIT: after posting this, I have just realized Matt Finch has written about this exact subject a couple of hours ago. See his insightful post here).

I'll discuss a few of my own solutions in the coming posts. Feel free to tell me yours!