A different way to figure Armor Class - Any feedback?

[OracleOfNarion]

I was thinking...

What if, instead of calculating armor class (AC) as per the PHB (10+armor bonus+shield bonus+dex mod+size mod), a defender, when attacked, rolled a d20, and used the result instead of the "10" at the beginning of the formula?

A high roll would represent a particularly good defensive manuver, and a low one could represent lost footing, or looking the wrong way, or misjudging distance, etc...

As I see it, this would keep the players involved in the entire combat round, rather than just waiting to see if they got hit, and how much damage they took.

Also, depending on how the rolls went, players could hit things they normally wouldn't be able to, and vice versa. I'm not sure if this is good or bad... Could it significantly effect game balance?

Possible game balance issues aside, the only negative that I see is more dice rolling, which could lead to slower combat, which could then lead to boredom...

I'm sure somebody's tried this. If you have, please share how it worked out.

Thanks.

 

[Maximillian]

There's a variant just like that in the DMG.

Advantages: All the things you just said.
Disadvantage: More dice.

I've toyed with it here and there, but only in one-on-one duels.

 

[Destil]

The game would be more random, which makes it intrinsicaly more of a danger to the PCs over the life of the campaign. And there's more rolling.

Not really much else to comment about, myself. You can do it with spells, too. (Psionics has it as the default).

 

[SylverFlame]

I'd use it but instead of rolling for every attack, why not every round of combat? That lessens the dice and upholds the realism the house rule goes for.

 

[Eternalknight]

That last one requires more work for the DM - trying to remember everyone's AC every round.