
![]() |

In a PC-Build exercise I came across something that concerned me.
The Luck Domain allows a Cleric to reroll a d20 in any situation before he or she learns about the result of the roll. So if the Cleric rolls a five on a Ref Save, for example, the Domain ability allows for a reroll showing the Cleric's luckiness.
Got it. It makes great sense -- well done.
But what about these situations:
The PC Fighter rolls an attack and totals a 23 -- misses.
The PC Rogue rolls an attack and totals a 26 -- hits.
The PC Cleric w/ Luck rolls an attack and totals a 24 -- misses.
Next round.
The PC Fighter rolls an attack and totals a 29 -- hits.
The PC Rogue rolls an attack and totals a 25 -- misses.
....Now the Cleric goes and knows exactly what is needed to hit the 26 AC. So the Player knows after the d20 is rolled and BEFORE the DM says whether it hits or not, if it hits -- thus creating a loophole in the Luck Domain ability to reroll a d20 before knowing if it's succesful or not.
And there's other times, too.
Reflex Saves for the Dragon's Breath Weapon:
The Rogue rolls and says, "I got a 19" -- and takes full damage.
The Fighter rolls and says, "Wohoo, I got a 24" -- and takes 1/2 damage.
So the Cleric rolls, calculates a 20 and since everyone knows a 19 fails....
And what about Knowledge Skills?!
The Wizard says, based on a Knowledge Planes check, that Demons & Devils have SR CR+11 -- even if they articulate it a bit differently, the Cleric -- and more unavoidably, the Player running the Cleric -- knows that means the Hezrou has an SR of 22.
So when the Cleric rolls to punch SR, the Player knows BEFORE the DM says anything if the roll was successful punching SR.
Is this a problem?

Jak the Looney Alchemist |

Solution: 1. All attack roll numbers will be handed in on little slips of paper so as to not compromise the integrity of the luck cleric.
Edit: Email is faster and saves trees.
2. Trust the cleric to roleplay his character and not metagame his knowledge of the opponents ac and only to reroll on really bad rolls.
3. Don't worry about it. If the pc is burning rerolls on attack swings he's probably wasting them.
Edit: The following concerns can easily be corrected by following solution 2. If you character is using out of game knowledge to influence his character then the problem isn't the power.

![]() |

Jak,
Solution 1 is not playable.
Solution 2 is obviously what I would do in a game like this -- but of course, other groups are different and I just wanted to see if anyone else has thought of, or dealt with this before.
(Ultimately, the whole concept of Luck allowing rerolls is COMPLETELY based on Metagaming -- there's no way to do it otherwise.)
Solution 3 -- well, not a solution but, yeah, I'm not even sure if it is a problem -- ultimately, who cares, right? But I wanted to bring it up.

Jak the Looney Alchemist |

Solution 1: Sure it is people do it all the time. I've done it myself on occasion. Irritating hell yes. Unplayable no.
Yeah solution 2 is ideal, but what are you going to do. Solution 3 is the question of is it worth it to chew out my player for using metagaming knowledge to reroll a save/attack/spell resistance check. The only answer to that is dm and situation based.
Personally I'd call them on it and let it slide the first time.
Your proposed solution 4 is alright, the only problem is the implementation of it. If you've got players that don't metagame then you're weakening their power. If you've got players that do metagame then you're installing a limiter on it. The problem is the dm isn't playing fair if they institute house rules on the fly. The character should know what their powers do when they take them. It becomes a question of foresight.

MC Templar |

once per day at first, a maximum of 3 times per day at 18th, I would say it isn't significant enough to waste your time as a GM to address.
If it helps, approach the luck domain on a macro level, the cleric of the luck god is supposed to be 'noticeably' lucky to a casual observer. The only way that can come out of a dice based rules system is if probability skews in the direction of the cleric. Therefore the class feature is designed to account for the specific type of metagaming that you are considering a problem, the re-roll is intended to turn a situation that should statistically be a miss and give it an additional chance to hit.
Instead of view the limitation of the rule as merely intended to restrict times where the cleric believes his or her roll was good enough for a success, only to find out it wasn't and then initiate the re-roll. View the limitation as strictly "keeping this free re-roll ability from slowing down play by forcing the GM to describe success or failure (which some may do at length) and then re-roll and reinterpreting the scene"
Think of it as merely a 'speed of play' issue, and the meta-gaming won't bug you so much.
(we had this same issue with ebberon action points and the GMs became playful used car saleman trying to talk players into spending action points in circumstances where it was mathmatically impossible for them to make a difference)

Mort the Cleverly Named |

I don't find this to be a problem. It mostly seems like characters using in-game knowledge for in-game benefit, to me.
In the first scenario, the Cleric has seen five attacks against the monster. He sees another swing coming in close, but based on observation of the fight knows it is likely to miss. So, he uses his luck powers to slightly tweak things.
In the third example, the Wizard has articulated in some way what the Hezrou's SR is. The Cleric, armed with this information, knows his spell probably isn't going to break through and uses a bit of luck to try and succeed. The issue here would be how, in-game, a wizard would articulate exact knowledge like that.
The second example is the only one that would be a bit odd, since in my experience players tend to roll saves at the same time or randomly, which gives the Cleric a slight metagame benefit for waiting to roll last. If one enforced a certain order of saving based on who was hit first you could avoid this issue, but that seems like a lot of hassle for an uncommon, and generally not game-breaking, scenario.