green slime
First Post
In DnD, it'd probably be more effective for the troll to just waste the door, rather than push against two fighters holding it closed.
Strong Fighter, Wimpy Mage
- Thread starter Nebulous
- Start date
Exactly. Unearthed Arcana was aware of this and included rules around using the 3d6 bell curve instead of a single d20. (blasphemy, I know, but there you have it)Stalker0 said:
I have read somewhere (DMG perhaps?) where it says for some skill checks to roll a single d20 and then add the individual PCs' skill modifiers to the die result to derive their individual skill check results. I think the discussion was around Spot checks, otherwise there is a high probability that at least someone will see the craftily hidden ambusher simply because of rolling 4 lots of d20 and the probability of obtaining at least one high result. Could you apply the same method in the situation with the door?
But what bugs me is when the 18 Strength fighter assists the 24 Strength raging half-orc barbarian in lifting a portcullis and gets a higher result, which may be enough to open the portcullis on his own, but is wasted as the fighter only needs a DC 10 result to assist the barbarian and the result is discounted. In these sorts of situations, we have house ruled that the higher of the two results is used with the +2 from the assistance if the other character made the DC 10 check.
I've seen other systems- don't ask me which, because its late and I've played as many systems as most pros- where to minimize the number of oddball rolls, you use a "dice pool" method of rolling.
In such a system, the PC doesn't just get a simple dice roll, he gets several. He then compares "successes" to "failures." To succeed, you must have more "successes" than "failures."
Thus, in a modified version of D20, the Str bonus would tell you how many dice to roll, rather than how much of a bonus to a single roll- so +5 would be 5 die, not +5 to a single roll. The wimpy Wizard would have fewer die to roll, thus a lower chance of success.
The good news: oddball results get minimized.
The bad news: rolling lots and lots of dice...and then lots more.
In such a system, the PC doesn't just get a simple dice roll, he gets several. He then compares "successes" to "failures." To succeed, you must have more "successes" than "failures."
Thus, in a modified version of D20, the Str bonus would tell you how many dice to roll, rather than how much of a bonus to a single roll- so +5 would be 5 die, not +5 to a single roll. The wimpy Wizard would have fewer die to roll, thus a lower chance of success.
The good news: oddball results get minimized.
The bad news: rolling lots and lots of dice...and then lots more.
prosfilaes
Adventurer
Dannyalcatraz said:
White Wolf systems do this. One problem is, it's statistically fragile; in several of their games, you're more likely to crit fail at times the more dice you have. I don't see any real advantages to this above rolling something like 3d6, which gives a pretty decent approximation to a bell-curve. You could also double the modifiers and up the DCs appropriately.
Honestly, I don't particularly want a game where only one character has any chance of making a certain check. It's nice for the rogue to sometimes remember a religious detail, or for the cleric to notice something.
I'm just trying to help the OP. Personally, I don't have problems with statistical anomalies.
While in the early days of studying for the Tx Bar exam, a buddy of mine came to visit. I had 3 years of law school, he was still in college, and yet for 20 minutes, he was ahead of me on the multiple choice questions (a section I aced on the real test).
While in the early days of studying for the Tx Bar exam, a buddy of mine came to visit. I had 3 years of law school, he was still in college, and yet for 20 minutes, he was ahead of me on the multiple choice questions (a section I aced on the real test).
glass
(he, him)
I think the key bit there is 'for six seconds'.FireLance said:
It's not a question of if the Str 1 child can hold the door against the child, it's how long. The most likely answer is 'not at all', but if she is really lucky, the child might hold the door for a round or two. I consider that a feature, not a bug.
glass.
Stalker0 said:
And maybe there are no such explanations.
The scenario is one of an unlatched door. How many doors have you ever encountered in life that are stuck? I suspect very few. Doors are designed to be easily opened. Otherwise, they would be called a wall.
The DM determines the environmental factors. So, if he thinks that the door might be stuck or rusted, he would give a bonus to the side trying to keep it closed.
But the example we are discussing here is how the rules model the holding of a simple door.
They model this concept very poorly if you use a Strength Ability check.
Just like they model a Tug of War very poorly. No, we do not go into "Maybe the rope gets caught on a rock" when discussing the fact that the Baby wins the Tug of War 9% of the time against the Troll. Such "environmental factors" are irrelevant as to whether the actual rule is a good model or not.
FireLance
Legend
Very poorly for realism, perhaps. But not so bad if you want the potential for a heroic outcome.KarinsDad said:
Only if the outcome of a tug of war depends on only a single opposed Strength check. At the risk of complicating the gameplay, I'd probably require the winner to be the first to achieve three (or more) successful Strength checks more than his opponent, or make the degree of success important, e.g. the rope moves 1 foot for every 5 points you beat your opponent's score, and winner must move the rope 10 feet.
werk
First Post
FireLance said:
Right, that's what we're saying, a single d20 is a huge variance, especially for low modifier or DC rolls. All you are doing by requiring multiple rolls is normalizing the data. In that vein, and to the OP's cause, I would suggest rolling 3 d20 for every check and using the mean average...it generates that bell curve and reduces extreme results.
Similar Threads
