I think the idea of giving a brief but tantalizing description of a room or scene, and then letting the players “bite” on the description, is the way to go. When I DM, I expect activity, not passivity, on the players’ part. Once their attention and interest latch onto something in a description, they get into the game. Rolling dice, for the most part, cannot do that; that’s about as personal as a game of craps in your local casino.

It is a social game. If the player isn’t a genius but his PC is, then there’s game mechanics or your judgement as a DM to make him spot something that only a genius would.

I also hope that we (as a community) can give the whole “old school” role-playing halcyon nostalgia a rest before I snap and start telling my stories about how we used to LARP walking ten miles up hill through the snow just to get to the dungeon. I was into D&D before there was such a thing as a Player’s Handbook, so I’ve certainly got the right.

I don’t think that Ben is suggesting getting rid of rolls entirely, just that there are many cases when rolling does not help. If something is important for the plot or sounds like a cool idea then it should probably just happen. But for things like swinging a sword in combat or leaping a chasm you should still roll.

The decision as to if the player succeeds, does not need to be just based on what the players can do. For example, the person that spots the important plot point should usually be the one with the highest spot check – even if you aren’t rolling. If the genius scientist is examining something then they should get more info then the brawny fighter.

The problem with the “My character is smarter than me, so they should know what to do even if I don’t” line of reasoning is that it can easily take away from the players being involved in the game. I think most folks would think that saying “My character is an expert tactician, so if I make my roll, move me to the best spot on the board” is obviously losing something. The key point is that the characters should be encouraged to be active in deciding what they do and then their success is determined by the skills of the character (with or without a roll) and what is best for the game. Having the players describe what/how they are searching “I think they might have hidden it in this room, and as a thief I’m pretty familiar with how to conceal things, so I’m really paying attention to places where it would be hidden by someone who knew how to hide stuff” is more involved than saying “I rolled a 23 on my spot/search check, what did I find?”.

When it comes to the player saying “I’m a genius scientist so I’ll whip up a device to invert the tachyon particles….” then the DM should either say “yeah in your lab with your skills that should be pretty easy for you” or have it be a roll. Depending on where the DM feels the task falls in the spectrum from Plot-related-spot-check thru swinging-a-sword-in-combat determines if you need to roll for it. If the cool device is a great way to defeat the monster and this is a good point in the story for it, then why not just say “yes you can do that”. If you think it’s a good thing for the story & game and he wants to do it (that is, if everyone in the game wants it to happen) why risk having him fail the roll and then feel like his scientist is an idiot?

Also look at the “virtual roll” post: rolling-for-roleplaying-the-virtual-roll as a good “it sounds cool, but I still think it needs a roll” compromise.

Every spot check is a series of difficulties. I will give an example of the party going through a forest and meeting a bear:

Spot DC -10: You are in a forest (okay joking on this one)
Spot DC 10: The bear have bloody paws
Spot DC 15: Thats probably from the dead animal lying behind it, hidden in the underbrush
Spot DC 30: It looks like it just ate a lot, so it probably isn’t hungry

The point being that the better the roll, the more information, but the party isn’t excluded from clues with a low roll, the clues just get better.

I agree that rolling should not replace thinking, and just wanted to point out that GMs got lynched back in the old days, too, when they forgot to think (Well you didn’t SAY that you looked at the floor, and that’s why you didn’t notice the pool of lava….).

Along with the spot check, I think the game has evolved from “don’t tell them unless they ask” to more of a “tell them everything upfront EXCEPT for this (Spot DC).” In the end, I think the latter premise probably keeps the game moving better.

But first, I’m going to draw an analogy with social skills: For a long time I disliked the idea of social skills. “If you can do it (i.e. roleplay it), why are you rolling dice?” But I’ve since come to realize that the correct application of social skills is actually advantageous because it removes arbitrary decision points.

One of the things I love about playing RPGs is when unexpected surprises spontaneously emerge out of gameplay.

So if the PCs lie to an NPC I have a decision to make: Does he believe it? Without a social skill, I have to either say “yes” or “no”. But if I add a Bluff skill to the game, I can suddenly say “maybe” and see what happens. Even more importantly I can vary the likelihood of the lie being believed depending on how believable the lie was.

Rather than closing down social interactions, it opens them up: Players become engaged with the game as they try to figure out how to use their skills to their best advantage.

Now, let me bring this back to Spot and Search checks: I vary the difficulty of the check by what the player tells me they’re doing. Is there a bit of writing carved into the lower lip of a desk? That’s difficult to notice if you’re searching the entire room, but far more likely to be noticed if you’re focusing your search on the desk itself and will almost certainly be noticed if you tell me that you’re specifically looking at the underside of the desk.

Perhaps the most common example of this is the infamous “I look up” instruction that happens whenever they get ambushed by something dropping down on top of them. But just because they’re looking up doesn’t necessarily mean the pirates should spot the ninja lurking in the shadows there — maybe the ninja is just that good at hiding. So, when you give me that instruction, you get a bonus for spotting things on the ceiling… but suffer a penalty to noticing things anywhere else.

This type of mechanical differentiation encourages the same type of proactive play that you’re advocating; but it doesn’t discard the advantages of using mechanics, either.

As for the example of a scenario falling apart because the PCs don’t notice one particular clue… well, that’s just as likely to happen without Spot checks. One solution is to make sure you always give them all the clues, but I find a better solution is the Three Clue Rule.