I have a cleric player. Every time the party has an NPC dialog, he announces something like "I looked at him carefully, trying to understand if he is telling the truth", implying using Insight.

Normally I ask him to roll the Insight check. Sometimes I try to give a hint, make it sound vague, but on clear success he just gets more information.

That doesn't feel right to me. Every roll has to have some kind of negative consequences, but in this case the character takes no risks. He either gets information (and draw his conclusions) or not (and continues the dialog normally). Repetitive announces feels unneeded and too straightforward.

On the other hand, I want to keep his ability to use his character's social skills. The cleric is proficient in Insight and is supposed to "read" people.

How can I improve this situation as a GM? Giving advice to the player is an option.

Before you mark this as a duplicate

I've seen this question prior to my asking. It has an answer, very broad, about how to make Insight checks as a DM. It doesn't cover two points though:

  • aside from misleading hints, what negative consequences a failed check should have
  • how can a DM handle the described "problem player" situation
  • \$\begingroup\$ Regardless of whether the first answer is palatable (remember that it has more than one answer), duplicates are judged on whether the questions are the same, not whether the answers are desired. These appear to be the same question, as they are asking how to handle the frequency of insight checks. This question doesn't appear to contain a problem-player problem that's different from “player likes to use it all the time” (which is what both questions are about). The bit about what risks, if that is important, is probably the thing to rewrite this to be about exclusively. \$\endgroup\$ Dec 6, 2016 at 17:53
  • \$\begingroup\$ @SevenSidedDie fair enough, it has a relevant answer rpg.stackexchange.com/a/62173/27377 \$\endgroup\$
    – enkryptor
    Dec 6, 2016 at 18:04

1 Answer 1


Good question!

I think that what you have to remember is that the DM decides when checks are called.

From the sounds of it you have already got your answer in your question! I think it's great that you have already established how insight would or should work, what you should consider is that if there is no negative potential then there is no need for a dice roll, you could simply explain what the PC gets a feel for from the conversation.

The negative outcome from an insight check could be a false positive. Example:

"If you go through that door, certain death awaits you!"

I want to look him over and see if I believe what he is saying

Roll insight

A high roll, may reveal that the NPC is scared or that he is being deceitful. A low roll however may reveal that as far as the PC can tell, the NPC seems confident or that he is absolutely telling the truth.

  • 1
    \$\begingroup\$ The problem is, players can see their rolls. For a player, a false positive is obvious it is a response on a "1" roll. \$\endgroup\$
    – enkryptor
    Dec 6, 2016 at 17:00
  • 2
    \$\begingroup\$ Hi Enkryptor, it sounds to me like you are having an issue with meta gaming rather than a problem with insight rolls. Perhaps you should ensure that your players are RP'ing what their characters know instead of what they infer from their dice rolls :) \$\endgroup\$ Dec 6, 2016 at 17:28
  • \$\begingroup\$ @enkryptor Not necessarily. Ability checks are not automatic failures on a 1. If the NPC is a bad liar, it may only take a DC of 5 to see through his lie. This is easy even for a level 1 character to pass: Prof +2 + WIS mod +2 + 1 on the roll = 5. \$\endgroup\$ Dec 6, 2016 at 17:51
  • 1
    \$\begingroup\$ @LegendaryDude I'm not saying "1" is an automatic failure, I'm saying it is low. The die roll becomes a spoiler. \$\endgroup\$
    – enkryptor
    Dec 6, 2016 at 17:53
  • 1
    \$\begingroup\$ A good solution to this is for the DM to roll a die behind the screen, take a second to think, then give the answer. Even if you're not actually using the number, the act of rolling a die behind the screen gives the impression that there are other variables besides their die and makes the meta-knowledge less absolute. \$\endgroup\$ Dec 7, 2016 at 13:58

Not the answer you're looking for? Browse other questions tagged .