The most important thing you can do while identifying the Traitor is to have "Show Me the Odds" checked off.

If you're feeling brave and you have turned off the odds, check for traitors by sending minions on missions where they should succeed from having a sum of 10-15+ points between their skills above the opposing character/location but wind up with bad/ambiguous results anyway.

• If you have a character who undergoes a mission by themselves and displays a 100% chance of succeeding but comes back with a failure or mixed results, that minion is your traitor. This is because there is a range of numbers where traitors will flub their results even when the game displays a 100% chance of success. This is NOT a bug, it is damning evidence.
  
  
• More often than not because you are assigning characters to missions that they're good at (Althea, Vonne, Mallory to Question; Rowan to Stalk; Thais to assign Detention, etc.), you're not looking for outright failures, but ambigious results. Make sure to compare the odds given by the game to the actual results you had the day before.
  
• Further Analysis by Franka_Scythe:
  
• The Traitor has a higher rate of bringing back lemons than your other minions.
  
• The traitor will not interfere with the results of a mission when paired with a partner.
  
• Until you have beaten the game once, Rowan cannot be the traitor. (Hanako is on record for saying that Rowan being the traitor makes the early game harder, so she is not available until you have won.)
  
• If you have gotten the traitor to 45 loyalty by the week following Midterms (October 24th?), she will confront you and tell you her motivations for her actions before Talmage makes you identify the traitor.
  
• On the other hand, if you get a girl's loyalty to 45 by then and she does NOT confront you, she cannot be the traitor.
  
• The traitor can be expelled or re-recruited (if you do not reveal her name and make her your queen for the festival). A re-recruited traitor will not subvert future missions*.
  
• (reported by Sung) The week of the mid-term exams, you can "joke" that there is a traitor amongst the student council. The traitor is always one of the two students who will react.
  
• (Consequently, if you save before that meeting and replay it, you may get a different combination of girls. But as the traitor is always the same, she will always be one of the two girls who react.)

*- There is one exception. If Rowan is the traitor, you keep her on board and betray her trust later on, she will return to making your life harder.

For this example, I'm starting a new game with a random traitor. I don't want to know in advance who's gunning for me.
I get my first case, and luckily there are five things for me to do.
Day 1:



First, I send Rowan to the chapel to find the supplies. I send her over Vonne to inspect, as Vonne has stronger charm than Rowan, which will be more useful in Questioning one of the four girls in the case.


She comes back with evidence the supplies were there, but the results were "Too Close to Call," so I can't outright eliminate her as a possibility even though she has a lead.



Next is Althea. She succeeds. So it's not her.



Vonne is also loyal.



Mallory comes back with ambiguous results, but it's not too surprising given that her target has high sneak. We'll still check her for now.



Thais comes back with mixed results as well, but she wasn't going to do well with questioning anyway. That being said, the actual girl she questioned gave a suspicious response. In the case itself, it tells me that I need to press this girl for more information. It also tells me that Thais might be the rotten apple. At the end of the day, we've eliminated Althea and Vonne from just a single day of activity.

Day 2:



I need to confirm Mallory isn't the traitor because she didn't succeed yesterday. So I check the results she had against the odds and they were actually kind of bad. This isn't incriminating, but doesn't clear her either.



I double check Thais' results from the day before and note that she was more likely than not to have succeeded. After my conclusions from yesterday, this is a red flag.



I decide that for today, making progress with this case is more important than getting more leads on the traitor. Harassing Trina opens up her room as a location for searching while the other girl names Trina and another girl as suspicious in his case.

Day 3:


I want to pin down Mallory today. I deliberately send her solo on a useless mission to see what she'll bring back. Meanwhile, I check Rowan to search the room for the case as she seems to have succeeded the first time.


Rowan gets another "Too Close to Call," but it seems to have succeeded as she found the box.



Mallory, meanwhile, has succeeded at her task. It's not her. It must be Thais (who I'm I have a red flag on) and Rowan (who has come back with results so far, even though none of them have been explicit successes).



Day 4:
I don't trust Thais at the moment, so I send Althea (who has the next best intimidate) with a pair of high heels to give the case's culprit detention. The mission comes back accomplished.

Conclusions: Althea, Vonne, and Mallory all had independent successes so they are explicitly not the traitor. Rowan didn't have any independent successes, but she still found the supplies even when the results were close. Meanwhile, when Thais questioned the girl who was the eventual culprit, she failed to get any evidence against her even though the odds were in her favor. Thais seems to be the traitor.

Again I choose a random Traitor.


Another 5 tasks.


Like in the previous example, I send Rowan to inspect over Vonne because Vonne's higher social rank makes her better at first conversations.


I get a lead in the case. Additionally, the Traitor isn't Rowan.


Althea seems to succeed but look closer. Her roll is a 1 and the other girl's roll is a 20. Doesn't that seem super contrived...?


Vonne succeeds, and identifies the two girls I need to check more extensively.


Mallory succeeds, but the girl has a generic response.


Thais succeeds, but the girl has a generic response.

At this point, it's all eyes on Althea. I decide to confirm my suspicions in one quick way:

Ooh, yeah it's her.

Coincidentally, the next day I sent Rowan to the new room that came up and she found all of the supplies and solved the case.

Can you say best case scenario?

This space is reserved for the first example of gameplay at a higher difficulty level.

As was pointed out in my ignorance, harder difficulties change the rules a bit.

While playing on Loaded Dice, you can identify a traitor by spotting mixed results on a case where your minion should have a 100% chance of success.

This is not the case on Normal and harder difficulty and some of the logic and assumptions we made that were were useful in the first two examples are outright fallacious here.

For starters, upgrading to Normal Difficulties from Loaded Dice adds a new bug in the ointment:

If the actual chance of success is 100% (your minion has 29 points more than the opponent), the traitor will succeed. It is only an artifact of playing with "Loaded Dice" that can make the traitor fail in spite of 100% chance of success.

Let's review these intervals for results.

If you are over 29 points more than the opponent, all minions will always succeed.
If you are between 11 and 29 points (inclusive) more than the opponent, all non-traitors will always succeed. Traitors will appear to have ambigious results.
If you are between 10 points higher or lower than the opponent, you will get an ambigious result.
If you are 11 points lower than the opponent, you will fail.