Confidence is a frequent topic of discussion in the judge program. We talk about the importance of being confident in a ruling. We tell people they're overconfident. We tell people they're underconfident. Imposter syndrome. How to practice your confidence and improve your self-esteem. Etc.

In my experience, these discussions have a tendency to get confused; people talk past each other because they're using the word "confidence" to refer to two different concepts. For the purposes of this article, I'll refer to these as social confidence and epistemic confidence.

Social confidence

Social confidence is how you come across to other people. Are you relaxed or tense? Do you speak clearly and with authority, or do you "um" and "ah" and trip over your words? Do you walk upright or slouch? Do you look people in the eye, or avoid eye contact? Are you comfortable being the center of attention, or do you shy away and try to hide? Do you interrupt people? Do you end your sentences with an upward inflection even when they're not a question? Do you respond to things right away, or do you freeze for a moment? These are all things that contribute to your social confidence. Someone with high social confidence is perceived to be experienced and sure of themselves. Someone with low social confidence is perceived to be newer and uncertain. Your social confidence has a huge impact on how much players argue with you and how likely you are to be appealed.

Social confidence is not the same as likability or charisma. That obnoxious judge who nobody likes because they're always short and to the point may have high social confidence and low charisma. That judge who's always stumbling over their rulings, getting appealed, and making self-deprecating jokes that endear them to their coworkers could have high charisma but low social confidence.People have different amounts of social confidence in different situations. A judge may have low social confidence when making a ruling, but high social confidence once they start talking about anime with their friends. For the purposes of this article, I'm talking about social confidence in their role as a judge.

Epistemic confidence

Epistemic confidence is how you perceive your own knowledge.Epistemology is the study of knowledge and knowing. How sure are you that what you're saying is correct? How strongly do you believe it? Is this something you've encountered 12 times before, and you know exactly how to handle it? Or are you seeing this for the first time, and you think your answer is probably right, but you don't really know? When you're handling a judge call, epistemic confidence is what dictates whether you should get a second opinion or go ahead and make the ruling.Just like social confidence, epistemic confidence is relative to the situation. You may have high epistemic confidence when you're asked what flying does, and low epistemic confidence when you're asked a complicated layers question. For this article I'm going to be talking about people's average epistemic confidence across all of the situations they encounter as a judge.

How they relate

These quantities are separate; someone can be high in both, low in both, or high in one and low in the other. But they are strongly correlated. The reason for this is because social confidence only exists as a useful concept because it's an indicator of epistemic confidence.

When we're interacting with other people, what actually matters to us is their epistemic confidence; can we trust them to know what they're talking about? But epistemic confidence is mental; I can't look at someone and know their internal state of knowledge. So we evolved ways to look at someone's external behavior and make an educated guess about their epistemic confidence.

If someone is speaking quickly and clearly, it's likely that they're thought about this topic before. If they're stumbling over their words, they're probably figuring things out as they go along, and their conclusions are unlikely to be as sound. So we treat verbal fluency as a signal that hints at someone's epistemic confidence.

This is why players are more likely to appeal if the judge displays low social confidence; the players will take this to mean that the judge isn't very sure of their ruling, and it's more likely to be wrong.

It's also more effective (on average) to be pushy/argumentative towards a judge displaying low social confidence, because if their epistemic confidence is lower, it'll be easier to change their mind. Often this isn't done via the exchange of facts, but in a form of dominance contest; whoever displays the higher social confidence "wins", and the other person meekly accepts their answer, even if it isn't well-explained or doesn't make sense.Riki Hayashi has written a bit about this effect here. The linked article therin by Nick Fang can be found here.


Social confidence is an imperfect proxy for epistemic confidence. This makes it very open to goodharting; people optimize for social confidence in order to be perceived as a better judge, even though epistemic confidence is what really affects their job performance. This is what leads to overconfidence.

There are two different ways to be overconfident. Either your social confidence is higher than your epistemic confidence, or they're both in step but your epistemic confidence is higher than warranted.

Having high social confidence and low epistemic confidence is not inherently a bad thing; in fact it's often desirable. As judges, we want players to listen to us. Judges who are frequently appealed are told to "act more confident" in order to get players to trust them. This is ok! The judge was hired because the tournament organizer thought their skills were good enough, and given that that's the case, they need to be able to make rulings that players will listen to.

This is, in some sense, deceptive. It's manipulating a social signal in order to get players to draw a conclusion that isn't what that signal normally means. But that's something people do all the time, and arguably the players were drawing the wrong conclusion in the first place, so all this is doing is correcting for that.

So "be more confident" is good advice for newer judges, as long as it's understood that it refers to social confidence only. When people start applying it to epistemic confidence too, this leads to the much more damaging form of overconfidence. Not only are other people misjudging you, but now you're also misjudging yourself. You'll take on responsibilities you're not ready for and give people bad adviceA few months ago I saw a great example of this. Judge Alice took a ruling about how Blood Moon interacted with Urza's Saga. They thought it would make the Saga die, but weren't super sure about it, maybe 70%. Alice went to judge Bob to ask for confirmation of this ruling, and Bob said they were 99% sure it would stay on the battlefield. Alice thought to themselves "wow, 99% sure is much more confident than I am; I can trust this ruling", and went and issued it. Is it possible that this was Bob's only mistake out of a hundred rulings they were this certain on? It's possible, sure. But far more likely is that Bob was overconfident, and led Alice to issue an incorrect ruling because of it.Meanwhile, Alice learned from this interaction that a lot of judges are overconfident and to not blindly trust their self-reported probabilities..

This type of overconfidence can stem from the fact that our minds are not always good at keeping things separate. When we act socially confident, we can trick ourselves into thinking we should be more epistemically confident as well. Deception of others can easily turn into deception of ourselves if we're not carefulAfter all, social confidence evolved as a "giveaway" of your epistemic confidence that you don't have complete control over. So it's easier to act socially confident if you're also epistemically confident, even if that confidence is unfounded.

The way to combat this form of bias while still remaining socially confident is to be explicit about your uncertainty; don't try to hide it. Ironically enough, people who are straightforward and self-assured about their uncertainty are often seen as more socially confident and more reliable, even when they're expressing low epistemic confidence. Rather than say "this is the answer", say "I'm 80% confident this is the answer". That won't make you look bad, and people will appreciate your honesty and openness. Getting into the habit of talking this way (and more importantly, thinking this way) prevents your high social confidence from warping your self-perception.


The inverse problem is that of underconfidence. "I need to be more confident" is one of the most common areas for improvement newer judges identify in themselves. Just like overconfidence, social underconfidence and epistemic underconfidence are related, and tend to come hand in hand.

Social underconfidence occurs when the judge knows their ruling is correct, but isn't experienced enough to convey this clearly to the players. They hem and haw, stumble over their words, and convey a general sense of incompetence, even if they've given this exact ruling 10 times before.

Epistemic underconfidence is where the judge thinks they're less accurate than they actually are. They may be getting 98% of their rulings correct, but still feel the need to ask another judge for help nearly every time.

Which is worse?

As mentors, we're often faced with the choice of which direction to nudge people; do we try to increase their self-esteem and risk making them overconfident, or do we point out just how far they have to improve and risk making them underconfident? To which side should we err? The same question goes for self-improvement; which should we be more concerned about?

Making this comparison is difficult, as they manifest themselves in very different ways.

Which has a worse effect on players? An overconfident judge will accept tasks they're not ready for, and fail. Incorrect rulings, poor planning causing an event to run overtime, etc. This can seriously damage a player's experience. An underconfident judge on the other hand will just not apply for the position, since they think they're not ready, even if they are. This is much less damaging to the players, but that doesn't mean it has no impact; the store may end up hiring a less skilled judge instead. Underconfidence also leads to excessive second-guessing and double-checking rulings, which can slow down the event. Underconfident judges can annoy players who feel their time is being wasted by an incompetent judge, but overconfident/arrogant judges can also annoy players who feel they aren't being listened to. Overall, overconfidence likely has a more detrimental effect on players than underconfidence does, but they're both harmful.

Which is worse for the judge? Overconfident judges might become embarrassed by a failure they could have avoided, but this is a minor and transient issue. Overconfident judges may be identified as such by their peers, but this is uncommon; the reason overconfidence is so prevalent is that it's a beneficial social strategy. We're attracted to confident people, and we aren't so great at telling when that confidence is warranted or not. Overconfident judges are often offered more advanced opportunities than judges who accurately represent their skills. In the long term, overconfidence leads to stagnation, as the judge has solidified their self-image and doesn't continue incorporating feedback and trying to improve. This may be frustrating for them ("I'm such a good judge! Why does everything keep going wrong at my events‽ I guess the universe must hate me."), but usually not. Underconfidence on the other hand tends to lead to the judge not advancing, not gaining new experience, and not improving. They give up too quickly when presented with a challenge, because they've already decided that they're not good enough. They don't stand out from the crowd, they pass up opportunities they're offered, and they may even end up deciding that the judge community would be better off without them. So underconfidence is far more detrimental to a judge's professional advancement.Note that this is itself a problem; ideally we'd like tournament organizers to be better at recognizing the difference between true skill and boastfulness. But right now they're not very good at that, and the people who loudly sing their own praises tend to be hired the most.

Which is harder to fix? Overconfidence tends to lead to clear feedback the judge can use to improve. "You got that ruling wrong." "You failed to explain to your team what they actually needed to do." Underconfidence doesn't get the same sort of feedback. An underconfident judge who doesn't apply to an event never gets feedback that they should have applied. If they double-check a ruling they didn't need to double-check, the judge they double-checked with doesn't have a great way to know about that, and won't give them the feedback of "you should double-check yourself less". Only a mentor who's been working with them for a long time can notice the pattern and recommend they become a little more independent in their rulings. So minor overconfidence easier to notice and correct for than minor underconfidence.

However, serious overconfidence can lead to people dismissing negative feedback. If I think I'm great, and someone tells me I'm not great, then clearly they're just wrong, and I'll search for justifications for that. But wait a minute; isn't underconfidence the same? If I think I'm terrible, and someone tells me I'm actually ok, clearly they too must be wrong. But the situation isn't quite symmetrical; overconfidence leads to you thinking you know more than other people, so their feedback should be weighted less heavily. Underconfidence leads to you thinking you know less than other people, so their feedback should be weighted more heavily. So while underconfident people are just as subject to confirmation bias when thinking about their own actions, they're less likely to dismiss other people's opinions, and tend to be more reachable by feedback.

So which is worse overall for the judge program? Overconfidence is the one that most damages our ability to run good events and fulfill our responsibilities to players and tournament organizers. Underconfidence is the one that most damages our ability to have fun, form friendships, and enjoy ourselves. Which of those goals you consider most important is up to you.It's the first one.

Imposter syndrome

A quick aside on imposter syndrome. The way imposter syndrome is usually presented is as a specific form of underconfidence. It's when your social confidence is high, you have some decent accomplishments under your belt, yet your epistemic confidence is lower than it should be based on your true skills.

But this can't be the whole explanation, because imposter syndrome often doesn't go away once you're made aware you have it. This means it can't just be underconfidence, since if it were, you'd just increase your confidence based on this new knowledge and be done with it.

Rather I think imposter syndrome is less about confidence and more about our inability to read other people's minds. I know all my doubts, I know all the times I only got to the right answer because of a mistaken reasoning process, I know all the stupid things I've done that no one ever found out about. But I can't see those things in other people, and when it's someone I respect who seems to be good at what they do, it's easy to assume that that's because they don't have such hidden failings. So we think "wow, everyone around me is perfect, but I have all these internal problems that other people don't know about; I'm a fraud!"

can i cure your imposter syndrome in 1 tweet? worth a try --->

— Rachael Meager (@economeager) October 11, 2021

Or if you'd rather hear it from Neil Gaiman: The Neil Story.

Just knowing that it's an attribution bias doesn't necessarily fix it either, because you can't get clear evidence of other people's internal mental state. You have to trust their self-reports, and there will always be that nagging worry that they're just lying to be nice to you.

So imposter syndrome sometimes isn't a belief at all, it's an alief. You may know that ghosts don't exist, but that won't stop you from being scared in a dark creaky house. This fear may become strong enough to compel you to leave, even if the whole time you're 100% confident there's nothing in the house that will harm you. Similarly, imposter syndrome is not an explicit beliefs about your skills, but rather a stubborn alief that you can't get rid of just by knowing it's not true. You may believe that you're competent and take responsibilities that reflect that, but still have a nagging mental itch telling you "you're not good enough" that you have to constantly try to ignore.

How to overcome imposter syndrome is a deep topic that I couldn't possibly do justice to here. The extremely short version is: Try to decouple your self-worth from your performance. Everybody is worse at some things than other people, and self-acceptance doesn't need to be linked to your accomplishments. Talk to your friends and mentors about imposter syndrome, and get their thoughts and advice. Ask them to tell you when they think your imposter syndrome may be holding you back. Consider seeing a therapist. If seeing one in-person is prohibitive, there are online therapy practices which can be much cheaper and more accessible. A few jumping off points for further reading: The STOPP technique, Overcoming Low Self-Esteem, Overcoming Perfectionism.

As community members, how can we help our fellow judges overcome their imposter syndrome? There are two methods that fit in particularly well with the judge program's ethos.

Firstly, we can help them see that other people have the same sort of failings, and they're not unique. Be public about your mistakes, explain your doubts to other people, tell stories about a time you did something stupid. Shatter the illusion of perfection that can exist around experienced judges, and help people see that everyone makes mistakes and learns from them.

Secondly, be honest. While it can be temping to blindly reassure someone with imposter syndrome when express self-doubt, they're not stupid, and they'll realize if you're lying to them. In the radical candor model of feedback, this is known as ruinous empathy. It makes you feel good about yourself in the moment, but sets them up for failure and distrust in the future. We don't want people worrying that they’re being silently judged for some error they made, and refusing to tell them about those errors will lead to exactly that. When you're consistently honest with people, they'll trust you when you tell them they did something well, and that positive feedback can be deeply validating.“

Disguised confidence miscalibrations

"A friend told me I have imposter syndrome, but I know that's not true because I'm actually an imposter."

"The people pointing out my mistakes are exacerbating my imposter syndrome; I should ignore them and focus on being more confident."

These people are both committing the same error. Epistemic underconfidence and overconfidence are biases that affect our perception of our own skills. This means that, unsurprisingly, we shouldn't entirely trust our perception of our own skills. By their nature, these biases are hard to notice; your unconscious mind is trying to hide them from you. And a convenient defense mechanism is for one bias to masquerade as the other one. If you're underconfident, believing you're overconfident can be one more thing to beat yourself up over. If you're overconfident, believing you're underconfident can be a convenient justification to act even more confident.

People loved her
almost as much as she disliked herself
which made her wonder
if she was good at hiding her bad
or bad at finding her good.

— A Small Fiction (@ASmallFiction) June 25, 2019

So if you think you're epistemically underconfident, first consider whether you might actually have sub-par skills.

And if you think you're epistemically overconfident, first consider whether your confidence may actually be justified.

Then stop just considering it, and take actions to figure out which is really the case.

Addressing epistemic confidence issues

Both epistemic overconfidence and underconfidence are harmful, and we don't need to pick only one of them to address. They're both a deviation from accurate self-evaluation, and accuracy is all we need to strive for. The exact same techniques to figure out if you're overconfident also work to figure out if you're underconfident. Teach people those techniques and practice them yourself, and they'll guide you to the correct place. The goal is not to be more or less confident; the goal is to be correctly confident.

What are these techniques? Well, that's a hugely complicated field, and millions of words have been written about it. The short version is that you try to find ways to test the accuracy of your confidence against reality. People selectively forget things all the time. If you're underconfident, you'll forget the things you did well. If you're overconfident, you'll forget the things you did poorly. You can't trust your brain to accurately report on the flaws in itself, so investigate and find out for sure.

Try new things. If there's no serious consequence for failure, just try it and see how it goes. You can never be certain in advance how something is going to go, so if you're never failing at anything, you're not taking enough chances. Are you working on L2 but are unsure if you're ready yet? Just take a practice test and see.

When there are serious consequences for failure, take things a little more slowly, but don't let it prevent you from exploring entirely. If you can't smoothly run a 30 player tournament, you're not ready for an 80 player one yet, but that doesn't mean you never will be. Focus on the 30 player events, and once they're consistently going fine, try one that's a bit larger, even if you don't feel quite ready. If it goes well, great! Keep going. If it goes poorly, take a step back, figure out what you need to improve, and work on improving it. Always be looking for opportunities to try new things, and try the new thing that's closest to a thing you've already done.

In contexts where you can't gradually make a transition and have to cross a big gap all at once (like going from Regular REL to Competitive REL), if you feel like you're not ready for the next thing, trust yourself, you're probably right. Don't use this as an excuse to give up; make a plan to study up and practice the necessary skills to get yourself to the point where you do feel ready. If you've done a bunch of studying and still don't feel ready, now consider that you might be wrong, and maybe try the thing anyway. If other people are telling you that you are ready, trust them over yourself, and definitely try the thing.

Listening to other people is one of the most effective ways to find out how you're really doing. Seek out the judges who have been doing this for a long time, who seem good at evaluating themselves and other people, and who are willing to provide honest feedback when asked. Ask them how you're doing. If they say you're doing a great job, believe them! If they say you're doing a poor job, believe them!

You're also perfectly capable of investigating things yourself. Want to know if you have imposter syndrome? Take an imposter syndrome test. Worried that your policy knowledge isn't good enough? Test it and see. Feel like you're worse or better than your peers? Take a practice test and compare your result against theirs.

Good judges learn to incorporate these sorts of self-assessments into their normal judging style, and are constantly integrating this information into their self-image.

Addressing social confidence issues

Improving social confidence is a simpler task, since the goal is just to make it as high as possible.

The one big pitfall here is that it's tempting to attempt to improve your social confidence via self-deception. All sorts of self-help guides will tell you to attempt to brainwash yourself into thinking you're competent, successful, etc.This works due to the illusory truth effect. Trying to fool yourself into thinking you're a better judge than you are can be an effective way to improve your social confidence, but it wreaks havoc with your epistemic confidence, and it will damage your judge career in the long run.

So what's the good way to improve social confidence? Social confidence is ultimately about being comfortable with your epistemic confidence and your place in the community. Someone who feel embarrassed about how much they don't know is going to have lower social confidence. High social confidence stems from the ability to say things like "I'm only 60% sure of this" or "I really have no idea, Bob would be a better person to ask" without feeling like this will lower your social status in the eyes of the people you care about. When you know yourself well enough to know what you do and don't know, and you don't feel pressured to hide your ignorance, you don't need to worry as much that you're going to say something incorrect, and you can more easily speak as though you're worth listening to.For more on improving your social confidence without it damaging your reliability, see The Scout Mindset.

The most impactful way to improve is just to get practice. Everyone is nervous at their first big tournament, but once you're at your tenth or 15th, you'll be a lot more comfortable with it. Conferences and workshops can also give you a safe place to work on these skills in a more direct fashion. Get used to what judging is like, and over time you'll find your place in the judge program and become more self-assured.


Social and epistemic confidence each have a strong effect on our ability to be effective judges, and we need to handle them in different ways. Social confidence is necessary in order to garner respect from players and keep the tournament running, but it's easy to accidentally use for evil. We don't want to have high social confidence, while being careful not to use it to convince people of things we're not actually all that sure of. In service of that, we should work to present ourselves clearly and self-assuredly, while being up front about what we do and don't know. Epistemic overconfidence and underconfidence are detrimental to our development as judges and make it harder to fulfill our responsibilities to players, so we should strive for accuracy and focus on objective metrics to counteract these tendencies.

The community plays a huge part in each judge's comfort level, which can in turn improve their social confidence and reduce their incentive to act more knowledgeable than they are. If another judge is struggling with some aspect of their confidence, we should offer to have an honest conversation about how we can best help them. Roleplay workshops can be a huge benefit to judges with low social confidence, as can regular practice at events.

I'm glad to be part of a community that recognizes these challenges and looks for ways to help us each with our individual struggles.