On Self-Appraisal (How To Become Great)

This forum is for discussion related to the game.
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

On Self-Appraisal (How To Become Great)

Post Post #0 (isolation #0) » Sun Mar 25, 2018 12:10 pm

Post by Mathdino »

On Self-Appraisal


aka "Everyone On This Site Needs A Goddamn Reality Check"


I'm not currently a Great Player.

I can carry games in the Newbie Queue and the Open Queue, and I'd like to think I'm pretty good at reading newbies/lynchbait players. Part of why I don't play in the Newbie Queue that much is because I don't think me using newbie tells on everyone is representative of the rest of the site.

But outside of those queues, if you, for example, told me to get a solid confident read on xRECKONERx, or RadiantCowbells, I'd look at you like "Are you crazy?" Same goes for most 2016/17 players.

Playing Town Is Hard

Playing town in forum mafia is a game of probabilities (credit to mastina). Your job is to process a MASSIVE amount of input, weigh the evidence, and come out with people more likely to be town and more likely to be scum. Charisma and getting lynches is a different skill. The first step is scumreading scum and townreading town.

But how often do you do that? Really. Actually. Go back through your game history on, say, D1s, and check your top scumreads and townreads. How often are they correct? Here's 3 games of my own:

Spoiler: My Game History
Micro 764: Marked For Death 2
: I was, at different points in D1, scumreading Not_Mafia (off policy), chesskid's slot, and TheGoldenParadox. 2/3 of these were town. I was townreading northsidegal, JaydragonKing, and Gamma Emerald, all correct.

Open 704: Switch
: Weird circumstance because I had a N0 cop guilty on Bins. Decided she was scum with Raya36 and havingfitz, both town. My top townreads were Almost50 and Lalendra, both not mafia (A50 was SK playing for town), with Elmo, mutantdevle, and Maxous as my next townleans. Elmo was scum.

Open 707: JK9++
: My D1 scumreads were RedFlavor, JaydragonKing, Assemblerotws, and Creature. Creature was scum, so that doesn't count, so we'll say RedFlavor scum means one of those was correct. My townreads were momo, Almost50, Hawk, and UnaBombaH, all correct.

These were strange circumstances at times, but we'll run with it.
3/8 correct scumreads (not counting cop guilty or Creature) = 37.5%
11/12 correct townreads = 92%

Completely random D1 reads gives 25% accurate scumreads and 75% accurate townreads. I did better than that. Should I be totally certain of my scumreads on D1? Of course not, I've been wrong before and have mislynched town on D1 a bunch. But what if everyone were 37.5% accurate?

I wrote a program that plurality lynches the town's top scumread on D1 with mountainous 10:3 (super scumsided setup). Scum's reads are random. My mode algorithm is slightly scumsided, so the program is about 1% off. Anyway.

If town is 37.5% accurate, scum is lynched D1 40% of the time.
That's VERY good.

If town is random (25% accurate), scum is lynched only 21% of the time.
Why? Scum won't lynch themselves, so they're likely to be wrong!

Feel free to input your own accuracy rates, or clone the program and change the player count and scumteam size.

The point is, while 37.5% accurate isn't the best thing ever, if we all followed our 37.5% accurate leads, town would easily win most games.

If your own reads are around random,
that's okay
. Mafia is hard, scum is good on this site, and forum mafia lends itself to bad reads.
But you need to recognise when they are essentially random.
What can you do instead?

Play To Your Strengths

When I started out, my scumhunting was godawful (like all 3 scumreads are town). Over time, I realised that what I
can
do is townhunt players with a >90% accuracy rate. I'm a towntell reliant player; in a 13p game there are almost always 3 or more people that bleed town.

What if you can lock 3 people as town and be confident about it?


If you can get to a 92% accuracy rate for your strong townreads, there is a 78% chance you're right about all of them. And if you can make sure town never lynches any of them?

The open setup Friends and Enemies with 3 masons vs 3 mafia is roughly balanced. So if you can make damn sure town doesn't lynch your 3 townreads, you've turned a 10:3 mountainous game into a balanced setup, 78% of the time.
That's the power of PoE.

Disclaimer: Of course, your townreads will probably also get shot, so it's not quite like masons, but you get the point.

There are a lot of strengths you can play to:
  • Better at townhunting? Lynch anyone outside your 3 townreads and you'll lynch scum 33% of the time in minis.

  • Recognise a player in game you know you can read? Sort just those players. Personal experience with players is invaluable; learn how to meta well and how not to meta badly.

  • Does a player look easily readable? Read through their previous games, and try to predict their alignments. If they're always town when you townread them, you can townhunt them. If they're always scum when you scumread them, you can scumhunt then.

  • Got charisma but bad reads? Convince the rest of the playerlist to play to THEIR strengths. Sort people by using everyone else.
Townhunting and confident meta-reads do a crazy amount of work for you. If EVERYONE could accurately sort 3 or more players with a high confidence, scum would almost always get lynched D1 or D2 in every game.

Maybe you've done all you can, but you still can't solve enough.

Know Your Weaknesses

I can name dozens of players that I can't D1 read for shit. When you start out onsite, almost every player will be near unreadable. But consider this scenario:
  • You're in a game, and Almost50 claims "northsidegal is definitely town this game". He has good reasons. You can't read Almost50 or northsidegal. But you notice that Almost50 has 90% accuracy at reading NSG when he's town.
This is a team game. Trust your fellow players' strengths.
The most probable scenario is that Almost50, if he's town, is correct. This means either:
A. They are scum together.
B. Almost50 is scum who knows NSG is town.
C. Almost50 is town who is 95% likely to be right on NSG.

Can they be scum together? Remember, it's hard for scum to confidently defend their buddies; they know all their reasoning is wrong! Ask around, see if others can rule out A50/NSG.

...so you found A50/NSG isn't the scumteam? Great.
You should be townreading her with 90% confidence.


Sheeping players who you trust when they're town will do a lot to boost your accuracy. You just have to find out whether the player is lying scum (scum don't always lie, remember).

And when a player flips town, now it's a game of "how likely is it this dead townie was right?"

Here's another scenario:
  • You're in a game, and Mathdino claims "I can read JaydragonKing with 90% accuracy, and he's scum here". But you found a towntell that makes JaydragonKing town with 90% confidence as well!
If and ONLY IF you have good self-appraisal skills
, you might be able to say "there's only a 10% chance I'm wrong here". If Mathdino is town, there's a 18% chance you'll ever disagree on a 90% confident read. Not. Good.

Burden of Proficiency is a valid tell. If you KNOW you can read Jay, and you KNOW Mathdino should be able to read Jay, it's HIGHLY likely in a Bayesian way that Mathdino is lying scum. The correct move here is to lynch Mathdino.
Disclaimer: Remember to back up other players' skill! If I claim confidence in a read on someone that I am bad at reading, this shouldn't tell you much at all other than "Mathdino is overconfident". See if their reasons are any good.

Regardless, a great way to get solid reads on extremely good players is to find out how likely it is that they could be wrong. I've lynched otherwise incredible scum players just by pinpointing where they should know better.

Always Room For Improvement, For EVERYONE

Self-appraisal is an ongoing process. There are still tons of players I still can't read, and the players that I CAN read will evolve their scumgames to get townread.

After every game, figure out why you were wrong and why others were right.
That is how to get better as town.

Spoiler: My Self-Appraisal (For example)
Micro 764: Marked For Death 2
: I shouldn't have lynched TheGoldenParadox when my scumread (chesskid) replaced out and started pushing him. And when Paradox was mislynched, I should've trusted that slot to be right on Paradox as town, and realised they were lying scum.

Open 704: Switch
: I shouldn't have assumed Elmo was town just because Bins tunneled her. Bins is a known distancer/busser, which others figured out. I also shouldn't have assumed Raya36 was scum by associations when she was really just playing her towngame.

Open 707: JK9++
: I could've caught scum-Creature sooner, and I should've stuck to my RedFlavor scumread going into D2 even after he hammered his buddy. I should've realised that I was scumreading Assemblerotws and JaydragonKing for playstyle.

Maybe you've used a tell for years and realised it went horribly wrong in your last game. The site evolves, tells change. We've had an influx of good scum players in the past year or two. Learn new scumhunting tactics. Test your old ones. If a tell works with an accuracy equal to random chance, it's not a good tell.

Tells are about what ACTUALLY works, not what SHOULD work. Armchair scumhunting won't catch scum. Learn from every game what town ACTUALLY does, and what scum ACTUALLY does.

Honestly, this kind of Bayesian reasoning can apply to anything. Identify your strengths and how confident you can be in them. Identify your weaknesses and when you shouldn't be confident in something. Identify the strengths and weaknesses of others, compare, and teamwork becomes easy.

And always, always, improve, staying one step ahead of the rest. That's how to become a Great Player.

Mathdino out.
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #1 (isolation #1) » Sun Mar 25, 2018 12:59 pm

Post by Mathdino »

Some fun results from the program:

13 players, 3 scum
Town AccuracyLynch Accuracy
0%0.55%
10%5.5%
20%15%
30%28%
34%34%
40%45%
43%50%
50%61%
60%76%
70%88%
80%96%
90%99%
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #4 (isolation #2) » Sun Mar 25, 2018 2:52 pm

Post by Mathdino »

I'm still learning that myself. I'm not the best player when it comes to metadiving.

Personally, the primary way I use meta is basically:

1. I observe a behaviour ingame that I think could be a scumtell.

2. I get paranoid, say to myself "I've seen town do that before..."

3. I go check up on their towngames. If they display the same behaviour, it's a nulltell (NAI). If I don't see the same behaviour, I become slightly more confident in my scumread.

4. If I wanna go further and I'm basically hopeless at reading a player otherwise, I can run stats on a player's towngames vs scumgames.
For example, I found an argument that I recognised from a player's previous scumgames. But they also used it in a couple towngames.
But they used it in a higher portion of their scumgames than their towngames over a large amount of meta. So even though it isn't DEFINITIVE, it can still be a scumtell in a Bayesian way.

The above is basically the exact same for players I'm townreading, just swap "town" and "scum", i.e. I check a player's scumgames to see what kind of towntells they're capable of faking.

Here's a fantastic video on Bayesian thinking:



Anyway, point is, I mostly use meta to check work and nullify reads (or make sure my reads are okay). Every so often I pick up on behaviours that are more common to a player's scumgames than towngames, and that can be a huge advantage in a game (especially smaller games).

Nothing can beat having actually played with the person before and understanding how they think. If they play pro-scum (as in they don't just do whatever they would do as town), knowing the player's thought process from experience helps a lot in getting a feel for the genuineness of their arguments.

And here's a mastina article.
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #6 (isolation #3) » Sun Mar 25, 2018 3:13 pm

Post by Mathdino »

Meta can also be used to figure out how good someone actually is. Townies can and will lie about their confidence or their own ability just to get their lynch to go through. It's all in the game, and doing that when they have reads that are better than random does help weaker scumhunters get on board.

But if you're gonna stay one step ahead of everyone, nothing wrong with a background check!
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #9 (isolation #4) » Sun Mar 25, 2018 4:54 pm

Post by Mathdino »

Sure. Here's a thing I wrote in another thread:
In post 33, Mathdino wrote:Oh yeah charisma always helps for sure. TSQ and Thor already talked about that, and working on charisma is a great way to improve both games. People are more often lynched/not lynched based on that, regardless of their alignment.

For charisma, I recommend (as general advice, not actually based on your play):
- emulating players you find to be charismatic
- avoid nervousness or too much public self-awareness and self-consciousness
- when talking TO people, always act like they're town. if you talk to people like they're scum, it looks like tunneling (no one likes that), doesn't help you if the are town, and scum will just act more like town anyway since they know you're scrutinising them.
- when talking ABOUT people, make insightful but not 100%-sure observations. "I" statements are pretty charismatic even in real life. "I believe this is scum chainsaw defending" rather than "This is scum chainsaw defending, guys!" People LOVE it when you make confident but willing-to-be-wrong comments.
- lastly, just do a search on the wiki for articles that mention charisma. Check the Mastin Academy thread, I'm pretty sure she's done a few bits on charisma and how to work with others.
mastina articles (both 4-5 years out of date):
Mafia As A Social Game: Argument About Charisma
On Balancing Logic And Charisma

The first article was a major factor in improving my playstyle years ago I think. I think a lot of points were made in it that definitely show the article's age (especially points about how site meta has moved away from arrogance, strongarming, and AtE), but a lot of it still apply. Arrogance/strongarming/AtE/other such uncharismatic tactics can either go very well or very wrong. It's hard to play badly following all the tips of playstyle articles even 5 years out of date.
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #11 (isolation #5) » Sun Mar 25, 2018 6:34 pm

Post by Mathdino »

Maybe!

I know that people acting that way toward me actually biases me against them. Bias is obviously not something to be proud of (it lost a micro recently), but I'm living evidence that it definitely hurts people's credibility with some players.

That said, it became a thing in the past couple years because people were responsive to it and it started working, for better or for worse.

The real key when it comes to charisma is knowing your audience, and figuring out what will and won't work in dealing with players.

And to shamelessly plug the original point, I think if players were better at self-appraisal and appraisal of others, those kinds of arguments wouldn't happen as much! :P

A lot of disagreements on reads can be settled by arguments of "If I'm town, I have a x% success rate in reading y", or "I trust you to read y, but I don't trust myself to read you better than random". Confidence is good, but so is openness on what you just don't know in a game.
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #15 (isolation #6) » Sun Mar 25, 2018 8:26 pm

Post by Mathdino »

After games finish, find out what you did right and what you didn't! If your hard townreads were all right, that's a great starting point. If you noticed that you've correctly read the same player a bunch of times, also great. And once a player begins to find tells and scumhunting techniques that work even a little bit better than random, that's exactly the point where they graduate from "sheep", "newbie", or "VI" to "scumhunter" and "content producer" in their own right.
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #24 (isolation #7) » Mon Mar 26, 2018 5:18 pm

Post by Mathdino »

In post 21, TheGoldenParadox wrote:I think you're a great player.
Haha, I appreciate that.

At the same time, saying "I'm a great player, follow my reads" in game to others, is very different from saying "I'm a great player, my reads are >50% accurate" to myself. A little known secret is that, as town, I literally expect every single flip to be town, no matter how hard I'm scumreading the slot (unless it's a hard guilty or the player has claimed scum). A lot of confidence is great for charisma, but bad for scumhunting. I assume the worst case scenario (and that I'm wrong) until proven otherwise.

So I guess what I'm saying is, I may come off that way, and I'm glad that I do, but I'm still improving.
In post 22, mastina wrote:
For the record--I have been, slowly, trying to revamp my older articles to be...well, less shitty, and have further reflections and refinement. Those are two I haven't gotten around to quite yet, unfortunately, but they are on the to-do list.

One mafia article I've been working on is actually one which works in tandem with articles about charisma, though--
I am working on writing an article (probably short enough that it'll be in the academy rather than a devoted thread) behind my philosophy of "Push Hard".

This has been my playstyle ever since switching to my mastina account and is the main difference between my play from my mastin2 days. A preview of the article which works as a bit of a fairly succinct summary--as long as you possess the ability to reevaluate, the best way to generate useful content is to push players hard. Most of what you push will be wrong, but by having made the push, it is easier to expose what is right.
In post 23, Mulch wrote:I disagree about pushing hard. That's the last thing mafia in general needs, and let alone mafiascum. You have to understand that every action has an advantage and a disadvantage. Pushing people can ilicit genuine reactions, but it creates a ton of negative things for the game state as well.
I did actually notice those articles seem fairly unaligned with your current playstyle.

FWIW, I think following the advice of those earlier articles for my own playstyle has helped produce a much healthier gamestate FMPOV, and I do endorse those articles as they are.

There's an argument to be made that chaos is inherently anti-town because while you personally might be better at reading people off it, it makes scumhunting and getting lynches that much harder for the rest of the team.

But that's an argument for another thread! :P
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #26 (isolation #8) » Mon Mar 26, 2018 5:28 pm

Post by Mathdino »

The problem isn't in scumhunting, the problem there is political. The harder you push something that was wrong, the more confidence town loses in you, and the more you fall victim to Burden of Proficiency.

Regardless, I patiently await the article so we can discuss it with more depth!
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #47 (isolation #9) » Thu Mar 29, 2018 12:14 pm

Post by Mathdino »

standard size for micros, minis, and larges is designed such that 25% of all players other than you are scum, and 75% of players other than you are town
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #50 (isolation #10) » Thu Mar 29, 2018 12:20 pm

Post by Mathdino »

In post 30, mutantdevle wrote:Tbh, I'd probably win more if I just didn't have reads. If you were to compare random to lynching to only lynching my scum reads then random lynching would be more successful :/
Don't sweat it! You're still improving.

One of your greatest strengths that I've seen in your towngames is your ability to utterly spew town in half your posts, removing yourself from the lynchpool. I probably should've caught that you were scum in Open 714 (Tit for Tat) sooner just by the fact that I wasn't hard-townreading you.

Early scumreads are basically always a shot in the dark unless you know the players around you really well (or the scumteam is bad). What I would work on is figuring out why your townreads were wrong. What is scum capable of faking on this site? What kind of logic/progressions are actually deserving of townreads?

I have a few answers, but as always, publishing tells all over the place is a great way to invalidate those tells. You pick those kinds of things up over time, whether by yourself, or by listening to the logic of other players.

So in the meantime, recognition that your early reads are often flawed is a great foundation for
- Avoiding overconfidence
- Working on what your strengths CAN be in early games
Try sticking to reading the players you know, or trying to rule out scumteams so you know when to sheep other players (see the Know Your Weaknesses section).
In post 41, GuiltyLion wrote:yo MD this article is great and overlaps a lot with the way I try to play the game, even put some ideas into words that were kicking around somewhere in my head but that I hadn't considered so explicitly before. Thanks for writing it
Thanks!
In post 42, GuiltyLion wrote:also make sure to Bayesian analyze individual posts

like if you see a post that is absolutely super unfathomably unlikely to come from scum - such that Probability(Player made post [x] given that Player is scum) is ~0, then it doesn't really matter how much decent evidence you had before for them being scum, they're probably town. I have the highest success for accurate townreads when I find good towntell posts like this.

Same thing goes for scumreads if you find posts that are almost impossible to come from town. Unfortunately, those posts are more rare both because people are trying their hardest to avoid making them and because generally in your experience (and in each game), the volume of townposts >> volume of scumposts
hell yeah my dude

i might make a post that's full on solely about bayesian reasoning in mafia, i don't see that much in the MD
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #64 (isolation #11) » Sun Apr 01, 2018 6:36 am

Post by Mathdino »

In post 59, beeboy wrote:You dont need good reads to win games and I disagree with your first section where you say its the first skill one should learn.

In reality good reads are the hardest skill and you can win games without them.
My intent wasn't to argue that having good reads is the first skill to learn. I think in writing this article before "How to Read People" (something I probably won't write) I'm arguing that Self-Appraisal is a skill that should come first :P

My intent was to argue that scumreading scum and townreading town is the first step toward lynching scum and not lynching town ingame.

Personally, I developed towntelling and charisma long before I developed good reads. Not getting lynched is great and all, but one more Innocent Child with so-so reads just doesn't swing games in the current meta, in my experience.
Last edited by Mathdino on Sun Apr 01, 2018 6:39 am, edited 1 time in total.
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #70 (isolation #12) » Sun Apr 01, 2018 2:29 pm

Post by Mathdino »

In post 66, Ellibereth wrote:I mean I'd take someone consistently okay vs. someone who swings between bad and great anyday.

Good reads are a way to "win" the game, I think if most people want to improve they need to learn how to "not lose" games first.
Definitely, no question. For example, someone who's perfect 40% of the time but awful 60% of the time when they're town is not worth betting the game over. Because there's only a 75% chance that they're even town to begin with -- that comes out to a 30% probability that what they say is right, and I know from experience I can, in general, do better than that.

That said, the point is that "great" isn't necessarily in the quality of someone's reads, but in the player's ability to evaluate the quality of their own reads, and use that as a way to both not fuck up in game, and improve as a player out of the game. If you know what you don't know, you're halfway there.

I think I disagree with the pessimistic take that townies nowadays should just take a step back and let the known good players just determine the outcome of the game. For one, it's no fun and isn't really my view of what a good game of mafia looks like, and for two, it seems kind of defeatest for the player's future skill level. Players with bad reads can absolutely evolve to be town leaders with good reads. I think I improved as a player the absolute most in 2014 after pushing mislynches and realising where my reads went wrong.

But you have to have those realisations! :P
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #74 (isolation #13) » Sun Apr 08, 2018 3:21 pm

Post by Mathdino »

In post 71, profii wrote:It’s interesting that this thread has largely focussed on scum reading and ignores the skill of identifiying potential PRs: A) whilst playing scum and B) whilst tying to protect them as town
That's actually one of my best skills!

I have a draft saved for an article about doing just that: PR/VT reading.
In post 73, Ranmaru wrote:This is a good thread. I generally approve of self reflection. I think it would be interesting to review your own recent games, with someone who is equal to or beyond your skill level. You go over your games together, and comb over your strengths and weaknesses. Go over bad habits you should break, and brainstorm how you can improve your weak areas and boost your strong areas together. It could even be a mutual review if both players have close skill to each other. Basically, like a hyperbolic time chamber, or deep training as I would use for super smash bros or any fighting game. (Three people take turns 1v1'ing each other, the person out of the current rotation takes notes of the play they have noted, and after the set, they reflect on the matches and then rotate)
This is a really great idea!
I don't really have anything to add to this point other than that I definitely approve :lol:
Go teamwork :D
User avatar
Mathdino
Mathdino
Survivor
User avatar
User avatar
Mathdino
Survivor
Survivor
Posts: 14337
Joined: February 24, 2013
Location: Right Behind You

Post Post #76 (isolation #14) » Sun Apr 08, 2018 3:39 pm

Post by Mathdino »

I can't say I'm in a place to give tons of accuracy advice while I'm still working on various tenets of my own reads, but there are lots more aspects to townplay than reading ability, as beeboy pointed out :P

hmu on Discord sometime

Edit: probably not in the super near future because i'm trying to get games off the ground on modside but definitely in a bit
Post Reply

Return to “Mafia Discussion”