Orca? This! is! Dance of the Title Fairy!

[ZONEACE]

NO one made a flaming homosexual joke? god, and i set it up so perfectly.

[Thok]

Scum win % is almost NEVER 50%. Are you purposefully being obtuse?

Have you even bothered collecting data for this? 50% is a good approximation.

Just for the record, 5 of those games are as an IC in Newbies in the C9 era, where I'd expect an IC scum to have something like a 80% winning percentage (so Elias was 5 for 5 as opposed to 4 for 5. Big deal.)

[Thestatusquo]

NO one made a flaming homosexual joke? god, and i set it up so perfectly.

<3


Anyway, thok, I have NEVER seen a title have a threshold of "statistically significant" set for it before. If you're really going to be that big of a stickler, then half the titles on the site would be gone. He wins an insanely high amount as scum, and a normal amount as town, and has done so for a long time. THAT is title worthy. YOU are being a douchebag.

[Thestatusquo]

Scum win % is almost NEVER 50%. Are you purposefully being obtuse?

Have you even bothered collecting data for this? 50% is a good approximation.

Just for the record, 5 of those games are as an IC in Newbies in the C9 era, where I'd expect an IC scum to have something like a 80% winning percentage (so Elias was 5 for 5 as opposed to 4 for 5. Big deal.)

No, because given that the balancing meta is that town have a 50% chance of winning weighted against ALL other factions, instead of each given faction having a equal win probability means that the actual probability of a given scum group winning is MUCH lower than 50%. In addition, each scum screw up is bigger than each town mistake, therefore it is harder to win as scum, and therefore the % is even lower. You're just being dumb now.

[KaleiÐoscøpe]

Eh... the game of mafia has a pretty large factor of luck involved into it.

[Thok]

No, because given that the balancing meta is that town have a 50% chance of winning weighted against ALL other factions, instead of each given faction having a equal win probability means that the actual probability of a given scum group winning is MUCH lower than 50%. In addition, each scum screw up is bigger than each town mistake, therefore it is harder to win as scum, and therefore the % is even lower. You're just being dumb now.

Except that Elias only plays Mini's, where we're typically talking one mafia group, or one mafia group and an SK.

Of all the times Elias has been mafia, he's had to deal with multiple scum groups twice: Mini 365, (which he lost) had an SK, and Mini 417 had a 6-3-3 distribution. So Elias probably only had a 35-40% chance of winning as opposed to a 50% chance of winning.

[Thestatusquo]

NO one made a flaming homosexual joke? god, and i set it up so perfectly.

<3


Anyway, thok, I have NEVER seen a title have a threshold of "statistically significant" set for it before. If you're really going to be that big of a stickler, then half the titles on the site would be gone. He wins an insanely high amount as scum, and a normal amount as town, and has done so for a long time. THAT is title worthy. YOU are being a douchebag.

[Thok]

Anyway, thok, I have NEVER seen a title have a threshold of "statistically significant" set for it before. If you're really going to be that big of a stickler, then half the titles on the site would be gone. He

wins an insanely high amount as scum

, and a normal amount as town, and has done so for a long time. THAT is title worthy. YOU are being a douchebag.

The bolded test is why statistical significance is necessary. We're trying to quantify if Elias's win percentage is really "insanely high". If it's not, then the title will seem silly when Elias has a cold streak as scum/hot streak as town.

I'm going to stop commenting about this until people actually suggest a title.

Edit: Actually, something like "TSQ says you're good scum" might work for Elias and works on enough levels that I wouldn't complain about it.

[JDodge]

If we assume scum winning percentage is 50%, there's about a 1/200 chance of having a scum record of 10-1 or 11-0 in their games (and really, since we'd have this discussion but backwards if the records were flipped, I should include the possibility of 1-10 or 0-11 and make is about 1/100.) That's good but we have a lot of people on this site, and I wouldn't be shocked if a person randomly was 10-1 or 11-0 or even better.

There's also the question of whether Elias's is actually good, or getting somewhat lucky. (Probably the answer is a combination of both.)

You'll have better luck finding some particularly cool scum gambit that Elias pulled off and basing a title off of that. It would likely give a more interesting title also.

I can't help it if people fail basic statistics.

You really need to stop undermining the point of the title. Whether or not luck has gotten him this far has nothing to do with it; the fact is, he wins when he's mafia, and to pass it off as if it wasn't true pins you as the current penultimate of douchebaggery.

does he get a cash prize

can we give him an all-expense paid vacation to lovely FRANCE, home of the douchebags

or seattle, that works just as well

[Adel]

nominate Elias for "not statistically significant"

better than a title that outs him as strong scum

[Thestatusquo]

nominate Elias for "not statistically significant"

better than a title that outs him as strong scum

I like that.

[Pie_is_good]

second "not statistically significant"

[Occult]

Third "not statistically significant"

[xyzzy]

Fourth "not statistically significant"

[Thok]

I'm OK with this if Elias is.

[Mr Stoofer]

I think "Not statistically significant" is a brilliant title for Elias.

Also, this bussing between TSQ and Thok is so blatant I have to

vote: TSQ

and if he comes up scum we lynch Thok tomorrow.

[mith]

For what it's worth, my thoughts on this:

11 games isn't statistically significant, given the expected scum win percentage in those games. (Thok is a mathematician, I can't believe you guys were actually trying to argue that with him.)

"Statistically significant" shouldn't be a factor in determining a good title.

Actually suggesting a title makes it more likely a title will be given.

I don't think the 10-1 thing makes for a good title. It has never been the point of titles to point out good/bad players, records, etc. That's what the Scummies are for.

This thread is

not

for insults and the like. I am once again growing tired of this. Maybe I'll just shut down the title fairy thread and go back to assigning them arbitrarily. It's more fun that way.

[Adel]

more evidence that a person's opinion can be more correct even if it is a minority position.

[gorckat]

Maybe I'll just shut down the title fairy thread and go back to assigning them arbitrarily. It's more fun that way.

You need to be modkilled for impairing the ability of the

rest

of us to enjoy this game

[Quagmire]

Maybe I'll just shut down the title fairy thread and go back to assigning them arbitrarily. It's more fun that way.

Personally, I think we should scrap the middle man altogether and just let people control their own titles after a certain number of game posts. Why do people not like this option again?

[shaft.ed]

Why do people not like this option again?

cocks

[Sethaniel]

Random opinion: Titles are more fun/meaningful if you earn them from others, rather than simply giving them to yourself.

[xyzzy]

Random opinion: Titles are more fun/meaningful if you earn them from others, rather than simply giving them to yourself.

[Glork]

Why do people not like this option again?

cocks

Well said.


Also, Quag, it just would turn "titles" into "five word sigs." We'd get ones that offend people, some would lobby for the option to turn them off, then they'd lose pretty much all significance whatsoever.

Having a title handed to you by

the rest of the community

was the backing, but if there are going to be multiple lengthy, stupid debates over what policies are "fair" and who should run them, I wouldn't be opposed to scrapping the nom/fairy policy altogether.

[Thestatusquo]

I want mine to be "tits and cooters"