|
On January 30 2022 11:45 bela.mervado wrote:oh hai thanks for the problem (while i let the whisky settle a bit) it is aktchually 1.17% (probability for the max EPT sum group to have no less than 10056 points, uniform draws per tiers). there are 3 of those scores, this current 10056, 10072 (Neeb+Dream) and the max 10122 (Nakajin: confirm, all these has Maru) i have assigned the first 4 players a group id (it is just a labeling for an extra * 4! which we dont care about ) the rest can be computed, we only have (4!)^4 = 331776 cases. these 3 scores each account for (3!)^4=1296 cases, so we have 3*1296/331776. sorry for my gnuplot skills ![[image loading]](https://i.imgur.com/DxcCDZk.png)
I think you're getting a different answer because you're asking a slightly different question, tbh.
Edit: Actually, you're right. I figured out that I had slightly overestimated the difference in averages as 297.33 instead of 297.13, and there's about 0.3%-0.4% of probability mass between those two.
|
I don't get what people complain about tbh.
As far as I can see, the groups got drawn in packs of four, right? 1 player from Top 4, 1 player from Top 8/12/16/20.
Yes, Group D is a group of death, but that just happens. There is always some "weirdness" happening with draws, but that doesn't mean it is rigged.
I also don't get how people think Serral "always has easy draws". Last year at Kattowice he had Dark and Innovation. Bad draws like that happen...if you want to avoid that, you probably had to put in some kind of seeding.
Just for the lolz - lets do a seeding, where we start in Group A with 1st and just go right:
Group A: Maru - Zest - Reynor - SpeCial - Dream
Group B: Rogue - Cure - Heromarine - Bunny - Zoun
Group C: Trap - Dark -Solar - Neeb - Showtime
Group D: Serral - Clem - TIME - Scarlett - Lambo
(yes, I know, the flaw in this system is that the "weakest" player (Lambo) gets seeded into the 4th players group...it gets balanced out after you seed the PlayIn Top 4 into it)
Is it just me or does Marus group still look really strong while Serrals probably got a bit easier? And we still have a group with four koreans, while we also got one with none.
And this is technically a "fair" seeding without any randomness.
|
On January 30 2022 12:02 Balnazza wrote:
I don't get what people complain about tbh.
As far as I can see, the groups got drawn in packs of four, right? 1 player from Top 4, 1 player from Top 8/12/16/20.
Yes, Group D is a group of death, but that just happens. There is always some "weirdness" happening with draws, but that doesn't mean it is rigged.
I also don't get how people think Serral "always has easy draws". Last year at Kattowice he had Dark and Innovation. Bad draws like that happen...if you want to avoid that, you probably had to put in some kind of seeding.
Just for the lolz - lets do a seeding, where we start in Group A with 1st and just go right:
Group A: Maru - Zest - Reynor - SpeCial - Dream
Group B: Rogue - Cure - Heromarine - Bunny - Zoun
Group C: Trap - Dark -Solar - Neeb - Showtime
Group D: Serral - Clem - TIME - Scarlett - Lambo
(yes, I know, the flaw in this system is that the "weakest" player (Lambo) gets seeded into the 4th players group...it gets balanced out after you seed the PlayIn Top 4 into it)
Is it just me or does Marus group still look really strong while Serrals probably got a bit easier? And we still have a group with four koreans, while we also got one with none.
And this is technically a "fair" seeding without any randomness.
Pls check the actual statistics on the group draws that Bela and myself did. You seem to be completely ignoring the points and going by your gut on what you think looks stronger or weaker. Also, I think a lot of people would disagree with the idea that a group with Clem is easier for Serral.
|
On January 30 2022 12:02 Balnazza wrote:
I don't get what people complain about tbh.
As far as I can see, the groups got drawn in packs of four, right? 1 player from Top 4, 1 player from Top 8/12/16/20.
Yes, Group D is a group of death, but that just happens. There is always some "weirdness" happening with draws, but that doesn't mean it is rigged.
I also don't get how people think Serral "always has easy draws". Last year at Kattowice he had Dark and Innovation. Bad draws like that happen...if you want to avoid that, you probably had to put in some kind of seeding.
Just for the lolz - lets do a seeding, where we start in Group A with 1st and just go right:
Group A: Maru - Zest - Reynor - SpeCial - Dream
Group B: Rogue - Cure - Heromarine - Bunny - Zoun
Group C: Trap - Dark -Solar - Neeb - Showtime
Group D: Serral - Clem - TIME - Scarlett - Lambo
(yes, I know, the flaw in this system is that the "weakest" player (Lambo) gets seeded into the 4th players group...it gets balanced out after you seed the PlayIn Top 4 into it)
Is it just me or does Marus group still look really strong while Serrals probably got a bit easier? And we still have a group with four koreans, while we also got one with none.
And this is technically a "fair" seeding without any randomness.
Yeah 100%.
|
either way,
I would love to see a live group draw ceremony, it could be a good show.
maybe not as nice as GSL group selection, but there could be some casters, discussing the groups as they are forming,
they could use a real dice to draw players in turns.
a live show would take away most of the doubts about seeding.
they could do stupid things like make it look like a card or poker game (I know Rotti is into that thing) where players to be selected would represent cards in the casters' mini games, whatever.
|
I just think they should add some more sophistication to the way the group draws are done so these lopsided groups don't get formed. That way, it wouldn't matter.
|
They should just use GSL format and have designated groups based on EPT ranking so the second there are no EPT points left we know what the groups are and there's never a question like this. All tournaments should do that for playoffs as well. There should not be a group draw or break before the announcement. The bracket should be predetermined before the groups are played out at all.
The fact that in 11 years of SC2 GSL is one of the only events to do that is dumb.
|
On January 30 2022 12:16 honorablemacroterran wrote:Show nested quote +On January 30 2022 12:02 Balnazza wrote:
I don't get what people complain about tbh.
As far as I can see, the groups got drawn in packs of four, right? 1 player from Top 4, 1 player from Top 8/12/16/20.
Yes, Group D is a group of death, but that just happens. There is always some "weirdness" happening with draws, but that doesn't mean it is rigged.
I also don't get how people think Serral "always has easy draws". Last year at Kattowice he had Dark and Innovation. Bad draws like that happen...if you want to avoid that, you probably had to put in some kind of seeding.
Just for the lolz - lets do a seeding, where we start in Group A with 1st and just go right:
Group A: Maru - Zest - Reynor - SpeCial - Dream
Group B: Rogue - Cure - Heromarine - Bunny - Zoun
Group C: Trap - Dark -Solar - Neeb - Showtime
Group D: Serral - Clem - TIME - Scarlett - Lambo
(yes, I know, the flaw in this system is that the "weakest" player (Lambo) gets seeded into the 4th players group...it gets balanced out after you seed the PlayIn Top 4 into it)
Is it just me or does Marus group still look really strong while Serrals probably got a bit easier? And we still have a group with four koreans, while we also got one with none.
And this is technically a "fair" seeding without any randomness. Pls check the actual statistics on the group draws that Bela and myself did. You seem to be completely ignoring the points and going by your gut on what you think looks stronger or weaker. Also, I think a lot of people would disagree with the idea that a group with Clem is easier for Serral.
I will be honest: Math is just not my strong point. I tried to follow it, but it wasn't working for me - not saying you or Bela are wrong though. But the one thing I do know about statistic is, that it isn't absolute. The chance to win the jackpot is incredibly low - yet it still happens rather frequently. Weirder things have happened. I've seen Hearthstone RNG that went down in the 0,00X%...
I would disagree however that I "went by my gut". The groups I mentioned are based on EPT points, so basically groups without any draw, just pure seeding. It is not what I think is stronger or weaker, it is what the points dictate. And overall I don't think the groups look that different compared to the ones we actually got. Because in the end, the difference between the 5th player (Clem) and the 8th player (Zest) is not that big, even though they have 600 points between them. On paper, either one of them looks strong in a group.
Though you might be right, Clem is a harder opponent for Serral, but in the end, it doesn't really matter. Either way (actual draw vs. "my" seeding) Serral is in a group that he easily should come out of, while Maru is still in a group that could go bad for one of the big three.
I would agree however with some comments that I really don't get why SC2 draws are always done in private. For example, I really enjoy the group-draws before the LoL Worlds. But correct me if I'm wrong, but isn't that kind of an ESL thing to not show draws? I don't watch much Counterstrike, but I'm sure I've never seen a group draw in any shape or form in it.
|
I think the issue boils down to the following things:
+Among all tier 1 ranker, Trap is falling flat in recent months, so much that I think he would struggle to beat any tier 3 players. So anyone falling in Trap group are generally happy about it.
+All tier 2 players have pretty much the same impact, so it should not matter who goes where.
+Reynor is a borderline Tier 1 player, the fact that hes falling down to Tier 3 meaning that whoever getting him into the group would already be on the way to be the "Group of Death".
+The rest of the lower tiers are pretty much similar in term of potential and impact, but that also depending on the matchup
Having said that, the fact that they got pretty much the same group B from last year IEM make me raise my eyebrown, not because is an "easy" group or anything, but really the process behind it.
|
I like these groups because every race is represented and there are very good match ups coming for us 
On a side-note: i played "kings" last year. Its a game where u roll the dice two times. to get two times a six has a probability of 1/36. i got this 3 rounds after another! this has a probability of 1/46656. all my friends laughed and accused me of cheating, but i didnt.
I think probabilities are deceptive sometimes!
|
On January 30 2022 15:39 DarkGamer wrote:I like these groups because every race is represented and there are very good match ups coming for us On a side-note: i played "kings" last year. Its a game where u roll the dice two times. to get two times a six has a probability of 1/36. i got this 3 rounds after another! this has a probability of 1/46656. all my friends laughed and accused me of cheating, but i didnt. I think probabilities are deceptive sometimes!
The probability is 1/36 every time though. It's not really comparable to this situation. Sounds like you fell into what's called the "Gambler's Fallacy."
|
On January 30 2022 15:46 honorablemacroterran wrote:Show nested quote +On January 30 2022 15:39 DarkGamer wrote:I like these groups because every race is represented and there are very good match ups coming for us On a side-note: i played "kings" last year. Its a game where u roll the dice two times. to get two times a six has a probability of 1/36. i got this 3 rounds after another! this has a probability of 1/46656. all my friends laughed and accused me of cheating, but i didnt. I think probabilities are deceptive sometimes! The probability is 1/36 every time though. It's not really comparable to this situation. Sounds like you fell into what's called the "Gambler's Fallacy."
???
the probability to roll the dice after another becomes less likely:
first time: 1/36
two times in a row 1/36 x 1/36
three times in a row 1/36 x 1/36 x 1 /36
this is basic math in school and is right. of course its always 1/36 to throw two times a six, but its not to do it 3 times in a row!
|
This looks like an easy group for Maru. The only threat would be Zoun. Yeah you can say Clem and Reynor are good but Maru will stomp them anyway.
|
United States33699 Posts
herO has a pretty high chance of getting out of the Ro36, and he's just so disproportionately stronger than any of the other Ro36 players imo. whatever group gets him will be so unlucky—similar to ByuN in the 2021 tourney
|
On January 30 2022 16:36 parksonsc wrote:
This looks like an easy group for Maru. The only threat would be Zoun. Yeah you can say Clem and Reynor are good but Maru will stomp them anyway.
Not an easy bracket for Clem and Reynor, but unsurprisingly lucky bracket for Serral over the years.
|
They should put in a rule so that for the top 16 players, those of the same rank within each tier can't be drawn into the same group
|
group and bracket draws should always be done live and publicly availible via a stream..
this shouldn't be a hard thing to do at all.
|
On January 30 2022 08:50 deacon.frost wrote:
Two years in a row(and that's just what I remember) with this conspiracy. Is it really that hard to record the process and show the video there's no conspiracy involved?
exactly.. this should be a commonplace already in the world of e-sports.. it's kinda ridiculous that it isn't tbh
|
|
|
Czech Republic12133 Posts
On January 30 2022 14:24 Balnazza wrote:Show nested quote +On January 30 2022 12:16 honorablemacroterran wrote:On January 30 2022 12:02 Balnazza wrote:
I don't get what people complain about tbh.
As far as I can see, the groups got drawn in packs of four, right? 1 player from Top 4, 1 player from Top 8/12/16/20.
Yes, Group D is a group of death, but that just happens. There is always some "weirdness" happening with draws, but that doesn't mean it is rigged.
I also don't get how people think Serral "always has easy draws". Last year at Kattowice he had Dark and Innovation. Bad draws like that happen...if you want to avoid that, you probably had to put in some kind of seeding.
Just for the lolz - lets do a seeding, where we start in Group A with 1st and just go right:
Group A: Maru - Zest - Reynor - SpeCial - Dream
Group B: Rogue - Cure - Heromarine - Bunny - Zoun
Group C: Trap - Dark -Solar - Neeb - Showtime
Group D: Serral - Clem - TIME - Scarlett - Lambo
(yes, I know, the flaw in this system is that the "weakest" player (Lambo) gets seeded into the 4th players group...it gets balanced out after you seed the PlayIn Top 4 into it)
Is it just me or does Marus group still look really strong while Serrals probably got a bit easier? And we still have a group with four koreans, while we also got one with none.
And this is technically a "fair" seeding without any randomness. Pls check the actual statistics on the group draws that Bela and myself did. You seem to be completely ignoring the points and going by your gut on what you think looks stronger or weaker. Also, I think a lot of people would disagree with the idea that a group with Clem is easier for Serral. I will be honest: Math is just not my strong point. I tried to follow it, but it wasn't working for me - not saying you or Bela are wrong though. But the one thing I do know about statistic is, that it isn't absolute. The chance to win the jackpot is incredibly low - yet it still happens rather frequently. Weirder things have happened. I've seen Hearthstone RNG that went down in the 0,00X%... I would disagree however that I "went by my gut". The groups I mentioned are based on EPT points, so basically groups without any draw, just pure seeding. It is not what I think is stronger or weaker, it is what the points dictate. And overall I don't think the groups look that different compared to the ones we actually got. Because in the end, the difference between the 5th player (Clem) and the 8th player (Zest) is not that big, even though they have 600 points between them. On paper, either one of them looks strong in a group. Though you might be right, Clem is a harder opponent for Serral, but in the end, it doesn't really matter. Either way (actual draw vs. "my" seeding) Serral is in a group that he easily should come out of, while Maru is still in a group that could go bad for one of the big three. I would agree however with some comments that I really don't get why SC2 draws are always done in private. For example, I really enjoy the group-draws before the LoL Worlds. But correct me if I'm wrong, but isn't that kind of an ESL thing to not show draws? I don't watch much Counterstrike, but I'm sure I've never seen a group draw in any shape or form in it.
The fuck is this logic? You realise this isn't the first time this come into discussion? And this isn't a jackpot situation per se, because you keep ignoring how many times they have to do the drawing and how many people have to take a part to actually get the winner of the jackpot.
But in here we have a situation where people complain about the groups EVERY. DAMN. TIME. So while weirder things have happen, they don't tend to happen with this frequency.
The frequency is the thing that people question the most and you graciously ignore it
Edit> the most pathetic thing is that all it takes is 5 or 10 minutes of few people life on a camera and then presenting the video.
|
|
|
|
|
|
EPT
