Maru v Hurricane:
Maru wins 3-1
Inno v Trap:
Trap wins 3-1
Dream v Rogue:
Dream wins 3-2
Assuming that’s correct,
Maru v Trap:
Trap wins 3-2
Dream v Trap:
Trap wins 4-2
Maru v Hurricane:
Maru wins 3-1
Inno v Trap:
Trap wins 3-1
Dream v Rogue:
Dream wins 3-2
Assuming that’s correct,
Maru v Trap:
Trap wins 3-2
Dream v Trap:
Trap wins 4-2
I’m not sure about predictions this time, except Maru beating Hurricane, the rest are not easy to know the outcome. I would say that Inno should lose against Trap, that’s the common sense, but something tells me this GSL will bring some surprise, so maybe Inno suddenly beats trap, same with Dream vs Rogue, Dream is on fire this tournament, and should win vs the late Rogue, but somehow Rogue started playing well again, so that next match could be either easy win for Rogue or easy win for Dream.
Yeah normally I’d side with Rogue in that BO5 but I got a feeling with Dream. With Inno v Trap I want Trap to win just so we can avoid the additional TvTs
Rogue vs trap finals. Trap loses 4-0 just to annoy the big bad batz.
Only a Zerg player could write this.
It is always so hard to predict Rogue. Rogue in top shape is the best SC2 player of all time hands down. The problem is that Rogue in a bad day looks like a total scrub.
Agree with you until Dream. I think Rogue beats Dream and probably wins the GSL. If Dream beats Rogue, I think Dream is the favorite to win, hes been on fire but dont blame you for going with Trap.
PS: Not much confidence for Maru. Injuries keep causing him to underperform.
He got back into the game and he said his shoulder was better than before, don’t know if he was just resting for some months or had some kind of surgery. I’m pretty confident that he will deliver some high level games.
Game 2 of bunny vs dream was amazing
The rogue reverse sweep was great as well.
Really great stuff last night and I highly recommend last night’s stream.
All the stuff is viewable on AfreecaTV youtube channel
Maru > Hurricane
Trap > Innovation (unless inno got WAY better at TvP over the break)
Rogue > Dream
Trap > Maru
Rogue > Trap
GSL ro4 onward is bo7.
good point, my bad .
I don’t think he’s playing like a scrub though. He’s really freaking good and to me hands down the best korean Zerg in legacy. I think people might think its dark at 2019 but man the more i think about it the more I think Rogue is better.
As for predictions
Hurricane 0-3 Maru.
I don’t think Hurricane can take a map off Maru with how he played in the TvP vs bunny. Bunny was dominant and this is Maru.
Innovation 2-3 Trap.
Inno has been really good in the past. He still is but he’s kinda outgunned nowadays. I think trap takes this one
R04
Dream 1-4 Rogue
Rogue is the strongest in a b07 format. In ZvT he loves mutas. And if you check the way he played at IEM he looked really good in the TvZ. I don’t see Dream winning this.
Maru 4-3 Trap
I think Maru’s TvP is insane. Trap is no slouch but I think Maru is slightly favored to win this. Trap lost a couple of times against ByuN and PvT used to be Traps best MU now not anymore i think. But this is the most 50-50 match in the r04.
If Trap advances.
Rogue 4-2 Trap.
Rogue is disgustingly good vs Protoss in a b07. Check his history. I don’t think its a 4-1 or 4-0 this time around as PvZ is not extremely Zerg favored anymore. But yeah I’m favoring rogue to win.
If Maru advances:
This will produce an extremely tightly fought finals i think this is the finals i want. Could go 4-3 either way.
Pretty much this exactly. Not as much belief in Maru at this time though.
Maru is the most likely individual to win at a 46.31% chance. Terran is the most likely race to win (typical for Code S historically) at a 62.07% chance. These predictions are made from a monte carlo tournament simulation based on a statistical analysis of the player’s match history (Elo analysis). It is a sample of 1,000,000 possible tournament outcomes. If a player’s match history is accurate, there is a 37.94% chance of a Rogue vs Maru finals:
Bunny:
Round of 8: 100.00%.
Round of 4: 16.18%.
Round of 2: 5.69%.
Round of 1: 0.55%.Dream:
Round of 8: 100.00%.
Round of 4: 83.82%.
Round of 2: 21.06%.
Round of 1: 4.92%.Rogue:
Round of 8: 100.00%.
Round of 4: 67.27%.
Round of 2: 51.60%.
Round of 1: 27.12%.Zest:
Round of 8: 100.00%.
Round of 4: 32.73%.
Round of 2: 21.64%.
Round of 1: 5.97%.Hurricane:
Round of 8: 100.00%.
Round of 4: 1.00%.
Round of 2: 0.12%.
Round of 1: 0.01%.Maru:
Round of 8: 100.00%.
Round of 4: 99.00%.
Round of 2: 73.53%.
Round of 1: 46.31%.Innovation:
Round of 8: 100.00%.
Round of 4: 49.77%.
Round of 2: 17.22%.
Round of 1: 10.29%.Trap:
Round of 8: 100.00%.
Round of 4: 50.23%.
Round of 2: 9.13%.
Round of 1: 4.83%.
Terran:
Round of 8: 50.00%.
Round of 4: 62.19%.
Round of 2: 58.75%.
Round of 1: 62.07%.Protoss:
Round of 8: 37.50%.
Round of 4: 20.99%.
Round of 2: 15.45%.
Round of 1: 10.81%.Zerg:
Round of 8: 12.50%.
Round of 4: 16.82%.
Round of 2: 25.80%.
Round of 1: 27.12%.
Note: this was based on the round of 8 seeding, so it predicted a strong chance for Dream to advance over Bunny (83.82% vs 16.18%), which has already occurred and is confirmed.
LUL 0.01% to win? XD XD DD
Yep Hurricane has a 1% chance to beat Maru. Inno only has a 50% chance to beat Trap.
These numbers don’t look too inaccurate, honestly. I think it overestimates Dream’s chances against Rogue and Innovation’s chances against Trap though.
This is the average outcome if this tournament were played on repeat a million times. Elo is very accurate IF it has up to date match history information.
That being said, nobody is an expert on the future and predictions are only as good as your starting assumptions. It is physically impossible to have complete starting information so you must make assumptions. Furthermore, just because a scenario is unlikely doesn’t mean it won’t occur - unlikely things are happening all the time.
I did the same simulation for the last Blizzcon and it showed a high chance of Serral winning, low chance for Dark. Dark won. All thanks to the fact that Reynor squeaked out a victory over Serral in the round of 4.
I understand how probability works, you don’t need to explain it to me, thanks. To throw it back at you, the outcome could go exactly as your model predicts is most likely and my analysis could still be more accurate than your model’s.