Home | About | Tournament Winning Decklists | Forums

SHL2 Stats now available


#1

https://drive.google.com/open?id=0B-gMiPlH3rBAR2Y0Y044TTc2a0U&authuser=0

Also, if you’re interested, I’ve also provided the anonymized 400k stats: http://octgn.gamersjudgement.com/wordpress/anr/400k-anonymized-octgn-stats-now-available/


~300k OCTGN game stats
#2

NEH still kicking butt.

Corp Win rates in SHL2

NEH 67.91%
Blue sun: 55.14%
RP: 54.23%

not even close.


#3

Runner Vs. NEH;

Lela: 40.45%
Kate: 36.22%
Andy: 35.47%
Gabe: 33.90%


#4

67.91% – NBN | Broadcast Center – 635 wins 300 losses
55.14% – Weyland Consortium | Powering the Future – 343 wins 279 losses
54.23% – Jinteki | Replicating Perfection – 237 wins 200 losses
50.42% – Jinteki | Personal Evolution – 179 wins 176 losses
49.69% – Haas-Bioroid | Engineering the Future – 241 wins 244 losses
43.43% – NBN | The World Is Yours – 43 wins 56 losses
42.85% – NBN | Making News – 69 wins 92 losses
42.02% – Haas-Bioroid | Infinite Frontiers – 29 wins 40 losses
37.14% – Jinteki | The Next Generation – 13 wins 22 losses
33.76% – Jinteki | The Secrets Within – 26 wins 51 losses
31.81% – Haas-Bioroid | Refining the Process – 35 wins 75 losses
30.76% – Weyland Consortium | Building a Better World – 16 wins 36 losses
30.00% – Haas-Bioroid | Guarding the Net – 12 wins 28 losses
29.16% – Jinteki | Biomedical Pioneer – 7 wins 17 losses
18.75% – Weyland Consortium | Because We Built It – 3 wins 13 losses
14.28% – Weyland Consortium | Power Unleashed – 1 wins 6 losses
08.69% – Haas-Bioroid | Stronger Together – 2 wins 21 losses
05.55% – Haas-Bioroid | Engineered for Success – 1 wins 17 losses

53.33% – Shaper | The Professor – 8 wins 7 losses
52.66% – Criminal | Andromeda – 356 wins 320 losses
51.43% – Shaper | Kate McCaffrey – 395 wins 373 losses
50.00% – Criminal | Disappeared Clone – 47 wins 47 losses
48.98% – Criminal | Gabriel Santiago – 145 wins 151 losses
48.04% – Criminal | Trained Pragmatist – 172 wins 186 losses
47.82% – Anarch | Whizzard – 33 wins 36 losses
44.72% – Anarch | Noise – 212 wins 262 losses
43.78% – Anarch | Quetzal – 74 wins 95 losses
41.62% – Shaper | Kit – 77 wins 108 losses
40.77% – Anarch | Reina Roja – 42 wins 61 losses
37.83% – Shaper | Nasir Meidan – 28 wins 46 losses
36.66% – Criminal | Retired Spook – 11 wins 19 losses
32.14% – Shaper | Chaos Theory – 63 wins 133 losses
20.00% – Shaper | Exile – 3 wins 12 losses
16.27% – Criminal | Stealth Operative – 7 wins 36 losses

PROFESSOR KILLIN’ IT!


#5

Professor best Runner. we were all wrong

PE is exactly a coin flip. Might as well just pull out a coin at the beginning of the game =)


#6

Also, here’s what you get if you look at only the games played between players who both ended with 1500+ ELO. This is somewhat messy as challengeboards names sometimes don’t match octgn names, but even not trying to fix that, the dataset size drops from 3566 games to 1507 games. Most notably, NEH drops from 68% to 60% among the “top quarter” of games.

100.00% – Weyland Consortium | Power Unleashed – 1 wins 0 losses
59.95% – NBN | Broadcast Center – 283 wins 189 losses
53.65% – NBN | The World Is Yours – 22 wins 19 losses
53.33% – Weyland Consortium | Powering the Future – 160 wins 140 losses
53.33% – Weyland Consortium | Building a Better World – 8 wins 7 losses
53.33% – Jinteki | The Next Generation – 8 wins 7 losses
50.00% – Haas-Bioroid | Infinite Frontiers – 6 wins 6 losses
49.75% – Jinteki | Replicating Perfection – 103 wins 104 losses
47.02% – Haas-Bioroid | Engineering the Future – 95 wins 107 losses
45.63% – Jinteki | Personal Evolution – 47 wins 56 losses
42.85% – Haas-Bioroid | Guarding the Net – 3 wins 4 losses
39.28% – NBN | Making News – 22 wins 34 losses
34.78% – Jinteki | The Secrets Within – 8 wins 15 losses
30.95% – Haas-Bioroid | Refining the Process – 13 wins 29 losses
12.50% – Jinteki | Biomedical Pioneer – 1 wins 7 losses
00.00% – Weyland Consortium | Because We Built It – 0 wins 2 losses
00.00% – Haas-Bioroid | Stronger Together – 0 wins 2 losses

66.66% – Shaper | The Professor – 4 wins 2 losses
55.55% – Anarch | Whizzard – 15 wins 12 losses
52.43% – Criminal | Andromeda – 151 wins 137 losses
51.21% – Criminal | Disappeared Clone – 21 wins 20 losses
51.16% – Anarch | Reina Roja – 22 wins 21 losses
50.58% – Criminal | Trained Pragmatist – 87 wins 85 losses
50.00% – Criminal | Retired Spook – 3 wins 3 losses
49.42% – Shaper | Kate McCaffrey – 171 wins 175 losses
48.48% – “Shaper | Rielle ““Kit”” Peddler” – 32 wins 34 losses
45.96% – Criminal | Gabriel Santiago – 57 wins 67 losses
44.84% – Anarch | Noise – 100 wins 123 losses
44.28% – Anarch | Quetzal – 31 wins 39 losses
40.00% – Shaper | Nasir Meidan – 12 wins 18 losses
35.71% – Shaper | Chaos Theory – 20 wins 36 losses
25.00% – Shaper | Exile – 1 wins 3 losses
20.00% – Criminal | Stealth Operative – 1 wins 4 losses

Yeah yeah, GRNDL/professor is the new meta, &c &c.


#7

I went looking for the SHL2 final standings, but couldn’t figure out how to see past leagues on challengeboards. Do you have the final standings? I’m interested in seeing what more might be gleaned here :).


#8

http://www.challengeboards.net/boards/standings/2155


#9

I played about 50 games with Ken, I’m proud of getting him to fourth place! NEH matchup was the worst, but Blue Sun, RP & HB was ok.


#10

I would be curious to see the standing with a glicko2 rating system instead :smiley:


#11

What is this? Some Star Trek saga statistics? I don’t even know those identities haha.


#12

Because the way OCTGN outputs stats it puts the faction of the corp/runner instead of the actual ID name. So instead of Blue Sun: Powering the Future, it replaces the first line with Weyland Consortium. Same with Nisei Division and Cerebral Imaging.

I’ve been annoyed by it since Creation and Control and it will probably never change.


#13

So much for NEH having a below 50% match-up against all the top tier runner decks I guess.


#14

Does anyone know whether the code that spits out this data on github somewhere? I have a mind to fix it so that it gives the actual ID name.
@db0 @bblum


#15

I made a quick updated standings based on glicko2, using the implementation found here. Used the following settings:

  • Initial rating: 1500
  • Rating deviation: 100, based on assumption most people fall in the 1400 to 1600 range. Increasing this seems to heavily favor people who played few matches but had a high win %.
  • Tau: 0.5
  • Sigma: 0.06

Top 10 is then:

matuszczak : 1647 (rd : 50.48)
bblum : 1635 (rd : 29.36)
crunchums : 1634 (rd : 52)
flickerwisp : 1625 (rd : 35.55)
paranoid31 : 1623 (rd : 42.33)
moioioi : 1620 (rd : 38.18)
calimsha : 1619 (rd : 28.89)
g_man : 1609 (rd : 35.22)
moreoverlords : 1604 (rd : 61.49)
yaba : 1601 (rd : 55.49)

Sorry @Calimsha, still not in top 4 :stuck_out_tongue: Full standings for those interested here


#16

Yeah, it’s interesting that the early IDs used the card name (e.g. “Kate McCaffrey”) but newer ones seem to use the card subtitle (e.g. “Stealth Operative”).


#17

I guess that makes me formally opposed to glicko :stuck_out_tongue:

I’m not familiar with the system and I expected it to be just a more refined version of ELO. What could be the reason for big shifts between both rankings?


#18

The main difference between the two is that GLICKO calculates an “uncertainty” level of your rating based on how many games you’ve played recently among other things. So, if you don’t play for a long time and then lose a game, you lose a lot more than if you had been frequently before taking the loss.


#19

I’m surprised that The Foundry did so poorly, I’ve always thought that ID ability was strong.

Also, dat professor win rate :open_mouth:


#20

Well, I played a Grail Foundry that didn’t really work, a Bootcamp Foundry that didn’t really work, and a Director Haas Foundry that didn’t really work.

Me and anybody else brewing with rough lists are probably responsible for most of those losses.

It’s unquestionably worse than HB:ETF just giving you free money constantly with no downsides. Every ICE you tutor makes R&D weaker, makes Beta Tests weaker, gives up information, and due to Parasite spam it’s hard to accumulate enough of your NEXT to do anything.

If you are running Bootcamp then you’re committing money to ICE without the Runner even having to make runs and potentially be surprised. If you’re not running Bootcamp then the Runner controls when your ability fires, which means you won’t get it when you need it most.

That said I don’t think it’s as bad as these stats say because the stronger players self-select into HB:ETF lists that are also more tuned. The ID is probably capable of being only 5% worse than ETF.