Recommend
14 
 Thumb up
 Hide
19 Posts

Android: Netrunner» Forums » General

Subject: [Stimhack Article] Discerning the Metagame: Spin Cycle rss

Your Tags: Add tags
Popular Tags: [View All]
Lysander
United States
Georgia
flag msg tools
Avatar
mb
Hey all, just published a Stimhack article, analyzing Spin Cycle and its effect on the competitive Netrunner environment utilizing moderately high ELO stats from OCTGN and Stimhack's Tournament Winning Decklist compilation.

http://stimhack.com/discerning-the-metagame-spin-cycle/

Feel free to discuss, and an extra special thanks to Miaowara Tomokato for the ELO graphs, db0 for OCTGN, and Sneakysly for article assistance and hosting!
15 
 Thumb up
0.02
 tip
 Hide
  • [+] Dice rolls
Grant Cain
msg tools
badge
Avatar
mbmbmbmbmb
Great article. Interesting the Jinteki droped in the spin cycle. My guess is that with Shock! and other new toys, more people finally got interested in taking them to the next level.... and failed.

Anyway, thumb for you.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Lysander
United States
Georgia
flag msg tools
Avatar
mb
I think the biggest reason Jinteki dropped off is because Kate rose so much. It's really hard to intimidate an opponent with access to Deus X + Clone Chips.

FFG, may we get unpreventable net damage?
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Zeb
United States
New Hampshire
flag msg tools
You found me!
badge
Hiss.
Avatar
mbmbmbmbmb
Lysander1 wrote:
I think the biggest reason Jinteki dropped off is because Kate rose so much. It's really hard to intimidate an opponent with access to Deus X + Clone Chips.


Definitely think this is the case, my Kate deck only loses to Jinteki in rare circumstances.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
John Fanjoy
United States
Virginia
flag msg tools
mbmbmbmbmb
Why did EFS drop off so much? (I had to look up what EFS stood for, I'm used to thinking of it as Custom Biotics. There's also an inconsistency there, the chart refers to that one as EFS but it also refers to NEXT rather than GTN).
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
John Fanjoy
United States
Virginia
flag msg tools
mbmbmbmbmb
I also think you're misusing the term "god tier," and it seems arbitrary that you put Gabe into that group with Andromeda, but dropped Kate to the next tier down. Judging just from the numbers you have presented, I think I would group the runner identities as

A: Andromeda, Gabe, Kate
B: others
C: Professor

and for corps:
A: HB ETF, HB CI, NBN MN, NBN TWIY, W GRNDL, W BABW
B: J PE
C: HB NEXT, J RP
trash: HB ST, HB CB, W BWBI
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Lysander
United States
Georgia
flag msg tools
Avatar
mb
CitizenFry wrote:
Why did EFS drop off so much?


Its Genesis sample size is so miniscule, its actual win rate is probably closer to that of its Spin Cycle rate. That being said, there are very few meaningful ways to spend that extra 7 influence (most are highly predictable upon seeing the ID hitting the table), and ETF's ability typically dwarfs anything you can spend the extra influence on anyways. That being said, I have a pretty good Custom Biotics Red Coats deck, but even then anyone can make the argument that it'd be as good if not better as ETF.

As for its initials on the graph, I realized that as well, but that's Tomokato's handiwork.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Lysander
United States
Georgia
flag msg tools
Avatar
mb
CitizenFry wrote:
I also think you're misusing the term "god tier," and it seems arbitrary that you put Gabe into that group with Andromeda, but dropped Kate to the next tier down.


Fair point, I had to assign a cutoff somewhere, and I didn't want to group Kate in the same category as Andromeda since they have a 4% differential from one another. I can definitely see the argument that she deserves to be with Criminals though.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Daniel Wray
United States
Colorado
flag msg tools
mbmbmbmb
Lysander1 wrote:
CitizenFry wrote:
I also think you're misusing the term "god tier," and it seems arbitrary that you put Gabe into that group with Andromeda, but dropped Kate to the next tier down.


Fair point, I had to assign a cutoff somewhere, and I didn't want to group Kate in the same category as Andromeda since they have a 4% differential from one another. I can definitely see the argument that she deserves to be with Criminals though.


What's the error range for these percentiles?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Lysander
United States
Georgia
flag msg tools
Avatar
mb
cynchwyrm wrote:

What's the error range for these percentiles?


Great question that hopefully Tomokato can answer. I assisted in defining the parameters for the graphs and the subsequent analysis, but he was the one that analyzed the raw data and constructed the graphs.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
R. Fetterkey
msg tools
Avatar
mb
I'm confused by your statement that the game is "all about speed" for Corps. If anything, the reverse seems true. The Spin Cycle has enabled a lot of strong, slower-paced taxing decks.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Captain Frisk
United States
Connecticut
flag msg tools
badge
MIND|GAMES
Avatar
mbmbmbmbmb
Lysander1 wrote:
I think the biggest reason Jinteki dropped off is because Kate rose so much. It's really hard to intimidate an opponent with access to Deus X + Clone Chips.

FFG, may we get unpreventable net damage?


Is there really enough Jinteki / Kate matchups to account for that large swing? Wouldn't the same jump apply to all of the other shapers as well?

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Kevin Anon
msg tools
Captain_Frisk wrote:
Lysander1 wrote:
I think the biggest reason Jinteki dropped off is because Kate rose so much. It's really hard to intimidate an opponent with access to Deus X + Clone Chips.

FFG, may we get unpreventable net damage?


Is there really enough Jinteki / Kate matchups to account for that large swing? Wouldn't the same jump apply to all of the other shapers as well?



I guess this begs the question: Are there stats/numbers on specific match-ups?

It would be interesting to see who wins/loses to who.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Dave Chandler
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Lysander1 wrote:
I think the biggest reason Jinteki dropped off is because Kate rose so much. It's really hard to intimidate an opponent with access to Deus X + Clone Chips.

FFG, may we get unpreventable net damage?


Don't forget Professional Contacts. That card is a bullet in Jinteki's head
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Lysander
United States
Georgia
flag msg tools
Avatar
mb
fetterkey wrote:
I'm confused by your statement that the game is "all about speed" for Corps. If anything, the reverse seems true. The Spin Cycle has enabled a lot of strong, slower-paced taxing decks.

True, but that's a more recent development, which is enabled by runners focusing entirely on winning early against fast advance and ignoring the prospect of a lategame. The more glacier builds gain prominence, the more runners will shift builds to combat them. Thus the meta will shift back and forth like a pendulum, and we'll start seeing more well-rounded runners to address both extremes.

Captain_Frisk wrote:
Lysander1 wrote:
I think the biggest reason Jinteki dropped off is because Kate rose so much. It's really hard to intimidate an opponent with access to Deus X + Clone Chips.

FFG, may we get unpreventable net damage?


Is there really enough Jinteki / Kate matchups to account for that large swing? Wouldn't the same jump apply to all of the other shapers as well?

I'm interpreting this question two different ways, so forgive me for clarifying my original point before explaining. First, Kate rose far more than the other Shapers simply because she has the best ability and can address the widest variety of Corp archetypes. Jinteki decreased because she became so good, not the other way around.

My other point would be that the number of Shapers rose from 5,227 to 9,235; whereas Anarchs and Criminals combined rose from 10,589 to 13,152. That means previously Shapers comprised 33.4% of the field and they now make up 41.3% of the field; an increase of nearly 8%. Not much, but keep in mind Jinteki's win rate decreased by only 6.4%. I think it's a fair assumption that an 8% increase can account for at least half of that 6.4%, considering it's easily Jinteki's worst matchup.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Captain Frisk
United States
Connecticut
flag msg tools
badge
MIND|GAMES
Avatar
mbmbmbmbmb
djc6535 wrote:
Lysander1 wrote:
I think the biggest reason Jinteki dropped off is because Kate rose so much. It's really hard to intimidate an opponent with access to Deus X + Clone Chips.

FFG, may we get unpreventable net damage?


Don't forget Professional Contacts. That card is a bullet in Jinteki's head


Yes - it is like Jackson Howard in that it really dicks over one specific ID while also being decent all other times.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
R. Fetterkey
msg tools
Avatar
mb
If Anarch grows more prominent, Wyldside is even worse for Jinteki than Pro Contacts is.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Bryan Goodwin
United States
West Linn
Oregon
flag msg tools
badge
Avatar
mbmbmbmbmb
cynchwyrm wrote:
What's the error range for these percentiles?


There's no sampling error, since these charts constitute 100% of the games being examined (OCTGN games meeting Elo & Competitive criteria). What you see in the charts reports exactly what happened, not an estimation based on statistical sampling.

By way of example: when you see a national survey in the news, it's likely based on the polling of ~1000 randomly selected people, and will include a +/- x% error due to the fact that their selected population may be skewed due to relatively small sample numbers compared to the population. If the polling company contacted the entire national population rather than a sample, this would not be necessary.

A better question to ask is "What is the likelihood that the observed differences between x and y factions/time periods are due to chance?" You can construct a hypothesis and validate it with statistical methods, in this case a Chi-square test:

"Assuming a 'true' expected win rate of 49%, the difference between HB:ETF's Genesis Cycle win rate (49.7%) and Spin Cycle win rate (48.5%) is attributable to chance". In this example case, there's a high probability that this hypothesis is true, p=0.42 (generally the probability of a difference has to be lower than 0.05 to be considered significant).

(there are mathier ways than this to precisely examine complex relationships, but that's above my pay grade)

5 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Bryan Goodwin
United States
West Linn
Oregon
flag msg tools
badge
Avatar
mbmbmbmbmb
CitizenFry wrote:
Why did EFS drop off so much? (I had to look up what EFS stood for, I'm used to thinking of it as Custom Biotics. There's also an inconsistency there, the chart refers to that one as EFS but it also refers to NEXT rather than GTN).


Yep, it's inconsistent alright; I'll fix it to show as CB in future treatments. Never noticed because, hey, Custom Biotics! Who looks at that? whistle
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.