Recommend
2 
 Thumb up
 Hide
12 Posts

Terra Mystica» Forums » Strategy

Subject: mending MEND rss

Your Tags: Add tags
Popular Tags: [View All]
Loon
United States
flag msg tools
mb
I've been thinking about the MEND meta (explained below), as brought up in another thread recently discussing possible tournament changes. I started wondering about what could be done *now* to shake it up, from a game theory perspective.

The reason that MEND is so popular seems to come from each person picking the generally best available faction of the remaining choices. Engineers have a playable board? First pick. Darklings are always good. Second pick. Nomads will be safe and have a decent chance. Third pick. Well, I guess Mermaids will be able to survive better than the others. Done.

The situation I'd like to look at is after the first two picks: Engineers-Darklings-?-? (or equivalently, Darklings-Engineers-?-?). If the last two players continue to pick the "best available faction", we already know they'll both be at a big disadvantage. So what if they team up... do they have better options?

Obviously Witches are the best at specifcally hindering the Engineers, but at quite a cost to themselves as well. If the Witches player is expected to cooperate, they have to be given a chance to win as well, and Darklings still need slowing down. So how about Brown? Currently both brown factions are very selectively chosen, with fairly high success. This screams to me that more could be done with them to shake up the meta.

Now, back to the scenario, the third player can't pick Witches, otherwise the last player will take Nomads, sinking the Witches and Engineers, and racing against the Darklings alone. So I propose more third-picks of brown factions, which would enable the last player to take Witches and still feel like they have a chance to win.

Of course, the fourth player might not play along...
If they choose Nomads (for Engineers-Darklnigs-brown-Nomads) then brown is in lots of trouble, Engineers will be great, and the other two play for 2nd place. This doesn't strike me as better than taking the Witches.
The fourth player would be open to pick Chaos Magicians with only the Engineers as neighbors, but that also doesn't sound particularly risky for brown (especially Halflings).

Now, I don't have the scripting skill (or patience) to get all the data I'd want for this. Can anyone find out the win rates between MEND factions in games with exactly those four factions? What about Engineers-Darklings-Halflings/Cultists-Witches? Engineers-Darklings-Halflings/Cultists-ChaosMagicians? (Are Chaos Magicians even playable in that situation? I sound like I know what I'm talking about but I just finally got promoted to D6, which means I still think all kinds of crazy stuff is winnable!)
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Juho Snellman
Switzerland
Zurich
Zurich
flag msg tools
Avatar
mb
I added a faction filter checkbox to https://www.snellman.net/tmp/tm/tune/ . I don't know if it really tells enough of the story though, given how many other things have an effect. (E.g. Halflings picked before Darklings are a lot better than the other way around).
5 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Paul Kessels
Netherlands
Bussum
flag msg tools
Avatar
mbmbmbmb
Until the beginning of this month there have been 332 League games with a MEND setup.

- 92 times Darklings won with an average score of 147
- 112 times Engineers won with an average score of 150
- 70 times Mermaids won with an average score of 143
- 58 times Nomdas won with an average score of 141
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Paul Kessels
Netherlands
Bussum
flag msg tools
Avatar
mbmbmbmb
There have been 1549 league games with Darklings and Engineers in the game, but in a non MEND setup (based on log files; # is slightly higher than the above link shows...maybe this excludes non finished games or dropped players, etc...anyway for the initial stats that would not be a big impact).

These are the results:
- Auren - 3 times winner (avg score of 130)
- Chaos Magicians - 74 (143)
- Cultists - 58 (147)
- Darklings - 584 (144)
- Engineers - 476 (149)
- Fakirs - 1 (141)
- Giants - 7 (133)
- Halflings - 65 (146)
- Mermaids - 84 (140)
- Nomads - 92 (139)
- Swarmlings - 64 (145)
- Witches - 41 (143)

So when Darklings and Engineers are in play, in ~18% of the cases it is a MEND setup. Of the DEXX games:
- Darklings win 36% of the time
- Engineers win 31% of the time
- When Mermaids also in play, they win 19% of the time
- When Nomads also in play, they win 18% of the time
- When Witches also in play, they win 13% of the time
- When CM also in play, they win 14% of the time
- When Cultists also in play, they win 5% of the time
- When Halflings also in play, they win 18% of the time
- When Swarmlings also in play, they win 16% of the time

4 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Robert
Germany
Bocholt
flag msg tools
badge
Avatar
mbmbmbmbmb
jsnell wrote:
I added a faction filter checkbox to https://www.snellman.net/tmp/tm/tune/ . I don't know if it really tells enough of the story though, given how many other things have an effect. (E.g. Halflings picked before Darklings are a lot better than the other way around).
I'd assume that early Halflings picks are mainly in setups with SPADE>>2 in round 4.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Paul Kessels
Netherlands
Bussum
flag msg tools
Avatar
mbmbmbmb
DocCool wrote:
jsnell wrote:
I added a faction filter checkbox to https://www.snellman.net/tmp/tm/tune/ . I don't know if it really tells enough of the story though, given how many other things have an effect. (E.g. Halflings picked before Darklings are a lot better than the other way around).
I'd assume that early Halflings picks are mainly in setups with SPADE>>2 in round 4.



In those 18% of Halflings wins, in 89% of the cases there was the SPADE>>2 in the game (14% in R1, 4% in R2, 10% in R3 and 72% of the times in R4).

Of those Halflings wins with Darklings and Engineers and R4 SPADE>>2 in the game, in 50% of the cases Halflings was chosen before Darklings.
BON1 was chosen by Halflings at the start in 21% of the times and BON8 29% of the times. Score 5 and Score 8 seems to be favorable in R1 with each in 20% of the time.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Luke J
msg tools
Do people really pick Engineers from Seat 1? Every time I've tried that someone ALWAYS picks Witches. yuk
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Gambia
Brest
msg tools
Avatar
ljwoodw wrote:
Do people really pick Engineers from Seat 1? Every time I've tried that someone ALWAYS picks Witches. yuk


If only Witches were a real counter to Engineers, people would pick Engineers less from an early seat.

With F&I factions, green volcano or variable factions do counter engineers effectively...
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Robin Zigmond
United Kingdom
Durham
flag msg tools
mbmbmbmbmb
Skyswooper wrote:
ljwoodw wrote:
Do people really pick Engineers from Seat 1? Every time I've tried that someone ALWAYS picks Witches. yuk


If only Witches were a real counter to Engineers, people would pick Engineers less from an early seat.

With F&I factions, green volcano or variable factions do counter engineers effectively...


My experience, at much more modest levels than those we're talking about here, is that Engineers are indeed a risky pick from seat 1 or even seat 2, because Witches do often still get picked. Often this results in both Engineers and Witches doing badly, so you can probably trust top level players not to pick Witches against your Engineers. Nevertheless, if both factions are in the game, while both do badly, in my experience Engineers generally do worse. This is because Engineers are strong late in the game but have a very fragile early economy, and are very vulnerable to the green and grey hexes they need being stolen from them by a green faction. Witches on the other hand have a more robust economy early on, as well as potential Witches' rides to quickly get to another part of the board if they get blocked somewhere.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Fallen Angel
msg tools
jsnell wrote:
I added a faction filter checkbox to https://www.snellman.net/tmp/tm/tune/ . I don't know if it really tells enough of the story though, given how many other things have an effect. (E.g. Halflings picked before Darklings are a lot better than the other way around).


actually this tool is great.

1) If you want to find out who counters engineers, just set the top 5 divisions, last few tournaments (i choose 7) and try conbinate engineers with every other faction. Only faction who beats engineers (except darklings) are witches, however they end up in this games bad too:


Faction - Count - Score
darklings - 85 - 4.041176470588235
cultists - 17 - 3.6470588235294117
mermaids - 23 - 3.3043478260869565
halflings - 14 - 3.0714285714285716
nomads - 63 - 2.4206349206349205
witches - 109 - 1.7889908256880733
engineers - 109 - 1.614678899082569


so the conclusion is, there is not faction who counter engineers well.



2) If you want to find out what is best pick when DE are present (as 3rd seat) you just click theese two faction and you immidiately see what other factions has best results - which is by far cultist, then mermaids (both still have theit stats lower the DE).

When DEC are present and you pick faction from 4th seat:
- if you pick nomads or mermaids, you kill cultist and yourself and victory goes to engineers
- if you pick witches you kill engineers, and you give small advantage to DC
- if you pick any other faction, you loose

So my conclusion when you go from 3rd seat and DE is in game is:
- if last player is smart and aware, you should pick cultist and he will take witches since it is his best shot (which actually negates my first conclusion):

darklings - 16 - 3.5625
cultists - 16 - 3.5
witches - 16 - 2.25
engineers - 16 - 0.6875



- if last player is NOT aware and there is big chance he will take nomads or mermaids you shoud choose mermaids, because DEMN gives you better chance the DECN

engineers - 127 - 3.2992125984251968
darklings - 127 - 2.6535433070866143
mermaids - 127 - 2.1023622047244093
nomads - 127 - 1.9448818897637796
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Fallen Angel
msg tools
jsnmthw wrote:

Of course, the fourth player might not play along...
If they choose Nomads (for Engineers-Darklnigs-brown-Nomads) then brown is in lots of trouble, Engineers will be great, and the other two play for 2nd place. This doesn't strike me as better than taking the Witches.
The fourth player would be open to pick Chaos Magicians with only the Engineers as neighbors, but that also doesn't sound particularly risky for brown (especially Halflings).


stats disagree:

darklings - 55 - 3.7636363636363637
cultists - 55 - 2.6363636363636362
engineers - 55 - 2.4545454545454546
chaosmagicians - 55 - 1.1454545454545455

_______________


engineers - 36 - 4
darklings - 36 - 2.861111111111111
cultists - 36 - 1.6388888888888888
nomads - 36 - 1.5

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Fallen Angel
msg tools
lets try shuffle game with theese 4 factions (E,D,C,W) to test what i mentioned above (sample in stats was only 16 games), since the are recently so popular. The rules would be:
- random setup
- first seat must choose D, second E, third C, last one W
- each player plays one game with every faction
- games will be unranked since it pretty much depends on luck who gets good setup for his faction.

Who is in?
1 
 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.