$10.00
Recommend
9 
 Thumb up
 Hide
16 Posts

7 Wonders» Forums » Strategy

Subject: Cards that Lead to Victory: Age III rss

Your Tags: Add tags
Popular Tags: [View All]
J. Alan Henning
United States
flag msg tools
designer
mbmbmbmbmb
Here are some of my hypotheses for which Age III cards are more likely to lead to victory. I’ve analyzed the hypotheses based on reviewing the tableaus of winners from 100 randomly chosen 3-player games from JSP. While the sample size isn’t large enough to be conclusive, it is superior to anecdotal evidence.


H. The winner will have played a Guild card more often than not.

Yes, 72% of the winners played Guild cards: 43% played 1 Guild card, 24% played 2 cards, 4% played 3 cards, 0% played 4 cards and 1 player played 5 Guild cards. The average winner played 1.5 Guild cards.


H. The most valuable Guild card is the Builders Guild (1 VP per Wonder stage completed by the player and his or her neighbors).

Yes, 20% of winners have played the Builders Guild (approximately 40% of the players who could have, as Guild cards only appear half the time in 3-player games). The Strategists Guild (1 VP per neighbor’s -1VP marker) was a close second, at 19%. Third was the Traders Guild (1 VP per neighbor’s Commerce card) at 15%.


H. After Guild cards, winners concentrate on playing Culture cards.

Not so. Culture cards come first. On average, a Culture card was played 31% of the time, compared to 22% for Guild cards (I’ve doubled the average of 11% per Guild card since each card appears half the time).


The relative value of cards in Age III is Guild cards, Culture cards, Military cards, Commerce cards and Science cards.

Not so. The relative value of cards is in fact: Culture cards 31%, Commerce cards 24%, Military and Guild cards each at 22%, and Science cards at 20%.


H. Those who win and scored 18 military will have accomplished this with 3 or less cards on average.

Not true. Of the 76 games for which I recorded military scores, 12 winners (16%) scored 18 points. One winner had 2 Military cards, 4 had 3 cards and 7 had 4 cards. The average number of Military cards earning 18 points was 3.5 cards.

Of all winners, 10% played no Military cards, 14% played 1 Military card, 15% played 2 cards, 39% played 3, and 22% played 4. Not a single player won the game if they had played 5 Military cards.

Winners scored on average 3.4 points per Military card.


Notes

For a full description of the methodology, see my post on Age I cards.

For a full list of some of the caveats with the methodology ;-) please see my post on Age II cards.

Here is the list of cards with the percent of winners whose tableaus included each card.

Incidence – Card (Color)
37% Senate (Blue)
32% Palace (Blue)
31% Town Hall (Blue)
28% Pantheon (Blue)
27% Arena (Yellow)
26% Haven (Yellow)
26% Gardens (Blue)
25% Fortifications (Red)
22% Arsenal (Red)
22% Study (Green)
22% University (Green)
20% Lodge (Green)
20% Builders Guild (Purple)
19% Lighthouse (Yellow)
19% Circus (Red)
19% Strategists Guild (Purple)
18% Academy (Green)
17% Observatory (Green)
15% Traders Guild (Purple)
12% Philosophers Guild (Purple)
10% Shipowners Guild (Purple)
10% Magistrates Guild (Purple)
9% Craftsmen Guild (Purple)
6% Spies Guild (Purple)
4% Workers Guild (Purple)
3% Scientists Guild (Purple)

This wraps up my series on “Cards that Lead to Victory”. Got any hypothesis of your own that you’d like me to analyze with my database of 100 games?
8 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steven Metzger
United States
Pullman
Washington
flag msg tools
designer
mbmbmbmbmb
Jeffrey Henning wrote:
Got any hypothesis of your own that you’d like me to analyze with my database of 100 games?
Yes.

Is 100 games too small of a sample size?
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jim Krohn
United States
New York
flag msg tools
designer
You keep using that word. I do not think it means what you think it means.
badge
Ahhh....my misspent youth...
mbmb
Jeff, I just wanted to say thank you for the work you put into this series. I have read all of them. No, your methodology is not perfect and, yes, it would be better to have a larger sample, but we are talking about a lot of work here. I have found what you have done to be helpful.

Thanks for your effort!
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
J. Alan Henning
United States
flag msg tools
designer
mbmbmbmbmb
Thanks for the kind words, Jim. I hope this series advances the state of the art in understanding 7 Wonders strategy while pointing out the need for further research. I went with 100 records because it made the math simple whistle and because it was all I was willing to do manually. I think whomever takes this to the next level will have access to data dumps from BSW or JSP.

Steven, with surveys the fallback is always: does it have predictive validity? Does it actually relate to the outside world in some way? I was able to use the strategies here to break into the top 100 players of 2100+ 7 Wonders players on JSP. So I have proven to my satisfaction, at least, that these results are representative of JSP games. Not that every individual finding is, but that the findings in general are. YMMV.

- Jeffrey
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Morten Brobyskov
Denmark
Odense sv
flag msg tools
mbmbmbmbmb
I have a one thing I would like you to check.
Is there any wonder and side that is better (wins more) than the other
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
J. Alan Henning
United States
flag msg tools
designer
mbmbmbmbmb
The sample size when divided 14 ways is too small to draw any conclusions from. So I wouldn't change how I play based on this data.

Here's what I found:

Alexandria A - 5%
Alexandria B - 9%
Babylon A - 8%
Babylon B - 3%
Ephesos A - 9%
Ephesos B - 7%
Gizah A - 9%
Gizah B - 10%
Halikarnossos A - 6%
Halikarnossos B - 4%
Olympia A - 4%
Olympia B - 11%
Rhodos A - 7%
Rhodos B - 8%


Sorted by percent:

11% - Olympia B
10% - Gizah B
9% - Alexandria B
9% - Ephesos A
9% - Gizah A
8% - Babylon A
8% - Rhodos B
7% - Ephesos B
7% - Rhodos A
6% - Halikarnossos A
5% - Alexandria A
4% - Halikarnossos B
4% - Olympia A
3% - Babylon B
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Morten Brobyskov
Denmark
Odense sv
flag msg tools
mbmbmbmbmb
its interesting to see there is such a big difference in the percentage of winning games
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Evil Roy
United Kingdom
Sutton
Surrey
flag msg tools
mbmbmbmbmb
Jeffrey Henning wrote:
Here's what I found:

Alexandria A - 5%
Alexandria B - 9%
Babylon A - 8%
Babylon B - 3%
Ephesos A - 9%
Ephesos B - 7%
Gizah A - 9%
Gizah B - 10%
Halikarnossos A - 6%
Halikarnossos B - 4%
Olympia A - 4%
Olympia B - 11%
Rhodos A - 7%
Rhodos B - 8%


Sorted by percent:

11% - Olympia B
10% - Gizah B
9% - Alexandria B
9% - Ephesos A
9% - Gizah A
8% - Babylon A
8% - Rhodos B
7% - Ephesos B
7% - Rhodos A
6% - Halikarnossos A
5% - Alexandria A
4% - Halikarnossos B
4% - Olympia A
3% - Babylon B


Each of these percentages are out of the total 100 games, but obviously not all of the wonders were used in every game. A better set of figures would be the percentage of games won out of the number that the wonder was used in.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Derek Cabellon
United States
Tacoma
Washington
flag msg tools
mbmbmbmbmb
I would like to see the difference between green, blue, and red in 3 player, 4 player ... 7 player games. I'm thinking that science becomes less valuable as more players are in a game and military becomes more valuable.

Just a thought I've had with the games that I've played.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Paul W
United States
Eugene
Oregon
flag msg tools
mbmbmb
I think it'd be interesting to see a histogram of winning scores in various categories, particularly military and science.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Bill Gallagher
United States
Torrance
California
flag msg tools
mbmbmbmbmb
Only 3% with the Scientists Guild? Count me as a winner with it. On the last turn of a game (7p) I played last night, I completed the final stage of my wonder, which allowed me to build a discarded card for free. Despite having 7 raw/manufactured goods cards in my tableau (yes, the guild providing a bonus for those was available as well), I took the Scientists Guild because 9 green cards had been played by my neighbors. My overwhelming military edge (18 to negative 6) allowed me to neutralize my left hand opponent's 25 point advantage in science (26 to 1 - my penultimate play was my only green card, as the other two cards were military, and I had that locked). I won by ONE point.

It also helped that I picked up 21(!!) points in blue cards in the first three rounds of Age III. Yes, my right hand opponent had passed me juicy blue cards on successive turns after I had played one in the first turn of Age III.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Bill Gallagher
United States
Torrance
California
flag msg tools
mbmbmbmbmb
derekjcabellon wrote:
I would like to see the difference between green, blue, and red in 3 player, 4 player ... 7 player games. I'm thinking that science becomes less valuable as more players are in a game and military becomes more valuable.

Just a thought I've had with the games that I've played.

That depends on a large extent on how many people decide to develop science. If only three or four make a serious effort (the others opting for a blue/red or perhaps yellow strategy), then it's still possible for one or two players to get 26 (two complete sets) for science (more is unlikely due to competition for the few remaining cards). In both seven-player games I played yesterday, one player was able to get 26 points in science.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Max Maloney
United States
Portland
Oregon
flag msg tools
"If trees could scream, would we be so cavalier about cutting them down? We might, if they screamed all the time, for no good reason." -Jack Handey
mb
The interesting thing about this that matches my experience is that I always score mostly blue cards when I win. My sample size is very small but I've nevertheless noted this with frustration.

What I end up feeling is that blue cards are points independent of any supporting strategy, which makes them a "bulletproof" strategy if you can buy them. I very frequently end up picking a blue card because it mathematically excedes any other option in my hand.

And it seems to work. So dull.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Flame Bird
msg tools
If you are in a game where all resources are available, or players are focused on access to all resources, blue cards become the most powerful in Age III.

I think the vast majority of games go this way. Right now I think the tempo created by resource selection is not being talked about enough. I've played in some very tough games with good players where key resource cards have been denied. For example, Rhodes B denied double stone or ore, Gizah being denied dobule stone, Manufactured goods being pitched rather than let through.

The game sharpens up a lot when players are actively counterdrafting each other. This is harder and harder to do with more players.

The 3er and even 4er games can be cutthroat if players are trying to minimise the potential of opposing picks. Unfortuantely, the majority of games I've played are very, very permissive, and even just stopping a runaway science player can be difficult to do alone.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
J. Alan Henning
United States
flag msg tools
designer
mbmbmbmbmb
Here are the average scores, Paul. Is this what you were looking for?

9.3 Military
3.4 Coins
7.6 Wonder VP
15.6 Blue Buildings
2.5 Yellow Buildings
8.6 Guilds
10.4 Green
57.4 Score

- Jeffrey
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Paul W
United States
Eugene
Oregon
flag msg tools
mbmbmb
7 Wonders » Forums » Strategy
Re: Cards that Lead to Victory: Age III
Not really...the averages cover up a lot of information. What I was talking about was a histogram for each category, where the x-axis was the number of points scored in that category and the y-axis was the number of games in which that many points were scored.

Something like this:

Coin Scores for Winning Players
 
 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.