A rating system has been implemented to evaluate the overall effectiveness of builds (meat, magic, spells). The rating is displayed when visiting the thieves' guild after the construction phase is complete.
The calculation principle is as follows:
(total value of purchased meat) / 10k
+ (remaining gold) / 20k
+ magic (value of guild levels: 0.2 + 0.3 + 0.4 + 0.5 + 0.6)
+ racial bonuses
Top-tier Human builds (for the first player):

Build notations:
- Weeks are listed separated by commas (first, second, third)
- Numbers indicate the tier levels
- Numbers with a plus sign indicate upgrades (if upgrades are placed separately from the base, on the following week, they are indicated in parentheses)
- Cap, Mag, Rat - Capitol, Magistrate, Town Hall
- Castle, Stronghold, Fort - Castle, Stronghold, Fort (if they are passage buildings, they may not be indicated for brevity)
- GuildN - guild of the corresponding level (if it is a passage building, it may not be indicated for brevity)
Capitol Start
(early capitol: maximum gold, minimum meat)
Approximately 50 effective 8.0+ builds were reviewed on the
stream, from which the best were selected:
Militia Debut
Cap123, Stronghold4+7, (1+)Guild4 = 8.4 (meat 4.4, magic 1.4, gold 5.1 / 2)
Cap123, Castle4+, Guild4 = 8.3 (meat 3.6, magic 1.4, gold 6.5 / 2)
Cap123, Castle56+, (1+)Guild1 = 8.3 (meat 5.7, magic 0.2, gold 4.9 / 2)
Griffin Debut
Cap25, Castle3+6+(5+), (2+)Guild1 = 8.4 (meat 5.8, magic 0.2, gold 4.9 / 2)
Cap25, Castle13+6+, (1+)Guild1 = 8.3 (meat 5.7, magic 0.2, gold 4.8 / 2)
Cap25, 34+7, (2+)Guild4 = 8.3 (meat 4.7, magic 1.4, gold 4.4 / 2)
Stronghold Debut
CapStronghold23, Castle(3+)46+, (4+) = 8.3 (meat 5.3, magic 0.5, gold 5.1 / 2)
CapStronghold23, Castle(3+)456, (4+) = 8.3 (meat 5.5, magic 0.5, gold 4.5 / 2)
CapStronghold23, Castle1(3+)56+, (1+)Guild1 = 8.3 (meat 5.7, magic 0.2, gold 4.9 / 2)
CapStronghold23, Castle4+5, Guild4 = 8.3 (meat 3.9, magic 1.4, gold 5.9 / 2)
Monk Debut
Cap124, (4+)57, (1+)Guild4 = 8.4 (meat 4.6, magic 1.4, gold 4.9 / 2)
Cap124, Stronghold(4+)7+, (1+)Guild4 = 8.3 (meat 4.6, magic 1.4, gold 4.6 / 2)
Non-standard debuts (not included in the top):
Progressive Debut (some options are listed):
Cap34, Castle(34+)6, Guild4 = 8.1 (meat 3.8, magic 1.4, gold 5.8 / 2)
Cap34, Stronghold(4+)57+, (3+)Guild4 = 8.0 (meat 4.5, magic 1.4, gold 4.2 / 2)
Cap34, Castle(3+)56+, (4+) = 8.0 (meat 5.2, magic 0.5, gold 4.7 / 2)
Progressive Debut, Stronghold variant
CapStronghold4, Castle(4+)57, (5+)Guild4 = 8.0 (meat 4.0, magic 1.4, gold 5.2 / 2)
Economic Start
(early Magistrate)
...will be reviewed later
Standard Start
(early Town Hall)
...will be reviewed later
Meat Start
(without an early Town Hall)
...will be reviewed later
Training Builds
(with the construction of training halls)
...will be reviewed later
The arena is a very convenient tool for testing builds, as it clearly shows any imbalances in the strength of buildings and prices. The arena should provide a good idea of the game on the map, both in terms of the strengths/weaknesses of a particular build and the overall game dynamics.
Through experimentation, the optimal day for the final construction phase was established for the arena - the 16th (for the first player) and the 15th (for the second player). During this time, 60% of the new buildings are constructed (15-16 buildings out of 25-26 + 3-4 starting buildings).
Accordingly, the full-fledged map will also be tailored to these timelines, plus or minus a week. On the fourth week, from a construction point of view, there is absolutely nothing to do (only to simply complete the city to its maximum). And on the first week, many builds have not yet had time to develop (a period for rushes).
If before this I thought that in theory it was necessary to focus on timelines of about 10-14 days, now I am convinced of this in practice (with a slight shift towards 15-16). This time is enough not only to complete the development of any thematic build but also to fine-tune it with a couple of buildings at the end (like an extra guild or an additional upgrade).
So far, I am very satisfied with the results of testing for Humans; the city scheme works 100% as intended, and the prices adequately reflect the value of the buildings. During the stream, I managed to analyze about 50 basic effective builds (not counting variations of final tuning), and this is only for the Capitol start for the first player (a rough estimate is that this is 10% of all possible builds!). Other starts (economic, standard, meat, training) and openings for the second player (which have a fundamentally different pace) remained behind the scenes. But I also conducted internal testing on them; they are approximately equal in effectiveness, without significant imbalances (at the moment).