When I discovered the double build strategy, I imagined that the probability of an event order which is suitable for double build would be very low. Through some number of games, I'm feeling that such event orders come up more frequently than I expected. So here let me try giving a rough estimation of how often such event orders would come up.
There are 10 months and the Droughts are 2 among them. So there are total C(10,2) = 45 possible placements of the Droughts. For the sake of simplicity, we'll primarily be looking at the Droughts, although in practice the distribution of the other events is important too (the Tributes in particular). Note that according to the rules, if two identical events are drawn in succession, the latter one will be shifted to one month later. This will be taken into account in the counting below.
Type 1: both Droughts early
You start double-building after, or even during, the second Drought. There are 2 chances (3-4 and 3-5) for the second Drought to fall on the 5th month, and 3 chances for the 6th month (3-6, 4-5 and 4-6). That's 5 chances total.
If the second Drought falls on the 7th month (4 chances), you will have to do double build actions on rounds 7 and 9 or it won't be fruitful. Double building on rounds 8 and 10 would yield only 16 VP, and it'd be less hassle to just get a couple of scholars and do research instead, or just get early Ladies. It's not very ideal, but it may be doable depending on how the other events fall.
Type 2: both Droughts late
You double build in the early game, on rounds 2/3 and 4/5 (depending on whether the events would kill your persons) so that you have 6 or 7 palaces by round 4 or 5. You should start with a Builder, and build on round 1 too unless you get a free chance to tax for $5 or farm for 2 rice. You'll probably need to farm twice before the first Drought. Even if you don't make enough rice for the second Drought, it may still be okay if it's late enough (round 11 or 12) since the persons are just 2 VP each with little extra utility that late in the game.
There are 3 chances (10-11, 10-12, 11-12) for the first Drought to fall in month 10, and 3 chances (9-10, 9-11, 9-12) for month 9. That's 6 chances total. If the first Drought comes on month 8 (4 chances), it becomes difficult to feed your palaces: 8-12 is probably doable, 8-11 a bit tough, and 8-10 probably bad (and if a Tribute is in between, don't!)
Type 3: the Droughts split wide apart
The idea is to double build in between the two Droughts, immediately after (or even during) the first Drought. If the Droughts are split wide apart enough, this gives you time to double build twice and farm twice.
There is 1 chance (3-12) for the Droughts to be 9 months apart, 2 chances (3-11 and 4-12) for 8 apart, and 3 chances (3-10, 4-11, 5-12) for 7 apart. If the Droughts are 6 months apart (4 chances), the feasibility is lower, and it really depends on what events fall between: non-action events (Contagions and Mongols) make it easier, while action events (Festival and especially Tributes) make it difficult. For now, let's count it as 6 chances.
In total, this gives 5+6+6 = 17 chances out of 45 for a double build game, or roughly 1/3. This means that, if you can't play double build, you'll be at a significant handicap in roughly one-third of the games. Even if you're not playing double build yourself, you should understand it well enough so that you can make the life of the double-builder difficult. (Double build is quite difficult to play in practice, but highly rewarding if successful. See example here. http://www.mabiweb.com/modules.php?name=GM_YearDragon&g_id=4...)
To be worth it, you have to double build at least twice; otherwise it'd be less hassle to just get a Lady instead of the second Builder! With the more favorable event orders, the opportunity of a third double build is very lucrative.
The above has not taken into account other factors such as how the other events fall (Tributes immediately preceeding Droughts make it tougher, etc.), or the small chance that another event pair drawn in succession might cause a Drought to be shifted (neglegible since that's a fraction of a 1/45 chance).
"One of the key problems today is that politics is such a disgrace, good people don't go into government." - Donald Trump
Actually, I believe its 36 possible starting placements for any given tile (can't be placed adjacently.) 8+7+6+5+4+3+2+1 = 36.
OIC, you took the shifting of duplicate adjacent events into account already!
- Last edited Thu Mar 18, 2010 5:59 pm (Total Number of Edits: 1)
- Posted Thu Mar 18, 2010 5:49 pm
That was an interesting analysis - and an impressive demonstration at the link you sent: I've definitely never seen anyone build 9 palaces before. (I think I've seen 7, but he lost to the person with 2). I'll have to keep double-building in mind!