Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Balanced matches against human opponents usually give 2 extra starting influence to the US player.
If you learn the game, you will find yourself able to beat the AI every single time, no matter how unlucky the dice rolls (maybe 1 in 1000 games the dice and cards might combine to make things unwinnable, I suppose).
Today after a good night's sleep and reading tips and stuff, I tried again and won as US in early mid war.
I think key to my victory was denying the Soviets from couping anywhere in Asia or Middle East by being ultra aggressive myself with coups, e.g. I couped Iraq as one of my first actions, thus throwing the soviets out of the majority of the Middle East early.
I've used the China Card to flip Thailand back to me with Influence, when Soviets went into it early with two Influence and I countercouped them, when they couped me in South America over Columbia, two times!
And I acted on a hint I saw in a video, that to not try to dominate the regions but just keeping the balance.
Except South America, In the final turn of that match I had to countercoup them so aggressively there over Colombia, because I had the South America Scoring and that card in the end granted me victory over them.
If you are on the back foot, or reasonably even in board state, the AI actions can seem "intelligent" (or even "prescient", if you are very new to the game), but as soon as the AI is on the back foot, it starts to flail about in increasingly desperate and nonsensical ways.
Basically, the better you get, the worse the AI gets, which goes some way towards explaining how much the solo experience diverges between new and experienced players.
Then again, as Benkyo says, you will almost always still be able to win in the end. There's some satisfaction in getting behind and slowly clawing your way up to victory. One of the very few times I lost was when I got 3 defcon lowering cards in one hand and found no way of disposing of them.
1. A perceived "bad dice rolls" or "bad card draws" pattern may be an issue of confirmation bias, since we all naturally tend to be more greatly discouraged by a crummy roll than we are reassured by a good roll. My understanding of the randomization is that it's actually harder to code a tilted randomized system to favor the AI side than it is to code a purely random one, and TS' dice rolls and card drawing mechanics are pretty close to what a purely random generator would produce.
2. Getting 3 scoring cards or very low-point cards are unfortunately a part of the game design. I don't like them either, but they do occur (and have even occurred in an International TS Tournament in Turn 1 for a player, who then very predictably lost the game). This could be an argument that the game is poorly designed, or designed with so much swinginess that it doesn't deserve championship tournaments, etc. But I would not impugn the digital app for this, since it's been a constant feature in the hardcopy board game from the start.
2.5 The low-point cards is especially egregious, since if you are unlucky enough to draw a low-point card, then you've thinned out the deck and your opponent is more likely to draw a higher-point card. Some TS fans I know have come up with a "High Deck" and "Low Deck" drawing mechanic to try to help both players get more even-powered cards, but it's finicky and in any case it's a problem with the original game design, not the game software.
3. When the AI decides to focus on a region that it had previously ignored: note that the software version of the game clearly marks which regions have already been scored and whose Scoring Cards are clearly in the discard pile - there's a little bulls-eye target icon beside the region name to show it's gone. (If anything returns the SCs back into the draw deck, that little icon vanishes, to show it's potentially scoreable again.) This interface feature was not present in the original hardcopy version of the game, and players had to mentally keep note of which cards had gone into discard. The discard pile is usually (though not always) reshuffled into the deck at two distinct turns: at the start of Turn 3, and at the start of Turn 7 (although sometimes if the China Card passes back and forth a lot, the Turn 7 reshuffle might be skipped). Human players can therefore predict with high confidence the composition of about half your cards in your hand at the start of Turn 3, and a similar level of card confidence at the start of Turn 7. The AI probably doesn't count cards, but it does keep an eye on which regions have been scored, and it usually tries to play to strengthen its position in unscored regions.
In my experience, if you're having trouble fighting the AI, be aware of a few clear weaknesses in its behavior (which you should not get too complacent with, since humans won't have these weaknesses):
- The AI tends to play influence into breaking Battleground control purely to reduce the scoring advantage of a region. It doesn't play long term, but plays round by round. So where a human might play all 4 points of a strong card into one BG to control or set up threat of control in a later round, the AI will more likely use the 4 points to break two separate BGs, meaning you can easily recapture them by spending a measly 2 influence next action round. Skilled humans don't do this.
- The AI tends to stop putting influence into countries once it achieves bare control, which means it's easy for a human to break AI control and maybe even set up a takeover to flip the country. Skilled humans will usually over-control vulnerable BG states (e.g. putting 3 influence into Thailand instead of 2) to make the cost of breaking their control prohibitively high. Sometimes skilled humans will also over-control one non-battleground state in the region to prevent easy loss of Domination scoring.
- The AI is extremely negligent towards an auto-defeat by losing full control of Europe. Skilled humans are so sensitive to the auto-loss condition of enemy Control of all European BGs, that they tend to react very quickly to such threats and as such humans almost never achieve European Control victories against each other. The AI for whatever reason has a huge blind spot to European Control (perhaps because it was trained on a library of games where humans never even attempted such a long shot victory condition).
If you're having difficulty, the main three beginner tips for general Action Round prioritization:
1. Prioritize your position in regions that have not yet been discarded to the discard pile;
2. Prioritize putting influence into BG states over non-BG states;
3. Prioritize putting influence into empty states over contested states.
This is on top of more general Turn-based advice, such as "try to coup at least once" and "be careful of cards that degrade DEFCON", etc.
The AI doesn't even know what Scoring Cards it's got in its own hand, sometimes. I've seen the AI play a load of Action Rounds and then straight up auto-lose the game because it held a Scoring Card.
Yeah, Ask Not... can do that even to a thinking human player, to say nothing of the AI.
Not to invalidate your experience, but I've seen the AI self-lose games by holding scoring cards often enough that it's clearly a flaw in its programming. I'd say maybe 10 times in total? Out of 1000+ games.