Dominion Strategy Forum

Archive => Archive => Innovation Articles => Topic started by: popsofctown on March 08, 2013, 05:02:56 pm

Title: Splayability types
Post by: popsofctown on March 08, 2013, 05:02:56 pm
Splaying is nice, because it helps you get more icons, and that's usually a good thing.  It's also one of few strategical considerations in Innovation, even though Innovation is mostly about tactics.  Usually you don't want to spend time splaying unless you get a short term benefit in addition to the long term benefits, but understanding when those long term benefits matter most is important.


The black hex throws quite a wrinkle into splaying strategy.  Depending on where the black hex is, some of your splay types will be messed up, some won't.  It's even harder to sort out when you realize there is no perfect, ideal placement for the black hex. 

There are four splay levels: unsplayed, left, right, and up.  It doesn't take long to figure out that even with worst case scenario black hex placement, upgrading from one level to the next doesn't hurt you (for the overall total, at least). 

One thing that is not glaringly obvious though, is the following theorem:

Quote
"For each card, exactly one of the following is true: the card reveals additional icons when its splay is changed from left to right, or the card reveals additional icons when its splay is changed from right to up."

In other words, either Invention or Railroad will improve a cards iconnage, not neither, not both.  Thus, every card in base Innovation is one of two types.  To most clearly convey the meaning of those types, I will call them "Invention Island" cards and "Railroad Sprawl" cards. 
Remember: neither type is "better" than the other.  These types don't convey how useful a card is, but it tells you how the card wants you to splay it.  An Invention Island card wants to be splayed right, and after that it is happy, it doesn't want anything more from you.  A Railroad Sprawl card wants you to splay it up.  Once you've splayed a Railroad Sprawl card left, it wants to be splayed up, and it has no interest in being splayed right.

Ironically, Invention is actually a Railroad Sprawl card, and Railroad is actually an Invention Island card.  Try not to get confused.


I've attached a chart of what color and age the Invention Island cards belong to.  Exactly one third of the Innovation cards belong to Invention Island.  There are a few things to be aware of, there are a couple patterns.

1. As BrokenTree has commented before in another thread, yellow has an unusually large number of Invention island cards.  The chart says red has more, but 3 of those are from age 1, which has more cards in it anyway.  Adjusting for that they are even.  And if you adjust for the large number of Castles appearing in low red Invention Island cards, it becomes even easier to declare yellow the king of Invention Island.

2. Blue goes an astounding 4 ages without a single Invention island card, the longest streak of any color, and its the first four ages.  And there's actually only one streak of 3, which doesn't really count because the streak includes Age 10, which rarely is covered.  So splaying blue left early on can pay off in the early game, you won't get the urge to upgrade from left to right anytime soon.

3. Age 4 has the fewest Invention Island cards, but it is immediately followed by age 5, with the most.  The invention effect, itself, belongs to age 4, but won't be very useful on covered 4's. 
If you've only covered up 1's, 2's, and 3's around the time you get Invention, you're looking at 5 red cards that can benefit from invention, 4 yellow cards that can benefit, 2 green cards, a purple card, and no blue cards whatsoever.  One moral of the story is that Invention is a pretty troubled card, stuck improving a stack of red cards that will yield you icons of a probably irrelevant type, or improving the yellow pile that can splay itself on its own with Reformation.  Perhaps a more useful moral of the story is that it might be a wise idea to spam Experimentation before returning to Invention later, since age 5 has the most Invention Island cards.

4. All other things being equal, select Purple for Feudalism, Purple for Industrialization/Emancipation, and yellow for Publications (blue tries to overcompensate for its strange behavior in the first four ages later on).

Title: Re: Splayability types
Post by: BitTorrent on March 09, 2013, 02:42:10 am
My own observation(or experience...whatever)

1. Any covered BASE set cards with a castle will give you a castle when splayed left. This is universal no matter that card is a castle dogma card or not, say mapmaking is a crown card but if it is covered and your green is left splayed, it will still shows a castle. Same thing applies on Fermenting and Machinery.

2. Basically I memorize cards like this:
There are 4 types of card which are classified with the position of the hexes, say...
(locate the icons from top left to bottom right, X is hexes and _ is an icon or echo or whatever)
Type A: [X][_][_][_] Type B: [_][X][_][_] Type C: [_][_][X][_] Type D: [_][_][_][X]

When I am good I can actually remember where the exact icon is for each card...at least for base cards.

3. Insight
Type A cards are good for splaying left or up, since splaying them to these direction would show no loss of icons due to the position of the hexes, same thing applies for Type C cards, where type C is good at splaying left and right.

Type D cards are only the best when splayed right, which is frequently spotted from the middle ages (frequently spotted from age 5 but there are examples in the more early ages such as Optics and Anatomy). This shows that you are invited to splay your board right rather than left in order to harness these cards, since splaying left with these cards would only shows an hexes but no icon.

Type B cards are rare. You rarely see a card which you gain reduced icons both splayed right or up, examples are Engineering, Flight, Socialism and Mass Media. Won't explain this further since there is only a few of them.

In other words, Type C and type D cards are the 'Invention island' cards mentioned by pops and Type A cards are those 'Railroad Sprawl' cards.

4. In order to harness these knowledge the bottom line is: you got to know when is the best moment to splay your cards. A left splay may give you more castles but you may not need castle at all. A right splay though, is usually good enough for the whole game, even in the later stages, since most cards you will have in the middle of a color stack is from the early middle ages, so usually you don't need to care about some of the late age cards since they should be on the top of your pile at that time. Of course there are other ways to harness these knowledge though, especially consider echoes, you may have other approaches.
Title: Re: Splayability types
Post by: Thisisnotasmile on March 09, 2013, 05:43:32 am
This analysis is all well and good but I don't really see that it's that useful in a game of Innovation. Generally, when you get to the point that you can splay your cards (i.e. you have a splaying dogma available to you AND you have 2+ cards in a stack that it would let you splay), doing so is better than not if it takes you further along the path of no splay -> left -> right -> up.

You've often got a whole stack of cards that would be affected by this splay, and they're not all of the same "type". You might not increase the icon count on some of them, but others you will, and the ones you won't won't be decreasing their icon counts so it's not exactly a bad thing for them. Sometimes you might not increase your icon count at all, but you could change the icons available to you which might still be beneficial (e.g. you are leading 14-3 in light bulbs but are 4-3 down in crowns. Making a splay that gives you -4 bulbs and +4 crowns is good).

Other times the splay doesn't really do anything for you now... but it will later when you add cards to that pile. You could argue that now isn't the time to make the splay in this situation, and I may agree with you sometimes. But sometimes your splaying card is about to get covered and if you don't make the splay now, you don't know when you're going to get another chance.

That's just my thoughts on the subject. I may be wrong (and probably am since I'm not a great player), but to me this analysis seems good in theory, but is of little use in practice.
Title: Re: Splayability types
Post by: BitTorrent on March 09, 2013, 06:28:19 am
This analysis is all well and good but I don't really see that it's that useful in a game of Innovation. Generally, when you get to the point that you can splay your cards (i.e. you have a splaying dogma available to you AND you have 2+ cards in a stack that it would let you splay), doing so is better than not if it takes you further along the path of no splay -> left -> right -> up.

You've often got a whole stack of cards that would be affected by this splay, and they're not all of the same "type". You might not increase the icon count on some of them, but others you will, and the ones you won't won't be decreasing their icon counts so it's not exactly a bad thing for them. Sometimes you might not increase your icon count at all, but you could change the icons available to you which might still be beneficial (e.g. you are leading 14-3 in light bulbs but are 4-3 down in crowns. Making a splay that gives you -4 bulbs and +4 crowns is good).

Other times the splay doesn't really do anything for you now... but it will later when you add cards to that pile. You could argue that now isn't the time to make the splay in this situation, and I may agree with you sometimes. But sometimes your splaying card is about to get covered and if you don't make the splay now, you don't know when you're going to get another chance.

That's just my thoughts on the subject. I may be wrong (and probably am since I'm not a great player), but to me this analysis seems good in theory, but is of little use in practice.

The major usage of this knowledge IMO is when someone share such things to you, are you going to follow or not. Since most splaying dogma effect comes with a 'may' condition so you can choose not to splay and, often this will determine whether your opponent gets a free draw from sharing or not.

Say someone shares a splay right for your blue but your blue only got 2 cards. Are you going to take it or reject it to prohibit the free draw of your opponent. Now that is a question, something which these kind of knowledge may take place. Say you know you are only short of clock from your Empire, both of you are in age7, and right splayed blue can get 2 clocks each from tucking either blue cards from Age 8 (Rocketry, QT), but if you just meld either of them directly you can get the 3 clocks you needed. There is no model answer on this one, all it takes is your wits and insight on that specific scenario.
Title: Re: Splayability types
Post by: popsofctown on March 09, 2013, 10:26:17 am
TINAS, Innovation is a tactical game, whereas Dominion is a strategic game.  It's a lot harder to generate useful theory regarding the game that immediately and powerfully makes you win games, for that reason.  As you identify, the tactical components of your decisions will usually greatly override strategy. 

I tried to point out a couple applications in the opening post.  BrokenTree pointed out another very good point on sharing, since that's a case where you are likely to be comparing splay of future cards against the relatively minor benefit of your opponent drawing a card.  If Experimentation is the card providing the lightbulbs you need to share Invention, you know that the benefit of sharing is pretty high because Age 5 is packed with Invention Island cards. 
If your opponent Mathed straight into Printing Press and is offering you an opportunity to splay right your blues that are already left, you're still in the lower tech ages and you know that blue won't care about that splay until age 5.

As you say it is somewhat minor stuff.
Title: Re: Splayability types
Post by: Jerk of All trades on March 13, 2013, 02:33:33 pm
The only thing REALLY worth tracking are the type A and type D cards, where you get a major benefit for a splay choice.

If you get a bunch of type A, then you might prioritize splaying up, because you go from 1 to 3 icons.  This makes a difference for cards like Flight(8), which is usually a weak action once you are drawing 8s.  But If you know that you have a bunch of type As, then splaying up might be a good idea.  Otherwise splaying right to up usually doesn't have a great benefit. (unless you tuck more type As, but you probably wont want to play for future possible benefits from something like Flight, because the game is almost over.)

Splaying right is almost always a good thing, just because of Wonder. It's pretty rare when you'd pass up the opportunity to splay right, even if you give somebody +1 card.

Of course certain echo effects change this a lot.
Title: Re: Splayability types
Post by: saboor on April 18, 2015, 01:14:27 am
If you buy the physical game, the rule is bolded in the rules, and is followed by a sentence to the effect of, "We bolded this sentence because this tends to be the most commonly missed rule in Innovation"