On 7/11/2017 at 4:25 PM, Daigotsu Steve said:

Do people not gain honor for winning fights in L5R these days?

You have to had a minimum 10 witness and at least local news coverage to gain honor, regardless of outcome.

On 11/7/2017 at 6:50 PM, kiramode said:

This segues into the discussion of how to weigh fate vs honor vs cards. The methodology of how to get to the equation of 2 fate=2 honor=1 CC was pretty busted. The fundamental problem of the methodology was assuming that all the cards in question were equal in power. You can't compare Honored Blade, Ancestral Daisho, and Height of Fashion to Ornate Fan and Fine Katana because the Fan and the Katana are both dramatically better that the other 3 cards.

If that is your objection to the equation, I think you got it backwards. Allow me to elaborate.

@phillos makes an interesting remark in another thread, so I hope he's not offended if I quote him here. Even if I don't necessarily agree with what he was trying to argue there (Rokugan's Shield thread, page 6), I think his observation is accurate:

" I'm fairly certain when they design these cards they probably have some function as a starting point to guess a card's cost versus benefit, but at the end of the day cards are shaped by the empirical data they gather from play testing. To do it any other way would be to either risk making a very boring mechanical game or risk ignoring the situational benefits of cards and create exploitable situations, which were easily avoidable. "

As I understand it, numbers are important as baselines, but there are some factors that can ovverride them.

However, game designers usually do take some values as baselines and design cards matching them. Since in LCG there is no need for unplayable cards, we can assume the designers' goal was to make every card at least useful. However, good and bad cards do exist: the difference between a them though does not (as a general rule) reside in the designers having ignored the baselines ( a 2 cost character with 4 Mil 4 Pol and no downsides, for example ) but because of hidden variables or situationally exploitable abilities. From a game designer's point of view, for many cards there's no real way to know how good they will end up being without some playtesting, since LCGs are very complex games, so much that even the game designer her/himself does not master their every aspect. Knowing how valuable a situational ability is, or getting to understand a previously hidden, or considered irrelevant, variable is a key aspect of the skills involved in playing such games.

Coming back to our discussion, the point of the equation was trying to glean what that baseline might have been for the design team (I explicitly state it " when designing games such things are generally used as guidelines"): the a posteriori judgment on the cards' power levels is precisely what makes you question the equation that was being used, but it is a consequence of you learning how to play the game - it is not what the designers used a priori, when they designed them (of course your judgment could also be wrong and the cards you considered weak are actually more useful than you thought for whatever reason).

In particular, regarding Fate, it is clear that putting a Fate token on a high cost character is worth more than that single Fate you spent - which sounds paradoxical of course, but that's because you have to factor in the tempo cost: you spend 1 Fate now to gain the equivalent of X Fate the next round (X=character's cost). Is it worth the tradeoff? The answer is "it depends": here lies the skill of a player. Good Omen is so good because it allows you to gain that sweet sweet long term (relatively speaking) gain without sacrificing your tempo (thus your board presence) as much. By the way, at the end of the article I clearly state that Fate is worth more than the equation would seem to indicate.

On another note, concerning the 2:1 ratio benchmark: the reason why pretty much the only cards that match your (extremely high) expectations are low cost is the fact that you are not factoring in the card cost - which is fixed and does not scale with the fate cost. Being efficient with your Dynasty slots will surely be key to victory, so playing 4 small guys every round surely is a viable strategy, but, for example, using Holdings and higher-cost characters is another effective way to use those slots. In addition to this, a board of 5-6 characters costs a whooping 5-6 Fate to mantain, whereas a board of 2-3 strong low-costers costs only 2-3. This generally means that the 'taller' board can afford to achieve less in a given round and outwait the 'wider' board, since, player skills being equal, the taller board will outperform the wider one in the long term. Of course it depends on how much time is left before the game ends which is, again, another tempo vs long term consideration. There's also the fact that, as you noted, different board states lead to different weaknesses (tall board is weak to removal/control instead), with which I completely agree.

Bottom line is: I think this game rewards making efficient plays above all - which surely makes it a very challenging game to learn and play.

I definitely agree with your bottom line. At a purely competitive tournament level all the FFG LCGs effectively are an exercise in making judgement calls that yield the most efficient actions.

My point was to distill it down to a mathematical model while not impossible (anything in the physical world can be represented in this way even if we have yet to fully discover the relationships involved) it will yield a very complex model. More complex than saying 1 honor = 1 card for example. Since so many variables are in play a card's value floats based on a number of things including the composition of your deck, your opponents deck, the current board state, when in the match we are observing, what information is known, the current tactics being employed. I means we've seen that time and time again small tweaks to the card pool or shifts in the meta can cause us to reevaluate the entire card pool. This is one of the prime virtues of LCGs since it keeps the "puzzle" dynamic in our minds. On the design side since it's quite an effort to understand all the variables in play I assume they instead just used empirical data from play testing and do some sort of statistical analysis to shape the game's card pool such that it's reasonably fun and balanced while at the same time growing many different styles of play to keep things diverse (which explains why sometimes we do get things that are over or under powered in reality). Therefore a card in a vacuum may fall off the expected curve of a derived relationship in the game (like fate cost versus skill totals), but a number of variables not being taken into account may prevent a card from being actually broken or worthless.

My intent wasn't to completely dismiss the effort to understand the underlying math behind the game. I am sorry if I came off sounding that way. The post was just to caution people against using too simple a model as the gospel. I like reading articles that attempt to analyze the game in this way. I think it gets more interesting once we start playing the game and someone is collecting statistics like we did for Conquest because then it puts all these theories to the test.

ADD: Also while I'm sure it's their goal to make every card functional I also think they have the agenda of supporting different types of play (i.e. a Johnny, Timmy, Spike type consideration). Therefore we need big exciting effects, janky combo possibilities and straight up efficient card choices. While a Janky combo piece or a silly big effect card may not seem efficient enough for a tournament deck I wouldn't consider that a design flaw exactly. Also the designers may need to meet a cut off date in their schedule and nerf a card to play it safe rather than do enough analysis to correct an issue found during play test. Stuff like that can give us mathematically sub-optimal card designs.

@Eu8L1ch

I think we have two fundamental differences when viewing the topic of value.

1) Does the designer's value equivalences matter when it comes to judging cards?

2) If you're not using the designer's equivalences, what standard should be used?

As for the first question; it's not hard to figure out what the designer's believe equivalences to be. 1 conflict card is worth 2 honor via the bidding system. Ancestral keyword is worth one fate plus two scrolls and clan restriction based on Daisho/Katana comparison. You can continue the exercise with many similar cost cards to find out what the absolute equivalence is in the head of designers.

But this game(like every card game ever) has many clear misses on card evaluation. That's not to say that the designers are bad. This happens in literally every game that ever been printed. People who make the game have an idea of what's valuable and the moment the game is played it's clear that the value standard was off.

As for the second question; this is a matter of opinion. I personally find card draw and honor to be incredibly hard to evaluate and am not trying to evaluate it. However, power per fate isn't very hard. I personally find my 2 for 1 standard to be really low. You find it to be very high. That's a matter of opinion. But I think players are much better off making value judgments based on what are established good cards as opposed to making them off of what the designers had in place.

15 hours ago, phillos said:

My point was to distill it down to a mathematical model while not impossible (anything in the physical world can be represented in this way even if we have yet to fully discover the relationships involved) it will yield a very complex model. More complex than saying 1 honor = 1 card for example. Since so many variables are in play a card's value floats based on a number of things including the composition of your deck, your opponents deck, the current board state, when in the match we are observing, what information is known, the current tactics being employed.

[...]

My intent wasn't to completely dismiss the effort to understand the underlying math behind the game. I am sorry if I came off sounding that way. The post was just to caution people against using too simple a model as the gospel. I like reading articles that attempt to analyze the game in this way. I think it gets more interesting once we start playing the game and someone is collecting statistics like we did for Conquest because then it puts all these theories to the test.

I think that's an accurate description of what's going on.

As for your intent, I did not read anything bad into it, no worries. I agree it can be hard to "understand the underlying math", but it's still a worthwhile effort IMO because we get to understand more of the game itself in the process.

RE the various types of cards: that's a good point. However, I think they usually try hard to make (almost) everything useful at least to a certain degree, especially in the Core Set. That certainly does not mean that all cards end up being useful though, since in practice the game ends up dictating some choices - which is precisely when it gets interesting for me.

14 hours ago, kiramode said:

@Eu8L1ch

I think we have two fundamental differences when viewing the topic of value.

1) Does the designer's value equivalences matter when it comes to judging cards?

2) If you're not using the designer's equivalences, what standard should be used?

1) Yes, in a way. They are supposed to match an "expected value", so when the designers' expectations and the card's power level do not match you have something to investigate: this is how you come to understand game dynamics that might not have been apparent at first.

2) There is no absolute standard to use. Your 2-for-1 or 3-for-1 ratio is as good as any, the interesting part is the reasons you give for those cards being good or not being good (even with such a "raw numbers" approach you have to deal with cards with text-based abilities).

My point is that, while the designers' equivalences will (most likely) end up being a bad standard to rate cards, they are useful as a comparison point and serve to shed some light on the game mechanics.

Also, let me be clear: your 2-for-1 standard is extremely high in general only if you mean it as "raw numbers only". If you start factoring in the possible effects of card abilities as something that influences those numbers (like, let's say, covert is worth a quasi +4 when attacking, on average) the situation changes. The main reason why yours is a problematic approach is that turning an ability into a numerical value (as Phillos too pointed out) is very difficult, if at all possible, since the values are constantly shifting in different situations. If you take into account some kind of "conversion" from text abilities, your 2-for-1 standard is still a high point but is close to the mark IMO. However, high cost cards need to have a lower fate efficiency: just imagine a 5-cost character with an equivalent value of 10-15 strength - it would provide too efficient of a return for your single Fate, the equivalent of buying 5 low-cost characters. Why would you ever bother buying those? Kill/bow effects would have to be everywhere to justify paying 4 extra Fate (i.e. 8-12 extra strength for the next round) to keep your board intact.

Considering your claim on the possibility of a future 3-for-1, since the best low-cost cards do provide a 3-for-1 ratio (Doji Whisperer, Matsu Berserker, Hida Guardian), and the best high costers are close to a 2-for-1 ratio, it's possible 3-for-1 (text abilities included) will become the new golden standard. However, if that ends up being true, high-cost cards would have to still fall below the 3-for-1 because of what I said above. That is, whatever the standard is, if you don't take into account the advantages inherently granted by a high-cost personality, none of those is going to look worthwhile to you.

So, coming back to your point number 2, your stats-cost ratio standard is only good if you provide the means to convert text box abilities into numbers that can be factored into said ratio and if you provide the basis for your reasoning - i.e. your equation is useful if it is used as comparison point (the same of course can be said about 2 honor = 2 fate = 1 card). Note that I'm not criticising your method, which I find as helpful as any other since you did attempt to find a "conversion rate" for textbox abilities - my remark is merely aimed at a specific issue (low cost chars efficiency vs high cost chars efficiency).

Now a few examples to show what I mean:

Akodo Toturi. Let's focus on his highest stat: he provides 6 Mil for 5 cost, so the ratio does not seem very good. However, he also has an ability which provides additional value: a Ring trigger - using your unit of measure i.e. strength - is usually worth at least 2-3 strength (think about Fire or Earth; they're likely worth more than that). So Toturi is a 6 Mil + 3 strength after the conflict, which is already decent. Now let's say you have a Way of the Lion in hand. Way of the Lion usually provides a 3 Mil boost (Matsu Berserker, Lion Pride's Brawler, Matsu Beiona, Honored General) which makes it a strong inlcude regardless of Toturi (Banzai is +4 but requires 1 Honor, and is generally regarded as a very strong card). If you have Way of the Lion in hand with Toturi on the board, you gain an extra value of +3 from it (6 Toturi - 3 Average), which puts Toturi at a conditional 6 + 3 + 3. Of course you have to fulfill a few conditions in order for this to happen, but it seems to me Toturi is actually a lot more worthwhile than it would seem at first: going by your 2-for-1 standard, Toturi warrants the inclusion as "3x no questions asked", since he's arguably better than most other contenders: those conditions are not that hard to meet.

Guest of Honor. She provides 4 Pol for 4 cost, which is way below the 2-for-1 curve. However, let's say you are attacking with a Glory 2 character and your opponent is waiting to fire off their For Shame! and Courtly Games: those two cards alone are worth at least a +2 each, which would easily put the GoH at 4+4; anything more than that (your opponent might want to play a Rout, for example) is just added value on top. Now you can say that not every time your opponent has that many events (or is willing to play them) and that you're not discarding them but merely preventing those from going off in this battle. However, in a game in which both players bid very high, availability of the events might not be an issue, willingness to play them might be dictated by the board state (important battle) and the ability to play them later might be irrelevant (endgame). The point here is that GoH has an ability which is very situational (many conditions to be met), with a low floor but also a very high ceiling (Events are very powerful in this game).

Your "numbers only, strenght-based" approach in the examples above is not very worthwhile, since it's very difficult to translate those aspects of the card into raw numbers. Taking the stats-cost ratio as a starting point is fine, but unless you're willing to equate cards to strength, fate to strength etc, that approach IMO falls flat entirely. Compared to that, the advantage of using an economic model instead (taking cards and fate as the basis, and trying to find equivalences for those) is that it allows you to make more extensive comparisons, since they're more fundamental elements of the game, whereas determining strength usually is the goal, the endpoint - but not even that method is immune to the difficulties of trying to find equivalences, of course.

@Eu8L1ch

When I use my 2 for 1 baseline, it's primary purpose is to try to find some sort of way to evaluate text. When it comes to characters, text generally falls into one of the following categories:

- Stat bonuses

- Bow/unbow effects

- Economy bonuses

- Card prevention/ability prevention

Stat bonuses and bow/unbow effects are pretty easy to quantify. You may need to meet certain requirements. Venerable Historian only works when you have more honor. Brawler is only as good as the biggest stat you can push off the table. Nitan Adept needs an attachment(which costs a card plus probably a fate) and needs to target a guy without anything on them. But the reason you put these characters in your deck is because you're presumably going to play your deck in a way to maximize the cards you put in.

It's really hard for cards with stat bonuses and bow effects to not be able to hit a 2 for 1 threshold if played correctly.

As for economy bonuses; those are really hard to quantify. Cards that give fate bonuses are easy because you can deduct the fate from the cost of the character and then recalculate value. Card and honor bonuses are pretty hard. Those bonuses are highly dependent on the deck. I don't think we have enough info to properly quantify.

Card prevention is tricky. Some card prevention effects like Covert and selected discard from the Kitsuki Investigator provide more or less guaranteed value. You know exactly what you are stopping and how to plan around it. Effects like Guest of Honor and Watch Commander are different. You might be stopping stuff, or maybe the cards that would have been used in one conflict simply get used in one of the other 3 conflicts. Getting card effects like that to really provide value requires that you craft situations where your opponent would need to use the cards you're preventing in a particular engagement. In my experience a good player can always stash cards until the time is right and avoid situation where they are forced to do anything. That's largely why I don't value effects like that too highly.

Lastly; it's important to remember that just about any big body can hit a 2 for 1 value ratio by simply adding fate to them. Anyone who has ever played the game knows the danger of adding more than 1 fate to anything with the existence of kill spells, void ring, and dishonor effects. You also run the risk of not placing enough skill on the board right now that could make the additional value you get over subsequent turns not as valuable. But this is where skill comes in. One of the biggest skills in this game is trying to determine whether you need value on the table immediately or if you need sustained value.

The reason I view the 2 for 1 standard as a low standard is because most of the cardpool can hit the standard. By my estimation something like 65-75% of the cardpool can quantifyibly hit the quota with very little help. Be that because they are efficient chuds(Otomo Courtier), have easy to quantify bow/unbow effects(Nitan Adept/Master), are unique, or simply are big bodies. My goal when I build a deck isn't to desperately search for guys that can hit the bar. It's to find enough guys that can hit the bar so I don't dead draw too often, but have enough guys that can punch well above that bar that become your game winning cards.

Sidenote: I really only use this standard as a rough starting point for dynasty cards. Conflict cards have much higher potential and also have the benefit of being able to be held for the right opportunity whereas dynasty cards kind of have to be played the moment they come out. I feel like you can take more situational cards for conflict, but require much higher value from them. Dynasty you need guys that are useful in most situations so you don't dead draw.