War #11 - Market Prices

[kaempfer13]

Should we apply similiar ratios to salvage like we do for transmutes? like instead of 6 fixed cards you get like 12 salvage points to distribute, market price>60 cards cost 2 salvpoints, >120 3 salv points. I dont think discards will have to be modified, I think every team (except air) ran out of fodder after some time.

[MyNameIsJoey]

#biased post incoming.

Looking at ''experimental market prices''

prices seem fairly fair, if i compare some cards to each other.

Charger compared to squid. I personally think squid is a much stronger card than charger, though this can be argued and I'm relatively fine with the prices as is.

Flying weapon definitely needs a price increase in my opinion. I'd bring it all the way up to 70. We've seen sooo many adrenastaves, catatitans, from all teams, and I think upping flying weapon is a much better option than upping both titan and druidic staff.

Shard of Freedom. I know it is dominant. I know it is strong. But twice as much as any other card is a bit much. Could be brought down to 450 easily.

Holy Light. Holy light is one of the if not the most situational card in the game. What warrants it being at 50 rather than 30?

I think that's about it for my market comments. I'll try to build a vault if I have time.

[Vineroz]

Since this has been brought up in chat and I looked at it for a while, I am feeling I should put some of my comments here. They are mainly about cards and they might be a little bit biased.

Firstly I drafted this table where it compares the experimental price with War 10 price:



The first thing I notice is that ALL cards have been given a cost increase, and I am immediately wondering if other elements are the same. Turns out is the only element that receives no discount. I thought this only happens when a team won the last war, or when the cards are so popular that other teams includes a lot of them as well. I don't see both cases happen in last war.

has always had the problem of lacking in-element damage dealers. The cost increase of GotP and Golden Nymph might still be justifiable while comparing to other elements' creatures, but others like Dune Scorpion and Pharaoh are just insane. They went from situational to unplayable. It is just unreasonable to have Pharaoh as the same cost as GotP.

Even SoR got a cost increase. I don't think anyone in other elements used it, and we didn't use a lot of them either last war. I have no idea why it has a cost increase if you look at SoV or SoG where they are actually used by other elements.

[MyNameIsJoey]

now that im looking at www's sheet...

Isn't trident hella cheap? Pulverizer as well?

Trident should be upped to 100, and pulvy to 120 imo, each reaching another ''tier'' of weapons.

[worldwideweb3]

https://docs.google.com/spreadsheets/d/1FGk8kQfrqiDnm5aoXZulxzeXxRmeL6yA95qM4HDEHdE/edit#gid=114453631

Might help with comparison ^

Spoiler for Weapons:

Pulvy too cheap compared to others. Arsenic and Titan should be one tier above lobo and fahren. Trident and staff (or increase AW) should be increased a bit, perhaps to 100. Rest looks fine.

Spoiler for Creatures:

Cards that are cheap: SoFo, Toadfish (compared to crawler), FFQ, Green nymph, shreiker (?), Angel, crusader (when compared to other cards, its cheap), Turq nymph, cockatrice (when compared to horned frog)
Expensive cards: scarab

Spoiler for Spells:

RIP Fire bolt, but I guess that's fair enough, it's a pretty strong card. But why is nova and fractal not on the same level?? (nerf bolt to 200, dont increase the other two!!) Sosac at 170 looks a tad cheap. Blackhole is expensive. It's useless outside DBH, and other DBH cards got a nerf already so BH should be reduced. Stoneskin, Rage potion and Shockwave can also use a nerf i feel, they are situational and not that powerful in all honesty. Pandemonium needs a nerf. It's better than thunderstorm and RoF, so should cost more than them. Flying weapon definitely needs a nerf, too cheap at 50. EQ may also need a nerf. Holy light is useless and could be reduced to 30. SoSer is acc pretty good, too cheap at 40. Plague also pretty cheap, probably around 50 is good for it. PU can also do with a buff (decrease in price).

Spoiler for Perms:

SoFree is too expensive. About 300, 350 is fine in all honesty. Dial's a joke at 200 (should be less expensive), with SoPa at only 250. UG at 50 is very weird, seeing blue nymph is at 200. Nightfall and cloak could perhaps take a raise but not overly fussed about that one.

Spoiler for Shields:

Dusk should be more expensive, hope can be less expensive. Fog and bonewall both at 170 is weird. Maybe put BW to 200, fog 170, dusk 150. Emerald shield is cheap when compared to reflective. Reflective is only used for spell reflection but emerald can provide good damage reduction too. I feel reflective is only that high cause its used in firestall, but all other firestall cards are getting a nerf so reflective doesnt need one, its useless outside of it. Rest looks alright.

Spoiler for Shards:

Uh, i've already commented on them before. Raise for sosac, sofo, soser, potential raise for sopa. Decrease for sofree.

[Zawadx]

https://docs.google.com/spreadsheets/d/1794ycRzDZxHk-2WYS2_LsD6pFK8ErXclggyIcyCDIcA/edit?usp=sharing

This is the first draft of the final market prices, built on the base of a new and improved algorithm. As promised by the experimental market, there will be heavy modifications to try to make market as perfect as possible. So please, leave as much feedback as you can.
Do note that this is a very rough draft - none of these prices are final, especially those with the highest cost. It has yet to incorporate the new distribution perfectly and the latest round of suggestions. I'll finish these up and explain the new algorithm in detail in the morning~

[kaempfer13]

When is the final first draft of the first final market prices coming out?

[Zawadx]

The almost ready market prices are here! Now we will just have a few slight tweaks and test it out vs potential vaults.

https://docs.google.com/spreadsheets/d/1794ycRzDZxHk-2WYS2_LsD6pFK8ErXclggyIcyCDIcA/edit?usp=sharing



Feel free to suggest any modifications which you feel are necessary to the prices. If you want to help in the further development of war, you can build a vault following these market prices and pm it to The Warmasters. None of the information such as who sent in vaults or their contents will be disclosed to anyone, so your secrets are safe with us.

Build vaults with a maximum point total of 26,000 points. And since it might affect your vault strategy, note that we will start at the full 6 salvage, 20 discards from round 1 and transmutations will be done based on market price of the card being transmuted into (1:1 for cards costing 60 or less, 1:2 for cards costing 70-130, 1:3 for cards higher than that and 1:4 for SoFree).

In addition to the vault, it'd be helpful if you could answer these questions:

1. How did vaultbuilding feel compared to before?
2. Was it challenging/easy to build a good vault?
3. Did the prices push you in any direction (e.g. towards more diverse/non-conventional decks)?
4. Do you think your vault size will be below, around or above the average? How do you feel about that.
5. Name one thing you liked and one you disliked about these prices.

If you have the time, please help us out with your take on a vault. Here's to an amazing war!

[Zawadx]

So people asked me how the algo works. I'll detail it to make people aware of how this market came to be and to help future WMs:

Code: [Select]

The algo cost column is:
=arrayformula(if(mod(row(A2:A471),2)=1,"",IF(E2:E471="Quanta", 0, (CEILING(0.3*12*(log( ( (Q2:Q471-6)*(Q2:Q471-3)*(Q2:Q471)*(Q2:Q471+2) + 180)*(H2:H471)^3/(SQRT(G2:G471+8))/(log(Q2:Q471+6,2)) + 288, 2))*if(isnumber(find("Nymph",B2:B471)),1.3,1)/if(B2:B471="Nymph's Tears",1.3,1)))*2 -30  )))


which in less spreadsheet-cell-names means..

Pillars cost 0,
Nymphs cost 130% of what's to be described (Because nymphs are underused in war for a significant reason not linked to power level, they would be underrepresented otherwise)
The final result is rounded up

NOTE: Shards and Nymph no longer have no of decks used doubled because it was a silly blanket nerf to all of them without considering individual power levels.

This leaves us with the core:
0.3*12*log( D^3*P[T]/sqrt(U+8)/log(T+6,2) + 288 , 2)

And the final result after all modifications is multiplied by 2 and 30 is subtracted from it. This linear transformation fits it into the range from experimental market nicely (tho in the algo's case the larger values are a bit deflated).

U = Total usage of card
D = Number of decks using card (The main measurement behind the cost)
T = Number of teams which used this card at some point
P[T] = (T-6)(T-3)T(T+2) + 180, a polynomial to weight the effect of T


The main components of the stuff inside the log are:

1. D^(3/2), The main measure of popularity/usefulness of a card
2. P[T], Which was designed to give a low change when a card was used by 3-6 teams (i.e. a versatile, well-liked card), slightly higher when used by only one or two teams (narrow application) and very much higher for higher values (ubiquitous cards which are prolly thoughtlessly put into vault).
3. sqrt(D/(U+8)), scales down the cost for cards which have more copies put into deck slightly (by about 10 points from 1x to 6x), since one-off or low U/D cards are more likely to be in vault in less numbers, their cost should be a bit higher to compensate.
4. D/log(T+6,2), a rough measure of how many times specific teams would use a card. Mainly serves to dampen the extreme effect of almost every team using some card smartly.   

Log is used because it scales these variables to price nicely. They are multiplied instead of added because if these are added within log, it is impossible to strike a balance where each variable affects the log with a good enough variation.

Some things to consider for future:

1. Introduce the amount of the card in initial vaults as a variable
2. Consider use in-element and off-element seperately
3. Change P[T] from a quartic polynomial since it makes for a very extreme change going from 10 to 11 to 12.
4. The linear transform/changed distribution might make nymphs cost too high with the 130% increase.
5. Simply round the values to nearest multiple of 5.

The min reason why I used the algorithm is not to provide a definite value to the prices. But to provide a baseline measurement which is unbiased (the main formula was set only based on test values and overall distribution, not individual cards) and thus would not carry my lackings in knowledge, while also giving us a good measure of reaction to prices and effects last war. These were tweaked quite a bit based on our thoughts and everyone's feedback on experimental market. One big change is that the higher values in the distribution have been decreased in price, which was received favorably. There is a chance that this might have made them too damp/low to be really effective - we'll see.

To end, a big thank you to Physsion and Serprex for making the amazing gdoc which let me easily develop these prices. Thanks extends out to all the people they built on top of as well. And a more thanks to the people who've sent vaults in so far! Your input is really invaluable. On that note, we are still happy to receive vaults, thoughts and feedback on these prices. There are a few questions listed above - please reply to them if you construct a vault, or you feel like returning feedback. here's to a great war!

[Zawadx]

The almost ready market prices are here! Now we will just have a few slight tweaks and test it out vs potential vaults.

https://docs.google.com/spreadsheets/d/1794ycRzDZxHk-2WYS2_LsD6pFK8ErXclggyIcyCDIcA/edit?usp=sharing



Feel free to suggest any modifications which you feel are necessary to the prices. If you want to help in the further development of war, you can build a vault following these market prices and pm it to The Warmasters. None of the information such as who sent in vaults or their contents will be disclosed to anyone, so your secrets are safe with us.

Build vaults with a maximum point total of 26,000 points. And since it might affect your vault strategy, note that we will start at the full 6 salvage, 20 discards from round 1 and transmutations will be done based on market price of the card being transmuted into (1:1 for cards costing 60 or less, 1:2 for cards costing 70-130, 1:3 for cards higher than that and 1:4 for SoFree).

In addition to the vault, it'd be helpful if you could answer these questions:

1. How did vaultbuilding feel compared to before?
2. Was it challenging/easy to build a good vault?
3. Did the prices push you in any direction (e.g. towards more diverse/non-conventional decks)?
4. Do you think your vault size will be below, around or above the average? How do you feel about that.
5. Name one thing you liked and one you disliked about these prices.

If you have the time, please help us out with your take on a vault. Here's to an amazing war!

The prices will be locked in 24 hours. If you have any suggestions, now would be the time.

[JonathanCrazyJ]

You know you have done a great job as Water General, when flooding is no longer in the cheapest bracket!

[JonathanCrazyJ]

On a serious note, SoP shouldn't be 300. It is comparable in strength to dims and fractal, maybe a bit stronger, but the key thing is that it is usable by ALL elements, which shouldn't put it so close to SoFree, which is only usable by air (properly) and maybe life. Imo it should be 250.