Your remark made me curious about what winrate you need to break even, so I made a little program to compute the chance distribution of total wins for 30%-70% winrate:
win
30.0%
35.0%
40.0%
45.0%
50.0%
55.0%
60.0%
65.0%
70.0%
0
34.3%
27.5%
21.6%
16.6%
12.5%
9.1%
6.4%
4.3%
2.7%
1
30.9%
28.8%
25.9%
22.5%
18.8%
15.0%
11.5%
8.4%
5.7%
2
18.5%
20.2%
20.7%
20.2%
18.8%
16.5%
13.8%
10.9%
7.9%
3
9.3%
11.8%
13.8%
15.2%
15.6%
15.2%
13.8%
11.8%
9.3%
4
4.2%
6.2%
8.3%
10.2%
11.7%
12.5%
12.4%
11.5%
9.7%
5
1.8%
3.0%
4.6%
6.4%
8.2%
9.6%
10.5%
10.4%
9.5%
6
0.7%
1.4%
2.5%
3.9%
5.5%
7.1%
8.4%
9.1%
8.9%
7
0.3%
0.6%
1.3%
2.2%
3.5%
5.0%
6.4%
7.6%
8.0%
8
0.1%
0.3%
0.6%
1.3%
2.2%
3.4%
4.8%
6.1%
7.0%
9
0.0%
0.1%
0.3%
0.7%
1.3%
2.3%
3.5%
4.9%
6.0%
10
0.0%
0.0%
0.1%
0.4%
0.8%
1.5%
2.6%
3.8%
5.0%
11
0.0%
0.0%
0.1%
0.2%
0.5%
1.0%
1.8%
2.9%
4.2%
12
0.0%
0.0%
0.1%
0.2%
0.6%
1.7%
4.0%
8.4%
16.1%
It seems you actually need a bit over 60% winrate to have a >50% chance of 5 wins, which is when the rewards are worth more than the cost.
This is assuming that your win rate is going to be consistent across all matches, which isn't realistic, since I expect opponents will be matched on win/loss count like Arena and Duels. So as a run progresses, your winrate should come closer to 50%. However, I don't expect this effect to be as strong for the first half of the run, where you're trying to get to 5 wins.
It would be nice to know what the actual values are for the {0} and {1} placeholders in the ability descriptions. In particular, the fact that Blazing Song at level 5 still gives +3 heal power (same as level 3) means that it's less interesting to prioritize that as an upgrade.
Holy comp was a good suggestion. I used a variation though with Anduin + Natalie + Kazakus as the starting party in most fights.
I like Natalie in combination with Anduin because with her 3rd equipment you can alternate between AoE healing and damage every turn. In normal fights you can keep her out of sync with Anduin's Holy Nova, so you have an AoE heal every turn, while in the boss fight you can keep their AoE damage in sync to dispose of the summoned skeletons.
Kazakus I mainly added to the party because I hadn't unlocked his equipment yet and that required defeating the same heroic boss. Summoning taunt golems repeatedly was pretty useful though, especially after I got a reward dealing 10-damage AoE every time a golem died.
I used Appetizers as the equipment for Cookie instead of Seasoned Pan, because I'm less worried about doing damage and more worried about a merc getting sniped before they can be healed. As I never actually put Cookie on the board in my run, I think it was the right call.
For me, Prime Gaming says the legendary card unlock ends in 22 days, which would be the 25th of May.
Do you know when the refunds for the last round of nerfs end? I have a few Sunken packs that I haven't opened yet because I'd like to claim the golden epics from the reward track first, but if it takes me too long to reach level 75 it might be better to open them anyway in case I'll unpack nerfed cards.
Reverting Kael'thas should be easy since it doesn't impact their e-sports and doesn't require new card text translations. However, as long as Switcheroo is a problem too, that revert wouldn't actually fix the Wild ladder.
Banning Switcheroo from Wild, even as a temporary measure, might still be better though than keeping it active until a long-term fix is implemented.
Switcheroo is broken in Wild; maybe the solution is not to nerf that card though but remove cheap Charge minions from the pool instead to avoid similar super quick high-roll decks in the future
Pufferfist: just too much power + stats for its mana cost
Defias Cannoneer (one shot or higher cost) and/or Mr. Smite (battlecry instead of aura): I'd like Pirate Warrior to be toned down a little; maybe not strictly needed from a balance point of view, but I'd rather play against new decks
Wildheart Guff: while I like this card, I can't deny that if a 5-mana card is an auto-keep off the mulligan, it's probably too strong
personal pick: any card in Quest Hunter, since I like playing minions and this deck punishes you for doing so
I had Gruul + Anduin + Natalie, since I was trying to complete a "heal 900 damage" task at the same time. Probably not the quickest approach, but I got it done in 21 turns, so well below the 30 turns allowed. Gruul makes it so easy, since you deal lots of damage while taking very little with the fire resistance item equipped.
The same comp worked for the second task (keep all your mercs alive) as well and was actually a better fit there.
The other three characters were Brightwing (the one that had the heal task), Wrathion (mainly hoping to collect some coins) and The Lich King (more healing, taunt + freeze gives healers more time to patch up a focused merc, Human synergy with Anduin). But I barely used these, it was Gruul + Anduin + Natalie doing most of the work.
You're right, using a centralized store is easier than using a blockchain. There's also huge speed advantages to using a traditional database. Though, once validated, folks should be storing these validations, for at least some period of time, in a database or cache to help with latencies.
You can keep a cache of filtered and indexed history to do quick lookups, but you can't solve the latency between the time a transaction is submitted for inclusion on the blockchain and the time consensus is reached about it through caching. Of course a game company could hide that latency by immediately giving you access to an in-game item you bought or earned and then settle the transaction in the background, but then there is again the question why they're using a distributed ledger at all.
My biggest problem with people who are anti-crypto/anti-nft is for the most part, they just simply are not informed. You see plenty of folks on Twitter (I know, a horrible example of logical individuals) who won't even for a second make an attempt to learn about them. They all scream YoUr KiLlInG tHe EnViRoNmEnT at the top of their lungs, meanwhile, I'm sure they've just Uber Eats'd their third meal for the day.
Twitter isn't a suitable medium for discussion, but for some reason people keep trying.
I have spent some time learning about blockchain tech, but the more I understand the tech, the less I understand the market. Which probably means that many of the people heavily invested in the market don't understand the tech.
I really like that you can own digital items via an NFT. As things get even more digital, having "proper ownership" is a really cool idea. It makes player-run economies super easy and the company that runs the game doesn't need to run a marketplace, but they can still get paid via % cuts of each sale of an NFT, meaning they don't have to worry about the financial risks that come alongside with running something like the Diablo 3 Auction House or the Steam Marketplace - a huge barrier to smaller companies when you need to worry about dealing with tax for every single player. Not to say I'd want to mint every single item I owned since there would be fees with doing so, but if I got a really cool mount drop in WoW that could be minted, you bet I'd mint my Ashes of Alar!
At their core, NFTs are a mechanism for artificial scarcity. Sometimes you need scarcity to help creators make money, that's why copyright was invented for example. But in other cases the scarcity is just rent-seeking, where companies try to extract as much money as they can with no relation to the cost of creation. Using copyright as an example again, having a 10 to 20 year copyright term gives authors enough time to earn their money, the fact that copyright terms keep getting extended way beyond that is just corporate greed.
I'm a big fan of open source software, which is the opposite of scarcity: people contributing to something that everyone can benefit from for free. That brings its own set of problems, with people taking on big responsibilities in maintaining software while often not getting compensated for that. But overall I think that being able to duplicate digital works is a very unique and useful property; bringing artificial scarcity into such a system is not always a positive.
I also don't think that everything needs to be an economy. Games are supposed to offer a way to leave the real world behind for a while and experience something different. If more and more games not only get in-game economies, but also have those in-game economies tied to real-world economies, every game world becomes some kind of projection or re-skin of the real world, instead of existing outside of it.
Another cool part that I like about the whole NFT portion with games is having potential for items to be useful across other games. There's a bunch of interesting legal issues that need to be figured out before that actually happens, but if I did get Ashes of Alar in World of Warcraft and another game company who really wanted to pay tribute to WoW made some sort of Phoenix pet, maybe that was given out for free to players who validated that they owned an Ashes of Alar NFT. It's not the greatest example but it's a showcase into how different companies could pay homage to each other.
You already mentioned legal issues, but that's just one of the obstacles. From a technical perspective, an NFT is only a proof of ownership: it doesn't tell you what an item would look like or how it would function inside a different game. For each additional game it can appear in, the item would have to be manually re-created. That's a lot of effort and I think developers are likely more motivated to spend their time adding items specifically designed for their own game.
There are cases where it would make sense, for example when taking items from a game to its sequel in the same series, in particular for sports games with annual releases that are mostly roster refreshes. However, in those cases publishers would rather sell the same items to players again and again instead of transferring them over. We know that because that's what they're currently doing with their centralized databases as well.
NFTs need to have value though. The JPEG trash that we see in 99.9% of them right now is hilarious. The NFTs which grant access to a secret club is cool, though when those clubs don't actually do anything, that's basically just a JPEG. There's a couple of neat clubs that do special events together and being a part of something exclusive like that is nice. It's like a digital version of a secret cigar lounge that only the best get access to.
And not even actual JPEGs, but links to JPEGs. Links to locations that might not be all that permanent.
I have some doubts though that money is a good way to measure merit. Sure, some people get rich because they're talented and hard-working, but there are also quite a lot who get there through luck or bad behavior.
Real world, I'd love to see blockchain voting honestly. There's a lot to figure out to make that really happen, with the big hurdle being most people would never understand it so realistically, why even bother, but it would be one of the things we have right now that would make sense and it would make the stats more open. Big privacy concerns though, harassment potential, and compromised private keys are a negative. I like to think if done right, we'd have a better audit trail than what exists now. Now, a malicious actor could pretty easily get setup at the polling station as a volunteer and shove some extra ballets into the box (I'm sure it has been done somewhere). Though, I also don't think that we desperately need blockchain voting either.
Privacy is an issue with blockchain in general. Many people think it's anonymous, but it's very much the opposite of that. You only have privacy for as long as you can keep your wallet disconnected from your identity, but if blockchain is to play a significant role in our future lives, that will be hard to manage.
Another issue is spam: if the transaction costs become low enough, we'll need mechanisms to prevent unwanted items from being sent to our wallets. Otherwise you'd have it fill up with junk or you'll have to pay yourself to transfer the junk out of your wallet again.
Excited to see where it all goes, less excited about the scams that keep plaguing us, and even less excited about good games being bought out by crypto bros because I don't believe we have perfected the use-case for blockchains in games yet.
Despite all my concerns, I do think the tech is exciting. I wouldn't mind seeing the bubble pop tomorrow though; I think all the get rich quick schemes are getting in the way of potential useful applications.
Even with proof of stake, maintaining a distributed ledger is a lot harder than a centralized storage of who owns what. I haven't seen any proposed use of NFTs in games yet that does something that cannot be done easier with a traditional database.
In theory I like distributed technology: the web should become more decentralized if we want to avoid a handful of megacorps running the world. In practice though, blockchain tech is mostly used for speculation and crime, and has solved very few real-world problems.
Maybe after a while the Blood Magic meta was all DH and Druid, but in its first hour I had some really fun games yesterday, like playing Darkest Hour on turn 1 and beating a Hunter with their own cards while playing Thief Priest myself.
I think the main issue was that they made it into a big reveal, creating high expectations that they couldn't meet. If they had announced it with less fanfare, I think most people would have reacted "Sure, I'll give it a try".
Maybe revisit old archetypes that didn't make it in Standard to see if they are playable now with cards released later. For example Beast Druid, Freeze Shaman, Discolock, Handbuff.
In the case of Freeze Shaman, second time around it did make it in Standard based on the new cards alone, so there the question would be whether any of the old freeze cards are good enough to make it into a Wild deck.
In the case of Handbuff, an aggressive Handbuff Paladin saw some play in Wild already, but Hunter and Warrior not so much, neither have I seen much of a more mid-range or combo Handbuff Paladin (I have a partially refined decklist, if you're interested). Also Warlock has handbuff potential; we saw that in Standard for a short time until Imprisoned Scrap Imp got nerfed.
I would also be interested in an article about deckbuilding, since we usually only see the end results, not the process that led to it. I really liked Firebat's Deck Doctor series, for example, or Chump refining Patreon decks. But starting from just an idea instead of a deck list would be even more interesting.
Something I struggle with in particular is how many synergies to include. There are a lot of possible synergies in Wild, but some additions strengthen the deck while others make it too hit or miss and it's often not easy to see which way it is going to go. From the viewer decks I saw in the aforementioned videos, often deck builders try to do too many different things in a single deck. But sometimes it does all fit together and that makes a deck special.
Another difficult decision is how much deck slots to spend on the core of the deck and how many on support (like card draw, removal, healing/armor). Often my first draft will be top-heavy since the coolest cards often have high mana costs and then I have to drop some of them to make sure I can actually survive into the late game to play those cards. Are there some guidelines for balancing such things or is it all trial and error?
Something that I didn't realize until I had built a lot of decks is the relationship between card draw/generation, mana curve and situational/tech cards. For example, in Handlock you can afford to run more situational cards, since you draw a lot of cards and aren't going to play the majority of your hand. If you're mostly going to play cards on curve, you don't need much card draw; Seek Guidance Quest Priest is an extreme example, but also many Elemental, Spiteful and C'thun decks play cards on curve.
Another interesting topic is one-of cards. Some people don't want to see any of those at all (except legendaries), arguing that you should always include two copies of the best card. I think that's too simplistic: the best card often depends on the situation and having two different options in your hand can be superior to having the same option twice. When is consistency actually required to make the deck work versus it being an excuse to build decks that can be auto-piloted?
If you're into math, it might be interesting to do an article on the chances of finding certain cards by a certain turn. For example, what is the effect of including a two-of when running Zephrys or Reno, or a non-dragon when running Kazakusan? How many dragons do you need in your deck to reliably have one in hand by turn X? How many tutored cards do you run relative to the number of tutors: I see 3 tutored + 2 tutors quite often, but is that close to the ideal ratio or does Standard just not have enough options to allow other quantities?
For combo decks, how far into your deck can you expect to find the full combo? MarkMcKz often remarks he's very unlucky to find the last combo piece in the bottom cards of his deck, but since many of his combos require 5 or more cards, that might actually be the statistically expected outcome.
Is the taunt from Niuzao's Bullish Fortitude bugged? Half of the time it is not applied for me. That was before the patch, but since they changed Yu'lon, I was expecting a fix for Niuzao as well.
Your remark made me curious about what winrate you need to break even, so I made a little program to compute the chance distribution of total wins for 30%-70% winrate:
It seems you actually need a bit over 60% winrate to have a >50% chance of 5 wins, which is when the rewards are worth more than the cost.
This is assuming that your win rate is going to be consistent across all matches, which isn't realistic, since I expect opponents will be matched on win/loss count like Arena and Duels. So as a run progresses, your winrate should come closer to 50%. However, I don't expect this effect to be as strong for the first half of the run, where you're trying to get to 5 wins.
Thanks for the explanation, worked like a charm first try!
It would be nice to know what the actual values are for the {0} and {1} placeholders in the ability descriptions. In particular, the fact that Blazing Song at level 5 still gives +3 heal power (same as level 3) means that it's less interesting to prioritize that as an upgrade.
Holy comp was a good suggestion. I used a variation though with Anduin + Natalie + Kazakus as the starting party in most fights.
I like Natalie in combination with Anduin because with her 3rd equipment you can alternate between AoE healing and damage every turn. In normal fights you can keep her out of sync with Anduin's Holy Nova, so you have an AoE heal every turn, while in the boss fight you can keep their AoE damage in sync to dispose of the summoned skeletons.
Kazakus I mainly added to the party because I hadn't unlocked his equipment yet and that required defeating the same heroic boss. Summoning taunt golems repeatedly was pretty useful though, especially after I got a reward dealing 10-damage AoE every time a golem died.
I used Appetizers as the equipment for Cookie instead of Seasoned Pan, because I'm less worried about doing damage and more worried about a merc getting sniped before they can be healed. As I never actually put Cookie on the board in my run, I think it was the right call.
For me, Prime Gaming says the legendary card unlock ends in 22 days, which would be the 25th of May.
Do you know when the refunds for the last round of nerfs end? I have a few Sunken packs that I haven't opened yet because I'd like to claim the golden epics from the reward track first, but if it takes me too long to reach level 75 it might be better to open them anyway in case I'll unpack nerfed cards.
Reverting Kael'thas should be easy since it doesn't impact their e-sports and doesn't require new card text translations. However, as long as Switcheroo is a problem too, that revert wouldn't actually fix the Wild ladder.
Banning Switcheroo from Wild, even as a temporary measure, might still be better though than keeping it active until a long-term fix is implemented.
Nerf candidates:
Faelin, but also Hooktusk and Ini.
Several of these mention an underwater form. Was that an idea that got dropped later in the design phase?
I thought I'd put it muted in a tab just to get the pack, but I've been listening for over half an hour now. Well done!
That Salt bbcode might be useful in the future.
I had Gruul + Anduin + Natalie, since I was trying to complete a "heal 900 damage" task at the same time. Probably not the quickest approach, but I got it done in 21 turns, so well below the 30 turns allowed. Gruul makes it so easy, since you deal lots of damage while taking very little with the fire resistance item equipped.
The same comp worked for the second task (keep all your mercs alive) as well and was actually a better fit there.
The other three characters were Brightwing (the one that had the heal task), Wrathion (mainly hoping to collect some coins) and The Lich King (more healing, taunt + freeze gives healers more time to patch up a focused merc, Human synergy with Anduin). But I barely used these, it was Gruul + Anduin + Natalie doing most of the work.
You can keep a cache of filtered and indexed history to do quick lookups, but you can't solve the latency between the time a transaction is submitted for inclusion on the blockchain and the time consensus is reached about it through caching. Of course a game company could hide that latency by immediately giving you access to an in-game item you bought or earned and then settle the transaction in the background, but then there is again the question why they're using a distributed ledger at all.
Twitter isn't a suitable medium for discussion, but for some reason people keep trying.
I have spent some time learning about blockchain tech, but the more I understand the tech, the less I understand the market. Which probably means that many of the people heavily invested in the market don't understand the tech.
At their core, NFTs are a mechanism for artificial scarcity. Sometimes you need scarcity to help creators make money, that's why copyright was invented for example. But in other cases the scarcity is just rent-seeking, where companies try to extract as much money as they can with no relation to the cost of creation. Using copyright as an example again, having a 10 to 20 year copyright term gives authors enough time to earn their money, the fact that copyright terms keep getting extended way beyond that is just corporate greed.
I'm a big fan of open source software, which is the opposite of scarcity: people contributing to something that everyone can benefit from for free. That brings its own set of problems, with people taking on big responsibilities in maintaining software while often not getting compensated for that. But overall I think that being able to duplicate digital works is a very unique and useful property; bringing artificial scarcity into such a system is not always a positive.
I also don't think that everything needs to be an economy. Games are supposed to offer a way to leave the real world behind for a while and experience something different. If more and more games not only get in-game economies, but also have those in-game economies tied to real-world economies, every game world becomes some kind of projection or re-skin of the real world, instead of existing outside of it.
You already mentioned legal issues, but that's just one of the obstacles. From a technical perspective, an NFT is only a proof of ownership: it doesn't tell you what an item would look like or how it would function inside a different game. For each additional game it can appear in, the item would have to be manually re-created. That's a lot of effort and I think developers are likely more motivated to spend their time adding items specifically designed for their own game.
There are cases where it would make sense, for example when taking items from a game to its sequel in the same series, in particular for sports games with annual releases that are mostly roster refreshes. However, in those cases publishers would rather sell the same items to players again and again instead of transferring them over. We know that because that's what they're currently doing with their centralized databases as well.
And not even actual JPEGs, but links to JPEGs. Links to locations that might not be all that permanent.
I have some doubts though that money is a good way to measure merit. Sure, some people get rich because they're talented and hard-working, but there are also quite a lot who get there through luck or bad behavior.
Privacy is an issue with blockchain in general. Many people think it's anonymous, but it's very much the opposite of that. You only have privacy for as long as you can keep your wallet disconnected from your identity, but if blockchain is to play a significant role in our future lives, that will be hard to manage.
Another issue is spam: if the transaction costs become low enough, we'll need mechanisms to prevent unwanted items from being sent to our wallets. Otherwise you'd have it fill up with junk or you'll have to pay yourself to transfer the junk out of your wallet again.
Despite all my concerns, I do think the tech is exciting. I wouldn't mind seeing the bubble pop tomorrow though; I think all the get rich quick schemes are getting in the way of potential useful applications.
Even with proof of stake, maintaining a distributed ledger is a lot harder than a centralized storage of who owns what. I haven't seen any proposed use of NFTs in games yet that does something that cannot be done easier with a traditional database.
In theory I like distributed technology: the web should become more decentralized if we want to avoid a handful of megacorps running the world. In practice though, blockchain tech is mostly used for speculation and crime, and has solved very few real-world problems.
Maybe after a while the Blood Magic meta was all DH and Druid, but in its first hour I had some really fun games yesterday, like playing Darkest Hour on turn 1 and beating a Hunter with their own cards while playing Thief Priest myself.
You'd still get 6 damage + 6 healing for just 1 damage. Not broken, in my opinion, but still good. However, they banned the card anyway.
I think the main issue was that they made it into a big reveal, creating high expectations that they couldn't meet. If they had announced it with less fanfare, I think most people would have reacted "Sure, I'll give it a try".
Maybe revisit old archetypes that didn't make it in Standard to see if they are playable now with cards released later. For example Beast Druid, Freeze Shaman, Discolock, Handbuff.
In the case of Freeze Shaman, second time around it did make it in Standard based on the new cards alone, so there the question would be whether any of the old freeze cards are good enough to make it into a Wild deck.
In the case of Handbuff, an aggressive Handbuff Paladin saw some play in Wild already, but Hunter and Warrior not so much, neither have I seen much of a more mid-range or combo Handbuff Paladin (I have a partially refined decklist, if you're interested). Also Warlock has handbuff potential; we saw that in Standard for a short time until Imprisoned Scrap Imp got nerfed.
I would also be interested in an article about deckbuilding, since we usually only see the end results, not the process that led to it. I really liked Firebat's Deck Doctor series, for example, or Chump refining Patreon decks. But starting from just an idea instead of a deck list would be even more interesting.
Something I struggle with in particular is how many synergies to include. There are a lot of possible synergies in Wild, but some additions strengthen the deck while others make it too hit or miss and it's often not easy to see which way it is going to go. From the viewer decks I saw in the aforementioned videos, often deck builders try to do too many different things in a single deck. But sometimes it does all fit together and that makes a deck special.
Another difficult decision is how much deck slots to spend on the core of the deck and how many on support (like card draw, removal, healing/armor). Often my first draft will be top-heavy since the coolest cards often have high mana costs and then I have to drop some of them to make sure I can actually survive into the late game to play those cards. Are there some guidelines for balancing such things or is it all trial and error?
Something that I didn't realize until I had built a lot of decks is the relationship between card draw/generation, mana curve and situational/tech cards. For example, in Handlock you can afford to run more situational cards, since you draw a lot of cards and aren't going to play the majority of your hand. If you're mostly going to play cards on curve, you don't need much card draw; Seek Guidance Quest Priest is an extreme example, but also many Elemental, Spiteful and C'thun decks play cards on curve.
Another interesting topic is one-of cards. Some people don't want to see any of those at all (except legendaries), arguing that you should always include two copies of the best card. I think that's too simplistic: the best card often depends on the situation and having two different options in your hand can be superior to having the same option twice. When is consistency actually required to make the deck work versus it being an excuse to build decks that can be auto-piloted?
If you're into math, it might be interesting to do an article on the chances of finding certain cards by a certain turn. For example, what is the effect of including a two-of when running Zephrys or Reno, or a non-dragon when running Kazakusan? How many dragons do you need in your deck to reliably have one in hand by turn X? How many tutored cards do you run relative to the number of tutors: I see 3 tutored + 2 tutors quite often, but is that close to the ideal ratio or does Standard just not have enough options to allow other quantities?
For combo decks, how far into your deck can you expect to find the full combo? MarkMcKz often remarks he's very unlucky to find the last combo piece in the bottom cards of his deck, but since many of his combos require 5 or more cards, that might actually be the statistically expected outcome.
Ah, that blocks your own taunt as well? In that case, that likely explains it, since I'm leveling up both of them in the same party.
Is the taunt from Niuzao's Bullish Fortitude bugged? Half of the time it is not applied for me. That was before the patch, but since they changed Yu'lon, I was expecting a fix for Niuzao as well.