Thread by @drewlevin

Seeing a bunch of tweets about how Alchemy on MTG Arena is broken because card changes don't refund wildcards despite card redemptions creating "true" versions + Alchemy versions. Also seeing a bunch of suggestions about how to fix it. Taking a shot at a thread on all that. 1/

https://twitter.com/drewlevin/status/1471342321538977793

Before we start: who am I to go into this? I've played Magic for over a decade, Arena since launch, drafted my way to full sets, and have worked in free-to-play games product, analytics, and monetization for the last 6-7 years. I have player and business perspectives. 2/

Here are my big arguments:
(a) Magic is a pay-to-play game that brought that business strategy into a free-to-play market. The free-to-player’s Constructed gameplay experience doesn’t match the marketing. Lots goes into this. 3/

(b) If WOTC was interested in making a better free-to-play experience, they would reweight toward time spent and away from skill as a driver of outcomes. 4/

(c) Alchemy - and the wildcard fracas around it - exacerbates the Constructed-player economy and experience issues inherent in Hasbro’s business model. 5/

(d) WOTC and Riot spend time and money on vastly different things, and so they end up selling different things. Arena and Runeterra are not going to converge on business models, nor should they. 6/

Starting with point A: Magic spent decades teaching people that 15 cards costs $3.99. Their internal accounting likely measures against that. MTGO’s economy is built around that. Arena’s economy is also built around that. 7/

The game “prints” a very small amount of money per day and gives it to people who complete their daily quests. If you win a skill-based reward, you’re getting more at the expense of the people you beat who are getting less. 8/

Your collective entry fees paid for the cards you’re getting. That’s the accounting. The draft experience is thus a close 1:1 with other incarnations: you all pay a big entry fee that funds a pot of packs for drafting + rewards. Win more, get more of the pot. 9/

The Constructed experience is not as simple. The real-world model that people are used to is the Format Challenge tournament: pay a lot for tournament entry into a top-weighted payout structure. 10/

This has some problems, mostly around how people are meant to build a collection for this in the first place. There are also $1 and $2 entry tournaments that pay back in ICRs + soft currency, but that’s not a meaningful way toward building your collection. 11/

The big deviation from real world Magic is The Ladder, that stalwart of digital CCGs. Typical execution is a monthly divisional leaderboard that gets harder to go up as you get higher + month-end rewards based on what division you attain. 12/

Still, these are monthly rewards, not daily or weekly rewards. You’re not going to build a proper collection from making Mythic each month, nor will you be able to make Mythic each month without alternate sources of investment. So how do you actually make this work? 13/

The answer is simple: just like in real life, you buy the cards you want. Eight-card packs cost ~$1 each, so you can get ~100 packs + 16 rare+ wildcards for $100 instantly. 14/

If you don’t want to buy packs, you can limp around with a precon-quality deck that has a nominal number of rares and mythics and mostly lose somewhere between Silver and Diamond. 15/

This may match the FNM player’s rough experience, minus all of the other on-ramps into Constructed magic (buying precons instead of packs, people loaning you cards, etc). 16/

It doesn’t give you access to one of the most fun parts of Magic: building multiple decks, metagaming, getting to play with different strategies, all that. I think that this is a fundamental failure of imagination from Magic’s digital studio side. 17/

I’ll return to this with the Alchemy section, but I would argue that there’s more to gain from converting nonpayers who are currently shut out of the deck/meta loop than there is risk on top spenders who might spend less if you give them more stuff for free. 18/

The rough thing is that it’s impossible to experiment on this front given the nature of the community. How to validate? 19/

I would spend a full set running a series of precon-driven events that are designed as economy sources + selling precons at a discount/renting + pushing participants toward limited-time offers that help them improve/complete their precons and ramp up into competitive play. 20/

Absent that on-ramp to the top of the aspirational system, my thesis is that many people simply don’t and won't try to Care about Constructed. 21/

For top payers, probably they’ll buy the precons because they’re a good rate, but they’ll care far more about League-equivalents (more on this in a sec) and other status signifiers. 22/

Even if you miss high on value, it’s going to be a $5-10 price point, so you’re creating something that behaves like a Battle Pass – hardly a terrible outcome. 23/

To summarize: the current Constructed collection experience falls short in the face of wanting to build one or more properly competitive decks each set. If you want to build constructed decks, you need to play with paid-for packs. 24/

If you’re not buying your own packs, you must get other people to pay for them. And so let’s go to point (b). 25/

To revisit point B: If Hasbro was interested in making a better free-to-play experience, they would reweight toward time spent and away from skill as a driver of outcomes. 26/

The quality of your experience and the rewards you get today scale directly with how skilled you are at the game. This is true in Constructed with Format Challenges, and it’s true in Draft. I don’t track my draft win rate, but let’s say I’m good enough. 27/

I have to spend maybe $100 a year to draft as much as I want + complete rare sets if I put in the time. For other people, that’s enough to get exactly 13 drafts.

The second set of people are quite literally buying me cards. 28/

That is Hasbro’s business model, and it’s why drafting is more rewarding per-minute than playing Constructed: if you win a lot, you can apply other peoples’ dollars as a force multiplier on your collection progress. 29/

By contrast, Constructed matches don’t generate or consume anything, and so the value of those tournaments must be lower (or the cost must be higher to allow for a larger rake + juicier payouts, e.g. Arena Opens and Format Challenges). 30/

All of this goes back to point A, which is that Arena isn’t actually a free-to-play game design, it’s an onramp to a payer experience with no trade function. In other words, it’s MTGO with a bunch of middlemen removed. 31/

Is it functional? Yes, but I would argue it leaves a lot more money on the table than WOTC appreciates. If you don’t want to Draft, there isn’t a time-economical way to build a collection. 32/

You can either take shots at 30-pack 7-win payouts or play in the $1-2 entry kiddie pool with no middle ground. 33/

By contrast, FNMs - likely the modal sanctioned Constructed experience for Magic players - are $5-6, give everyone a pack (i.e. 2/3s of your entry goes back to you), offer 3-4 matches, and reward top performers with more packs. 34/

I’m confident that Arena’s monetization and live ops designers can figure out a Constructed tournament format that feels worth playing, both in terms of time in and rewards out. Free idea: run end-of-format events that pay out in next-set packs. 35/

Of course, then you run into the buzzsaw of “why do all that if everyone’s just going to play on the Ladder?” Well, about that… 36/

My proposal to the Arena team would be to position these flatter-payout, mid-priced tournaments as competitions where skill matters AND where you can make progress against the core Constructed aspiration of Mythic + status. 37/

I’d start with the Leagues model from MTGO and tweak from there – it’s clear that Ladder has some shortcomings in terms of both economy inflows (doesn’t create any) and outflows (doesn’t feel rewarding enough). 38/

Mythic also doesn’t seem particularly interesting or fun once you’re up there, but I will admit to being lower-first-hand-info on that front. 39/

So now that we’ve guided people toward a positive deckbuilding experience through selling something that WOTC has proven competency on (precons) and given people flatter, more rewarding play structures, let’s talk about Alchemy and point C. 40/

Point C is that Alchemy - and the wildcard fracas around it - exacerbates the Constructed-player economy and experience issues inherent in Hasbro’s business model. 41/

A lot of what I've seen re: Alchemy is enthusiasm from Constructed hardcores (fresh formats, new content, yay!) as well as frustration about how Arena's wildcard model interacts with a format that proposes to churn through metagames faster. 42/

The metagame/format freshness component is welcome and in line with stated goals, and obviously from a business perspective it makes sense to create sets where people need to buy a greater % of rares over the life of the set - that's how packs and WCs monetize, on breadth. 43/

The “problem” with Alchemy is that it’s going to be really, really good at sinking wildcards through creating more consumable content via card rebalancing. To be clear, that’s also the business opportunity, its raison d’etre. 44/

But absent a valid aspirational pathway to being competitive and deck-swapping in the first place, Alchemy looks like a really fun theme park that you can’t afford to get into. 45/

After all, if you weren’t getting to play multiple decks in the slower metagame, you DEFINITELY aren’t doing that in the faster one. 46/

A big conversation inside the Magic community in recent days is whether Alchemy should, upon nerfing cards, effectively refund players those cards. That point is what inspired this thread. 47/

A bunch of well-meaning people argued that WOTC should, and my point is that doing so would invalidate the business model of the entire format, which is to get people to Buy More Cards. Refunding people cuts against the entire thesis – you can’t do that. 48/

You CAN make it easier for people to go from the cheaper, slower format to the faster, more expensive format, and it’s no surprise to me that they’ve done exactly that by giving people both "versions" of a card when they get one. 49/

The reason people are mad is it’s hard to build their first deck, so getting it nerfed is painful. Solve: make the first 1-2 decks in a format easier to build + keep $ upside of selling complete set + create context where that’s a common desire -- where Alchemy is going. 50/

With the Alchemy point out of the way, I do want to close by comparing Magic and LoR, mostly as a way of making it clear why these comparisons are unwarranted. 51/

Point D is that Hasbro and Riot spend time and money on vastly different things, and so they end up selling different things. Arena and Runeterra are not going to converge on business models, nor should they. 52/

Let’s zoom out for a second. What is the primary thing that Magic makes and sells? I’ll give you a second. 53/

The answer is cards. You can tell because that’s where they’re spending the vast majority of their capital. 54/

Set, format, and card design, manufacturing, production, packaging, supply chains, all of it so that they can sell cards, which they design at a furious clip and have dedicated a lot of resources to. That is their core product. 55/

The business rationale for their digital products can be encapsulated as “we spend all of this effort making these cards - what if we could sell them again?” And that’s what MTGO and Arena do: they sell cards. That is their business model. 56/

Every few months, Magic releases another set that changes the face of a format and creates many different experiences centering on those new cards. The cards are the stars here, always. 57/

Let’s zoom back out and try this again with Riot. What does Riot sell? Take a moment. Think about what they put front and center in their storefronts. 58/

You already know the answer. It’s cosmetics. And if you look at Runeterra’s production processes and merchandising strategy, it’s clear where they’re focused. 59/

As a game, it’s a horizontal extension of League of Legends, a massively-popular character-driven IP with a ton of revenue driven by character skins. After all, you can sell people Garen only once, but you can sell skins for Garen many, many times. 60/

You can see how this plays out in their draft format. I got back into Runeterra relatively recently, drafted a few times, and remembered pretty quickly that Riposte is completely busted. Totally cracked. 61/

The Runeterra team is making new content, but their draft format’s best common (IMO) is the same thing that it was a year ago. For Magic, I would have missed multiple formats and have to relearn all the cards. 62/

Here, I had to pick up a few new regions and mechanics, but I was basically good to go upon returning. The emphasis is on the cosmetics (and the characters that drive them) rather than on the cards themselves. 63/

They’re not trying to sell me a bunch of packs that are filled with draft chaff so I have a harder time getting to my preferred Constructed rares, and so they’ll happily sell me the exact cards I want. Why? 64/

Because they know that most of their revenue comes after I attach to a favorite champion, and they resource their teams and release content and drive player aspiration and frame rewards and status along those lines. 65/

Because they’re selling deep investment in characters and IP, they want you to find your favorite champion and go deep. Magic wants you to keep buying new decks, because it’s running a content game rather than a cosmetic game. 66/

All of that means that “just let me buy exactly the cards I want with no inefficiency functions” is willful ignorance of what Magic’s business model actually is. Please stop comparing the two. 67/

Magic will make a ton of money off of you trying to get every card. Runeterra will make their money off of you buying $10-20 cosmetics and sell/give you a bunch of cards so you play enough to form attachments to their IP that drive that kind of cosmetic-purchasing behavior. 68/

That doesn’t mean that Magic couldn’t do this – I think they have a very strong character-driven IP with an awesome cast of characters... 69/

...but their current cosmetic executions are fairly limited and they can’t justify charging that much for 2D character art, which is what their avatars are. /70

Their best execution to date is board-side pets, which probably crush it when cute/when animation knocks it out of the park. But they're not - and won't be - planeswalkers. 71/

Figuring out how to sell Garruk or Liliana or Chandra or Jace cosmetics to people who want to pay a lot to show how much they love those characters is a summit they have yet to entirely reach, but I’d bet a lot that they’re trying. 72/

I didn’t expect this to turn into a 73-tweet thread, but here we are. Many thanks to Kate for bouncing ideas around, and good luck to everyone in navigating your relationship to money, Magic, and hobbies. 73/73

Back to the top of the thread:

https://twitter.com/drewlevin/status/1471370439645548546?s=20