Guest Post: Dragonflight Alchemy

This article on Dragonflight Alchemy was graciously written by Beta and ThetaJay!

Alchemy in early Dragonflight

1: The Concept

Like all other professions in Dragonflight, alchemy has had a complete overhaul. However, perhaps unique to alchemy is the introduction of rafted intermediate reagents, namely Omnium Draconis and Primal Convergent. Both of these reagents are used in crafting the top tier potions and phials.

Due to the limitations of gaining profession knowledge in the early weeks of the expansion, an alchemist could theoretically only be specialized in one of the three following categories early on:

  • making intermediate reagents;
  • making potions;
  • making phials.

By having access to all three specializations at the same time, it would be possible to have a complete crafting pipeline without having to rely on a middle man, only the purchase of gathered materials from the auction house. While this was our initial idea, when executing the plan we discovered many ways of optimizing the pipeline to maximise potential profit.

The first task at hand was to plan how we would allocate the limited knowledge points for each specialization. At this point it is worth discussing how the inspiration node is designed to work. With a maxed out inspiration node, an inspiration proc will always provide the exact number of skill points required for a full-quality upgrade. For example, the skill required for making a quality 2 Phial of Elemental Chaos is 225, and for quality 3 one needs 450 skill. A full inspiration node provides exactly the 225 extra skill points required to bridge the gap between quality 2 and quality 3. This leads to the following key observation when assigning knowledge points:

The most efficient way of spending early knowledge is to spend points until quality 2 crafts are guaranteed, and then the remaining points are assigned to maxing out the inspiration node.

1.1: Building the intermediate reagent crafter

Specialization for intermediate reagent making was simple:

  • 15 points in Alchemy Theory;
  • 40 points in Chemical Synthesis;
  • 30 points in Inspiring Ambience.

This was achieved within the first week of Dragonflight very comfortably. Additional points over the following weeks then went into Resourceful Routines.

Importantly, with this build one is able to use all quality 2 herbs to guarantee making quality 2 Omnium Draconis. Further, one can check that it is impossible to craft quality 3 Omnium Draconis without inspiration, so this rendered quality 3 herbs completely useless for this build.

1.2: Building the tertiary crafter

Our original plan for potion and phial crafting was naive. We planned to max out Batch Production and Inspiration, and then use a mix of quality 2 and 3 Draconium Vial, Omnium Draconis, and Primal Convergent to guarantee making quality 2 potions and phials with inspiration procs allowing us to craft quality 3.

However, the whole situation changed at 1 AM on December 5th, when we realized that it was actually possible to use only quality 2 materials to hit exactly 213 skill points for potion crafting, and 225 skill points in phial crafting, which is the skill required to craft quality 2. Even with finding these extra skill points, we were still able to obtain maximal inspiration and multicraft, allowing us to inspire into making quality 3 finished products. We call this the 2/2/2 build. The completed build for potion crafting was achieved on week 1 of release, and the build for phial was achieved on the reset of week 2 due to the extra 12 knowledge points required.

The finished builds ended up as follows:

  • 10 points in Potion (resp., Phial) Mastery;
  • 10 points in Potion (resp., Phial) Lore;
  • 20 points in Batch Production;
  • 5 points in Alchemy Theory;
  • 30 points in Inspiring Ambience.

Depending on profession tools and skill level, one needs some extra skill points to obtain the true 2/2/2 build. For example, for phial creation this was achieved by allocating excess knowledge to Alchemical Theory to gain access to decay phials for further first time crafts.

1.3: The overall plan

After the builds were online, our plan for alchemy was as follows:

  • Buy quality 2 herbs and awakened elements from the auction house.
  • Craft Omnium Draconis and Primal Convergent.
  • Sell all quality 3 Omnium Draconic and Primal Convergent obtained from inspiration procs on the auction house.
  • Use the quality 2 Omnium Draconis and Primal Convergent to craft potions and phials, and then sell them on the auction house.

Let us make a few comments on this plan.

  • Since quality 3 Omnium Draconis and Primal Convergent is not required for us to make potions and phials, we sell all of them. This action has two advantages.
    • Anyone who buys them immediately loses in the market, since they are inherently making consumables at a higher cost.
    • This is essentially a hedge against the price movement of herbs and awakened elements.
  • As more and more people realise that a 2/2/2 build as described above is possible, inspiration on intermediate reagents will slowly lose its value. Eventually, quality 2 and quality 3 intermediate reagents should effectively have the same price. As such, at some point resourcefulness will outperform inspiration on intermediate reagents, and we need to prepare for that.
  • A big, but overlooked advantage of this plan is that we only use the auction house twice:
    • Buying quality 2 herbs and awakened elements;
    • Selling quality 3 intermediate reagents and all end products.

    If someone instead relies on buying Omnium Draconis and Primal Convergent from the auction house for crafting their potions/phials, they are paying an extra 5% of the auction house commission fee to do so.

In what follows we will describe how we managed to achieve this build, the math behind the trading, and some thoughts for the future.

2: Grinding the profession

There were two essential grinds for the execution of the plan: knowledge points and artisan’s mettle.

The need for knowledge points is evident with the builds described in the first section. For the reagent making we needed access to 85 points, for potion making the end number came in at 100, and for the phial crafter we needed to squeeze out 110 knowledge points.

Knowledge grinding is a standard concept to most of us now, and of course we heavily recommend the guides on WoWhead. However, we point out two overlooked ways of squeezing out essential knowledge points that people may not consider:

  • It is possible to buy recipes from the auction house and reputation vendors to get the first craft bonus;
  • In the first weeks of Dragonflight it was possible to profession shuffle to hit 4/5 reputation with the Artisan’s Consortium for an extra 10 knowledge points.

Artisan’s mettle is important for two reasons. Firstly, profession gear requires mettle to craft, and the skill bonuses from these were essential to see the builds through. If you are on a high-pop server, then you probably had access to all rare professional equipment on week 2, which makes things a lot easier. But unfortunately, this is not the case for us (EU-Doomhammer), so we had to rely on some harder grinding to see the build through.

Secondly, and especial important for the phial maker, artisan’s mettle is required for experiments to discover new recipes and to obtain skill points. The recipe for the phial that we wished to craft could only be obtained through these experiments.

After our realization that a 2/2/2 build for potion and phial making was possible, we spent a lot of time and energy preparing the builds (especially in the case of the phial crafter which required extra skill points) before the first EU reset so that we could capitalise on the potentially novel build. And we were blessed by the RNGod for getting the Phial of Elemental Chaos recipe two days after we started. The grind, of course, is something that paid off in the end, else we would not be here writing an article about it!

3: A spreadsheet

On the Saturday before the raid release, we began testing the build. It was then that we realised that there were even more small optimization that could be made. For example, by using sagacious incense (which sells extremely cheaply on the auction house) it is possible to have an extra 2% inspiration, however, was it worth doing?

We immediately realized that we needed to be able to rapidly estimate all of our costs to make the right decision, especially when deciding how to weight our profit margins and ratios of potion to phial crafts. The solution to this problem was a simple one in theory, we sat down for a few hours on Saturday morning and made a spreadsheet that would calculate everything we required.

The spreadsheet takes a few inputs:

  • Market price of herbs and awakened elements.
  • Market price of quality 3 Omnium Draconis and Primal Convergent.
  • Market price of quality 2 Draconium Vial.

We then needed to consider how the three secondary stats: inspiration, multicraft, and resourcefulness should be encoded into the spreadsheet. Notice that for every 10 points of secondary stat it gives 1.1% of resourcefulness, 1%of inspiration, and 0.9% of multicraft respectively.

  • Inspiration is quite straightforward. When it procs, you get quality 3, otherwise, you get quality 2. So in addition to the above inputs, we also take the market price of quality 2 and 3 potions/phials into consideration. Since we want to undercut them evenly, we use the ratio of prices of quality 2 and quality 3 potions/phials on the market to determine our price.
  • Multicraft required some trials to pin down. After about 10k crafts, we are confident that if you spend all knowledge points in multicraft, each proc on average gives you an extra 200% of products. For example, each multicraft potion proc gives you on average 15 potions in total (5 baseline and 10 from multicraft). So if your multicraft is 10%, you are expecting a 120% productivity.
  • We simply ignored resourcefulness in the beginning. But some trials tell us that without spending points in resourcefulness, one proc gives you about 0.3 worth of a craft, and the number becomes 0.45 once you fully spec into the tree. For example, if you spec fully into the tree and have 10% resourcefulness, then after 1000 crafts on average you will save mats for another 1000 * 10% * 0.45 = 45 crafts.

We quickly noticed that multicraft is usually better. Inspiration beats multicraft only when the quality 3 price is more than 3 times of the quality 2 price (2.3 times if you don’t spec into the multicraft node). Inspiration beats resourcefulness if the quality 3 price is at least 1.45 times of the quality 2 price (1.3 times if you don’t spec into the resourceful node). And multicraft is always better than resourcefulness. Take this paragraph with a grain of salt, because we are doing linear approximation here. If you want an extremely precise computation to decide which secondary stat you need to go, write down the complete formula of the cost of your pipeline, then compute its partial derivatives on secondary stats. Time to go back to school.

Inspiration is better than resourcefulness almost all the time for potion and phial making, but resourcefulness will outperform inspiration eventually in intermediate reagent making. This comes in handy in choosing enchantments for professional tools. Notice that enchantments for professional tools scale with your character level, so level all your crafters to 70.

It took a few hours for us to be sure that the spreadsheet formulas were correct, with a few mishaps along the way. For example, for a while we ran the spreadsheet without including the 5% auction house fee, something we caught later on when our profit margins seemed a bit too good to be true.

The spreadsheet is not for estimating how much gold we can earn, but for helping us make the decision whether we should trade or not. If we can have a 20% margin, then we can do it casually, if 0% then we just stop, and if 50% we can really spend an hour or two crafting nonstop.

If gold-making via crafting, or even just playing the auction house is something you want to get into, we cannot stress enough how important it is to have some form of a spreadsheet which can quickly tell you the relevant pricing information. It allowed us to craft without any ambiguity or risk.

4: Trading on the AH

Since we are buying a lot of materials and selling six different products (quality 3 Omnium Draconis and Primal Convergent, quality 2 and 3 potions and phials), it involves some interaction with the AH. We are both new to this part of the game, so a lot of mistakes have been made, and a lot of experiences have been learned.

The biggest mistake is the lack of patience, especially in selling quality 3 Omnium Draconis and Primal Convergent. Since they are byproducts, we usually sell them at either the market price or even at a discount. But this market is far more active than we expected.

The second mistake is spending too much time undercutting, especially before the raid release. After a few days of trading, we started to realize that any loss of profit mostly came from time spent not crafting, as opposed to not undercutting. So eventually we just put the consumables at a reasonable margin and stopped worrying about them at all.

Our principle is not speculating, that means we buy mats, craft, and then sell all of the products before we start a new buying round. Ideally, this is an arbitrage, but the price movement is very volatile. Eventually, we found out that by having a small bank of mats, we can make the procedure much faster: we can send Omnium Draconis and Primal Convergent to potion/phial alchemists before we start crafting them (2000 Omnium Draconis crafting takes about 20 minutes, and this was very often the bottleneck in maximising profits).

It was very interesting to see the liquidity and the behaviour of the regional AH. Before the raid release, almost all our sales are in big chunks (it was common to see a single order of 100 to 200 phials). We believe they are either big guilds buying consumables or individual players speculating on price movements. After the release, sales were mostly in small chunks (10-20 per order) which is clearly individual raiders and dungeon runners buying their personal consumables. This results in different trading strategies: Before the raid release, we often undercut the market by 10% of our margin, to make sure we can be easily hit by big orders and then move on. After the raid release, we are happy to do much smaller undercuts because things sell immediately after they are on the shelf.

The price volatility of herbs and awakened elements was insane. A well-timed purchase would result in more than 10% extra margin. However, it naturally was very difficult to predict price movement. We believe that with some data and analysis, a reasonable quantitative model could be built to indicate buying signals.

5: Some final thoughts and the future

Some final thoughts:

  • It would not be a gold-making article if there are no numbers. We each bought a token just before the release of Dragonflight, which at the time gave us 600k total gold on day 0 of the expansion. This grew to 1.5m gold essentially from crafting and gathering while levelling. We launched the production of our build four days before the release of the raid, and we had made 25m gold profit by the time the raid released. Most of time was simply spent AFK crafting. During these days we pretty much always had a comfortable 50% margin on the market which we pretty much dominated (one could visibly see when we placed our products on the regional AH charts). Of course, these profit margins could not last forever, but at the time of writing we have made in the region of 50m profit from this strategy.
  • The pricing spreadsheet was, by far, the most important part of the process. It gave us quick and precise cost estimates under the current market condition, and allowed us to make easy and swift decisions. Our final strategy is rather involved. It buys 9 different mats and sells 6 different products, so simple mental math would not work well in this volatile market.
  • Selling quality 3 Omnium Draconis and Primal Convergent hedges against price movements of materials. If on average the price of herbs and awaken elements go up by 20%, our cost will go up by roughly 15% (depending on inspiration rate and the price difference between quality 2 and 3 Omnium Draconis and Primal Convergent). There were a few times when the potion/phial market overreacted to severe material price movement, and our margin skyrocketed as a result.
  • Since all secondary stats are multiplicative, it is much safer to underestimate their effect (thus overestimate the cost) than the opposite. After all, not trading is better than trading at loss. It also means that an error in estimating one term can cause disaster in the cost and profit approximation.
  • On the other hand, since people inherently fear trading at loss, our precise pricing spreadsheet allowed us to keep trading when other people were cautious. Combined with the high demand in the alchemy market, it gave us very exciting opportunities at times.
  • When our products were selling well, as we have already mentioned the bottleneck to profit is simply crafting speed. Blizzard has capped crafting speed at 75%, which is 4x speed than 0% (standard Blizzard math…).
  • The regional commodity market gives traders in low-pop servers a unique chance to compete, something we are thankful for on our server. Even though the server-based part (especially the crafting order) is still a downside, we are exposed not only to a larger market but also to bigger buyers. We joked a lot about  Gingi being the only buyer for our consumables, there are many Discord messages between us about our entire stock of products being “Gingi’d”.

Although we have described how we set up our build to dominate the market within the first weeks of the expansion, some readers might be surprised to learn that this build is still providing us in the new year with comfortable profit margins and we are still dominating the alchemy regional auction house. Of course, we have had to make some adjustments already to ensure this, and we are actively trying to predict the future of this build. Let us share some of our thoughts here.

  • Understanding all secondary stats certainly not only helps alchemy. Similar calculations can be done in other professions, especially those that require multiple steps.
  • Quality 3 Omnium Draconis and Primal Convergent market will eventually collapse due to them not being needed t craft quality 3 end products. This is already happening to Primal Convergent in the EU. When the price gap is less than 5%, we will use Quality 3 reagents in crafting instead of hedging, thus simplifying our strategy.
  • As an extension of the above, Quality 3 Primal Convergent can be used in making Sagacious Incense and Potion Absorption Inhibitor. That means we can further diversify our strategy by adding these two products (both Quality 2 and 3) into our sale list, something we have already begun implementing to great success.
  • A better understanding of material prices in terms of productivity and demand rather than historical data will greatly improve buying timing of materials, which becomes more and more important as out profit margin decreases.
  • By pushing our knowledge gain every week, we are still finding new edges. We have already described the 2/2/2 build, but as of now we are actually working with a 1/2/2 build, where the quality 1 material is the Draconic Vials. Quality 1 Draconic Vials are simply bought from the alchemy supplier for a flat gold price, which removes any volatility from vial prices on the auction house. This actually equates into a large profit gain, especially in making potions, where five vials are required per craft.
Advertisement

Author: Stargrace

Just another gamer with too much time on her hands.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: