Ice Deckathalon
How I Accidentally Won

Before I start, the purpose of this article is to give some insight into my thought process during the recent Ice Deckathalon. There will certainly be some parts meant to educate readers on Deckathalon mechanics, deck building, and card conservation. Do note, though, that not everybody will have access to the same pet, deck, or jewels that I did here. My goal is to illustrate concepts that will help people trying to complete this sort of “endurance” form of Deckathalons. If you’re trying to learn how to finish each individual floor of a Deckathalon tower, that’s not the focus here. When trying to finish the tower, you have a lot more flexibility and deck space to use stronger utilities (dispels, stuns, and plagues come to mind).

On the Competitive Deckathalon

Let me say first that I don’t think competing in the Deckathalon is necessarily inaccessible for anyone. It requires a consistent deck setup, the ability to follow a flow chart of options when complications arise, and time. This, in essence, leads to it being relatively simple if you have sufficient resources (high tier decks, gold for treasure cards, a specialized pet, and time). For this reason, I didn’t try to win the Deckathalon before now; I didn’t, and still don’t, personally find it interesting as a competition, because it seems like winning is as simple as having a high tier deck and specialized pet.

Ice Deckathalon Champion badge
The Deckathalon Champion badge and trophy

My suspicion is that if anything, it will get much more difficult to win as time goes on and more people get the higher tier decks and get more resources to compete. At that point, I would be much more interested in it. If the event ever gets to a point where you have to aim to be perfectly optimal with your play to win rather than simply having the best deck and pet, I would probably be more interested in it.

In other words, I think it’s very easy to win right now by just out-gearing people rather than being perfectly optimal with deck building and in-battle techniques. I don’t personally like that in a competition meant to test deck building and card conservation, which is why I hope things will change when more people get the 9th and 10th Deckathlete decks.

Like I mentioned, my priority has been to craft certain Deckathalon decks for use outside of the Deckathalon. For this, it’s easier and maybe more efficient to farm on multiple characters. That being said, I decided to try to win this one. This was mostly because I decided to change my rune farming strategy from using all of my wizards to just a few. Because of this, I could tell early on that I would win comfortably if I could use all of my credits even without playing optimally; it’s much easier to see where you are relative to other players when you only play on one character (leaderboards). I finished this event with 3211 points, and won by the equivalent of about 2 and a half credits worth of points.

If I had played all of my runs optimally (i.e., went into the event knowing I would compete), I would’ve gotten 3616 points at best (probably around 3500 accounting for some of my mistakes and my starting with deck 9). That would’ve put me a full 5 credits above the next closest player. The purpose of my saying this is to show that everyone (including me) competing for this Deckathalon was notably sub-optimal in terms of accumulating points. Although I didn’t pay too much attention to the leaderboards of previous events, I suspect they were similar in this regard (maybe even less optimal because casual players had less time to craft high tier decks).

I hope with this article to encourage more people to optimize their play to make for what I would consider more interesting competition. Of course, how optimized you can be in terms of point accumulation is relative to resources; as time goes on, play will naturally get more optimized because players will craft better Deckathlete decks. This article is meant to be a supplement to that natural optimization.

Before I get into strategy, I’ll give a quick description of how my week went.

My Deckathalon Timeline

So there I was on the first day of the Ice Deckathlon event. My goal for this event was to farm up enough Fire runes to craft the Ice Deckathlete’s Deck 10, which meant I needed about 600 more runes (I already had deck 9 and about 200 Fire runes). My usual strategy for farming for runes is to farm the first 4-5 stages on several different wizards so I was less restricted by the 3 credit per day limit. This time, however, I decided to choose one wizard and finish as many floors as possible with each credit. And that’s how it all started.

Naturally, for the first 3 days, I got the 20 treasure card points and collected credits on every wizard even if I didn’t go into the tower. The purpose of collecting points is to get the first 3 rewards on all wizards: the first reward is a Deckathlete’s Deck 1, and the third reward is 10 runes, both of which are incredibly useful for crafting decks and can be traded between wizards. For the treasure card points, I just bought 20 Keen Eyes treasure cards and enchanted 20 cards. After 4 days of this, you get 80 points, which is just enough for the 10x rune reward (you need 75).

The purpose of collecting credits is not as concrete. For me, I did it in case I wanted to farm on every wizard like I usually do. It worked out well for me because the wizard I won the event on was not the wizard I was initially farming on, and I still had all of my credits when I made the switch.

Into the Tower!

When I first went into the tower, I wanted to do it on a wizard with a lot of gold. The wizard I had the most gold on was my level 80 Death wizard, so I figured I’d start with him. I didn’t bother socketing jewels on the Ice Deckathlete’s Deck 9 that I had from the first event. If I was going to trade the deck between characters, I shouldn’t bother. I grabbed my Deckathalon pet, threw together a deck setup, and started farming the tower. Before I go on, note that I reference always starting a run through the tower at stage 1. This is the optimal strategy for farming basic runes, and as I’ve learned, for gaining points as well.

During my first run, I managed to go from stage 1 to the first battle of stage 6. By the end of day two (sixth run), I managed to go to the second battle of stage 7 consistently. That’s when I noticed that I was in first place on the leaderboards after day 2. This was on my level 80 Death wizard with no jewels on my Deckathalon deck. I didn’t think much of it, and just kept on farming on other wizards. This is sort of why I say I won “accidentally.” I wasn’t trying to win, and my winning strategy was initially created to help me farm for runes.

By day 4 (Friday), I checked up on the leaderboards again. My level 80 Death wizard was still in the top 10, which shocked me. I found that the top player was getting about 102 points per credit. This meant they were finishing around the first battle of stage 8 each credit. I knew I could do better than that, and my initial projection was that I could probably get 120 points (finishing stage 8) per credit. If I was able to finish stage 7 battle 2 without jewels on my deck, I should be able to beat stage 8 with level 130 jewels, right?

Getting the Points

So, I essentially knew by day 4 that I could win if I used all my credits. The question was if I would have time to use them all. Thankfully, I already had 80 points and 12 credits on my level 130 Storm wizard because I kept up with redeeming points and credits. At this point, I was more competing with my willingness to use my credits than anyone else. Each of these “endurance” runs took a while, so I ended up doing them while keeping busy with other things.

Ice Deckathalon leaderboard
The Ice Deckathalon leaderboard at the end of the event


My first run on my level 130 Storm was very successful in finishing stage 8, and I figured I could push myself to win the first battle of stage 9 as well. It surprised me that I not only finished the first, but also the second battle of stage 9 without preparation. This ended up being my baseline. I consistently was able to go from stage 1 to the second battle of stage 9 each credit. This meant 138 points per credit, more than enough to compete with the 100-105 on the leaderboards. I soon realized that I could craft Ice Deckathlete’s Deck 10 and probably finish stage 9 entirely for 144 points per credit. I eventually was able to do this successfully and consistently; however, I waited to craft the deck until I could do it without transmuting Fire runes.

And so that was that. There were very few runs that ended prematurely. When they did, it was usually only one battle before I was expecting. In the end, I was able to convert all of my credits to points. I finished around where I expected at 3211 points.


First, I’ll go over a few terms I use. See below:

To start, a stage is a set of 3 Deckathalon floors. Like I mentioned, assume that I always start a run through the tower from stage 1. When I talk about specific battles, I format it using a dash. For example: stage 7-2 would mean the second battle of stage 7.
By calling something a “roll,” I just mean that due to the damage range of a spell, it has a chance to defeat the mob in question, but is not guaranteed to.
This stands for Programmed Random Occurence. It is often used as a verb, and I will typically talk about it with regards to pet may casts.
By a “marathon” run, I just mean a run of the Deckathalon tower starting at stage 1 and aiming to go as far in the tower as possible.
By this, I just mean the school you’re attacking with. This should be the opposite school of whatever Deckathalon you’re doing, or for the Balance Deckathalon, Balance or a Spirit school (you might also attack with Storm here, but it doesn’t get a damage boost).

Deckathalon Stats

Before I get into what stats I think are optimal, I’ll do a quick review of the mechanics of flat stats.


In contrast to the rest of the game, the Deckathalon (as well as Beastmoon) relies on flat stats rather than percent based stats. Therefore, it’s essential to understand exactly how they work. The interaction between flat damage, shields, and flat resist is incredibly important to having good defensive instincts in the Deckathalon.

First, let’s talk about standard one hit damage spells. The order that boosts get applied is flat damage, percent based outgoing alterations, percent based incoming alterations, and finally flat resist (in the rest of the game, percent based damage boost is applied before flat damage and percent based resist is applied after flat resist).

For example, if I get hit with a Frost Giant on stage 9 when the opponent has a 35% Balanceblade and I have a 65% Snow Shield (a very common interaction given monster pierce, to be discussed later). Frost Giant does 460 based damage. First, the monster’s flat damage is added (this is 135 on stage 9). Next, the percent based alterations (blade and shield; order actually matters because the game truncates values to the nearest whole number after every added buff) are taken into account. This puts the damage at (460+135)*(1.35)*(0.35), which comes out to 281 damage. Finally, flat resist is factored in. I had 204 flat resist with my final setup, so that Frost Giant would do 77 damage to me.


DoTs work similarly, with one key difference: flat damage is applied in full to the initial hit of a DoT, and then applied again to the entire DoT portion, which is then divided by 3 and truncated to get the per tick damage. Flat resist, on the other hand, is applied in full to both the initial hit and each individual DoT tick. This means that flat damage is twice as effective on DoTs (with initial hits), but flat resist is four times as effective.

With high enough flat resist, this makes monsters using DoTs in the tower do very little damage to you. Other than that, the same order of operations applies. Note that this likely isn’t a glitch, but instead illustrates the idea that individual DoT ticks are treated as individual damage hits to the recipient, but as a single entity to the attacker. This idea is also what causes each individual DoT tick to have a chance to proc a pet may cast (more on this later).

For multi hit attacks (Minotaur, Hydra), flat damage and flat resist are both applied in full to each hit. In essence, the game treats a DoT as a 2 hit multi hit attack to the caster (initial hit is one, secondary DoT is the second), and as a 4 hit multi hit attack to the recipient (initial hit is one, each of the three ticks is another one).


As for what this means in terms of defending: damage buffs are much more valuable for monsters in the tower than they tend to be in the rest of the game. Since damage buffs are applied after flat damage, both the base damage and monster flat damage is boosted by a percentage based buff like a blade. When compounded with flat resist, this typically causes buffs to seemingly boost damage by much more than their percentage might lead you to believe.

For example: consider getting hit with a Frost Giant (460 base damage) on stage 9 (135 flat damage), and you have 200 flat resist. Without any buffs, it will do 395 damage. However, if the monster has a 35% blade, the Frost Giant will do 603 damage. This means that adding a 35% blade actually will give a 53% boost to the Frost Giant’s damage. This effect gets stronger as you get higher flat resist, which means that especially on higher floors when you have more optimized stats, you should be careful when monsters buff up for attacks.


Luckily, the same effect holds for debuffs. Percentage based debuffs can also be much more effective than you might expect when flat stats are involved. In the tower, the optimal debuff is usually school specific shields, but in some cases weaknesses are optimal (any Deckathalon using Plague or Virulent Plague for double battles, or using Weakness in the Myth Deckathalon specifically for Minotaur or in the Balance Deckathalon against Hydra).

If you look back to my earlier example, a Frost Giant on stage 9 with a 35% blade and 85% Snow Shield (pierced down to 65%) deals 281 damage before flat resist is factored in. If you have 200 flat resist, then, it will do 81 damage. However, in the previous paragraph I noted that without any buffs, a bladed Frost Giant on stage 9 does 603 damage against 200 flat resist. This means that adding a 65% shield (pierced down from 85%) is actually an 87% reduction in damage. Without the blade, the reduction becomes 98% (from 395 to 8 damage). Although this effect of shields becoming more efficient when you have more flat resist exists in the rest of the game, it is amplified in the Deckathalon just because flat stats are all that’s available, and you get a lot of flat resist.

As for how you get stats in the Deckathalon – there isn’t a whole lot. You get your deck, deck jewels, and pet may casts. Next, I’ll talk about what I think is optimal for these gear pieces. Please note that you don’t need to have the best deckathalon setup to just finish the tower. This section is aimed at showing what I think the optimal stats are for marathon runs.

Deckathalon Jewels

Obviously, you should be prioritizing getting the best deck possible to increase your health and flat resist. Higher decks also give you access to jewels, which are equally important. I’ll quickly go over what I see as the optimal jewels.

Ice Deckathalon deck 10

Tear: This unlocks once you get to deck 3. Obviously, you should be equipping health here.

Square: This unlocks at deck 6. I would say you want to aim for deck 6 as fast as possible to get this slot. For most Deckathalons, a single school defense jewel is optimal. If you don’t have one laying around, you should try to craft one at your appropriate level. If you’re level 115+, you can buy a 45 defense opal for gold for any school in Caravan in Mirage. The one exception here is the Balance Deckathalon – having done that one plenty (my first Deckathlete Deck 10 was the Balance one), I think a defense opal is optimal. You can buy those for arena tickets.

Triangle: This unlocks at deck 9. I think accuracy is optimal here. What accuracy jewel you choose depends on what spells you’re using. Of course, the first priority should be to not fizzle your attacking school. Deckathalons give enough base accuracy to put attacking school standard spells at 95% accuracy. This means that if your level’s best school specific accuracy jewel is at or below 5%, you should use that. However, if you’re at a high enough level that universal accuracy jewels are 5% or above, you have that option too. If you’re using spells with below base school accuracy (e.g., Firezilla, Arcticzilla, Leafstorm), you should use a school specific accuracy jewel to make sure those don’t fizzle. If not, then using a universal accuracy jewel at 5%+ will help with not fizzling Sprite, Steal Charm, etc., while still making sure you have perfect attacking school accuracy.

I didn’t end up using spells below standard accuracy, and I was using Sprite for many of my runs. Therefore, I used a universal accuracy jewel. The accuracy jewel is incredibly important if you’re trying to repeatedly finish marathon runs. This is because the value of having 100% accuracy goes beyond just not fizzling. It allows you to be confident making plays that would otherwise be risky. These sorts of situations might not come up often enough when playing normally, but during 24 marathon runs, they add up faster than you might expect.

Deckathalon Pets

An optimal pet has 4 may cast talents and a will cast talent, along with a may cast jewel. I didn’t have a suitable may cast jewel, so I only used four may casts. I’ll talk about may casts and will casts separately.

May Casts

Obviously, your first thought should be may cast heals. I like Spritely and Sprite Queen the most. The reason for this is that whenever they proc, they have a very good chance to have high value because the will heal future damage as well as current damage. This means that if, for example, a Sprite procs off of a shielded attack that does 0 damage to you, it still has potential to be useful if you get damaged in the next 3 rounds. Other good heals are may cast Fairy and may cast Healing Current, both because they carry the potential for an incredibly high value heal. I personally used those four may cast heals on my pet. I don’t think any other heals are optimal, though Unicorn isn’t awful if you have it.

The next type of may cast to consider is auras. The main options are Fortify and Amplify, with Fortify being my personal choice if you’re using an aura. Fortify will slightly reduce the damage you take on lower floors and increase the effectiveness of shields on higher floors. Another option is Empowerment, which is an interesting choice that is only really useful on higher floors (4-5 and up). Of course, if you have a may cast Brace jewel, it’s worth using.

In general, I don’t like offensive utility may casts. I would advise against using may cast Locust Swarm, Snow Serpent, Sunbird, Balanceblade, Dragonblade, Amplify, etc. This is because I find it optimal to calculate exactly how much damage you need when building your deck, and may cast offensive utilities will throw that off significantly. Put simply, these may casts are not easily reactable; you can’t easily react to one of these may casts mid battle to gain significant value in terms of card conservation like you could with a may cast heal or Fortify. I think that defensive utility may casts (shields) have the same issue. Of course, they aren’t bad; if you have these talents you can definitely use them. Keep in mind that I’m talking about optimal strategy for 24 marathon runs.

Will Casts

Deckathalon pet

For me, will casts are pretty simple. I value versatility very highly in general, and so I swear by will cast Vampire. Will cast Vampire gives you a consistent method to heal, deal chip damage, break shields, and can even defeat lower stage monsters without using any treasure cards. In general, I think you should have a will cast attack, if only because you save deck space by not needing any cards to finish the early stages. I would say take will cast Vampire for any Deckathalon except Death, and take will cast Leafstorm for the Death Deckathalon. Will cast Frozen Kraken is also a very good option for non-Ice Deckathalons, and may be optimal for the Fire Deckathalon.

If you aren’t familiar with how will casts work, they cost pet happiness to cast and have a cooldown that depends on the pip cost of the will cast and/or happiness cost. For Vampire, that cooldown is 3 rounds. In the Deckathalon towers, you can feed your pet snacks for happiness, so there isn’t much of a limit to how many times you can use a will cast in the tower. The only limits are the round cooldown and the hard limit on how many times you can use your will cast per battle (for Vampire, that hard limit is 3 casts per battle).

Plenty of pets with these talents are in the kiosk if you need to get one. A will cast talent is probably the most important Deckathalon pet stat. You absolutely need a lot of pet snacks to feed for happiness. Shanta Puddings work for this, as does buying snacks for arena tickets if you have a lot.

Deckathalon Monster Stats

The monsters in the tower also have stats. Apart from them gaining starting pips as you move through the tower, they also gain flat damage and armor piercing (along with critical block and resist to their own school, though these aren’t as relevant).

The gain in flat damage is relatively straightforward, and seems to be at 15 per stage (i.e., 15 at stage 1, 30 at stage 2, etc). A useful shortcut is to look at a stage’s corresponding deck to see how much flat damage monsters will have: monsters on stage 6, for example, will have the same amount of flat damage as flat resist is given by Deckathlete Deck 6 (that being 90).

The gain in armor piercing is slightly more complicated, but still not difficult to understand. On stage 1, no monsters have armor piercing. Beginning on stage 1-3, monsters gain armor piercing. From this point, at every X-3 floor (i.e., at the third floor of each stage), monsters get an armor piercing boost. At these floors, monsters alternate between gaining 2% and 3% pierce (in other words, they gain 2.5% pierce which is truncated to the nearest whole number). For example, from 1-3 to 2-2 the monsters have 2% pierce, from 2-3 to 3-2 they have 5% pierce, from 3-3 to 4-2 they have 7% pierce, and so on. The short of this pierce gain is that your shields will be less effective as you go up in stages. This is compounded by most Deckathalon monsters beginning to use spear spells to boost pierce at the highest stages (9 and 10).

Deckathalon Deck Setup

Moving on to specifically this Ice Deckathalon: take a look at the decks I used at the beginning (up to stage 9-2) and end (up to stage 9-3) of the event:

Notice that the second deck is exactly the same as the first, with the addition of 4 cards. Those 4 added cards are a Snow Shield, a Fireblade, a Heck Hound, and a Brandon minion. Those four cards are enough to win stage 9-3. I also traded a Sprite for a Fairy, and a Phoenix for a Koto (based on availability). I’ll talk each card below, as each has a specific purpose. It may be worth circling back to these explanations after reading the rest of the article. See the tabs below:

You might’ve expected more Snow Shields. I only carry 8 in the final deck, which last 27 total battles. The key here is that shields aren’t needed in every stage. With high flat resist, you take no damage from stages 1-2, and on stages 6-9 the primary attack used by the monsters is Frostbite, which gets destroyed by flat resist and a healing pet. Therefore, I only really need to consistently use Snow Shield on stages 3, 4, and 5. I used it to block buffed Evil Snowmen/Ice Wyverns or prevent giving monsters a free Balefrost (more on this later). I would also sometimes use it on stage 9 to block potential Frost Giants. The most key difference between my deck and what I would recommend for a player with a lower tier deck and/or less confidence in their strategy would be to be liberal with how many shields you take into the tower. Especially before deck 6 (which comes with a square slot) or if you’re not max level, definitely carry more shields than this.
These are emergency heals. I avoid healing unless I’m in serious danger of being defeated to give my pet a chance to heal me instead. I also would try to heal up if needed before a double battle (7-2, 8-2, or 9-2 typically).
Used to buff Fire Dragons, Sun Serpents, and Heck Hounds as needed.
Used to buff Phoenixes, Helephants, and Fire Dragons. I also sometimes used it to prevent the free Balefrost. If you’re wondering why I didn’t use Balanceblade or Elemental Blade, that’s because Amplify also buffs will cast Vampire, making it more flexible.
Used to buff one specific Phoenix and Helephant, and give other flexible buffing options
Used as an extra buff on 8-2 and 9-2 specifically.
Krokomummy/Mander were widely available and do a consistent 375 damage, so I used those over the variable Sunbird. Sunbird specifically forces a roll where Krokomummy doesn’t on 2-3, which makes the constant damage more desireable. I used these on lower floors.
I used Meteor Strike specifically on 2-2. The second copy was a backup hit, namely on stage 8-2 or 9-2.
Koto/Krokenkhamen both do 665 damage, making them more desirable than the variable Phoenix, but are somewhat uncommon. I didn’t always have them, but they prevented having a roll on stages 6-1 and 6-3.
Ember Everburn does 800 damage, which is much more desirable than Helephant. However, Ember Everburn was much rarer, meaning I didn’t have many. It prevents a crucial roll on stage 7-1; if Helephant hit minimum (725) damage, Feint+Helephant+will cast Vampire didn’t kill with deck 9. Once I got deck 10, it wasn’t a roll anymore with Helephant, but I kept it in anyway.
Fire Dragon was my attack for any X-2 floor, where I was up against 2 monsters.
Sun Serpent was one of my two options for higher stage single battles.
Heck Hound was one of my two options for higher stage single battles. I typically used it for maximum damage, to avoid shields, or in case I didn’t have enough power pips to use Sun Serpent.
Brandon minions are probably the best minions for any Deckathalon. The specific school isn’t always relevant. I had success with every school minion this Deckathalon (except Ice, obviously). I think the strongest Brandon minion for this Deckathalon were Balance and Fire. Balance is good because it consistently blades you once its starting pips run out. This is usually round 2-3 after casting. It also does solid chip damage with Sandstorm, Judgement, and Loremaster. Fire was obviously good here because it got a boost both from the Ice opponents and my Wyldfire. However, it’s important to note that the Brandons boost to their opposite schools and use Azteca auras (i.e., Furnace). This makes the opposite school Brandons glass cannons.

In general, I think the best Brandons for any Deckathalon are Balance, Fire, Myth, and Death.

All of these schools will do consistent damage. The Storm Brandon, while usable, will often fizzle, use low pip Tempests, or use Wild Bolt, making the damage far from consistent. Additionally, it often uses Supercharge round 1, which inevitably is wasted the next turn on a Wild Bolt or 1 pip Tempest. The Life Brandon is definitely good, and you will not be defeated when this is out. However, it will not regularly attack; it will occasionally Sacred Charge or Eternal Leviathan. It will also cast Sanctuary, which means you need to be careful casting damage bubbles. Ice Brandon is definitely decent in general, though it has the lowest natural damage boost and will Stun Block and Legion Shield/Tower Shield. I prefer minions that will do notable chip damage to more defensive ones.

Note that Brandon minion cards can be farmed with a 100% drop rate from Neberyx in Dragonspyre. It’s a fairly simple boss with 44000 health. Which school you get is random, so it’s probably worth farming a bit.

General Deckathalon Strategy

The general strategy of marathon runs should be to conserve as many cards as possible. You want to finish each battle with as many cards remaining as possible. The strategy for doing this will differ based on which Deckathalon you’re running; I’m here to focus on what I did to win the Ice Deckathalon, so I’ll talk about that. I thought of each run through the tower as split into 3 phases: early, middle, and late floors. Roughly speaking, early floors spanned 1-1 to 3-1, middle from 3-2 to 5-3, and late from 6-1 to 9-3. I’ll break down my general strategy for each section below.

For early floors, I pretty much needed to wait on my will cast Vampire to either defeat the monster on its own. I used will cast Vampire exclusively to win floors 1-1 through 2-1. For 2-2, 2-3, and 3-1, I had to dig for specific cards to defeat the monsters. A note on discarding: you should be discarding a lot. On earlier floors, discarding and drawing 7 cards per round is not uncommon at all and should be done. You need to know how many copies of cards you have and how many you need to use. This will tell you exactly what you can discard. Discard aggressively and cycle in new cards.

For middle floors, this is where the endurance happened. These were the most dangerous floors in terms of taking chip damage. Specifically for the Ice Deckathalon, this is because these floors have good enough attacks to do significant damage (Evil Snowman, Ice Wyvern), but don’t use Frostbite with regularity yet (more on this later). These floors were about waiting to set up my attack while defending when necessary when the monsters buffed up. The monsters can’t use Steal Ward until stage 5, so stage 4 was a hotspot for using Snow Shields to defend. The monsters also start using Tower Shield here, but my planned attacks didn’t suffer much from being shielded since the monsters still have low health.

For late floors, I shifted from timely defense to timely offense. Here, will cast Vampire was integral to breaking Tower Shields, which would now negatively affect my attacks. Choosing when to cast Vampire was key, as is reacting to monster pip level to gage when they have the most potential to shield. Defense was not as much of an issue here because the monsters main attack is Frostbite. Generally, I preferred a monster using Frostbite to them passing. Frostbite being the monsters preferred attack was a huge help.

My reason for wanting monsters to use Frostbite was simple: with high enough flat resist, Frostbite does very little damage per tick (under 20 per tick at stage 9 with my final setup). Additionally, Frostbite damages you 4 times. This means that it gave my pet 4 opportunities to proc a heal. More often than not, I would gain health from an enemy using Frostbite on me. Since Frostbite was the most preferred attack at these floors, you can see why I didn’t need to shield much. The one exception was on stage 9, where the enemies use Frost Giant. I saved at least 2 shields for blocking round 1 Frost Giants on stage 9.

Apart from Frostbite, the next most common attack at late floors was Angry Snowpigs, which only deals 2.5 ranks of damage. This doesn’t translate to much damage against 204 flat resist, even when considering enemy flat damage (+15 per stage, 135 on stage 9). They also could use Ice Wyvern, which was important to keep in mind if they got buffs up. Colossus is also a possibility, but it’s incredibly rare. In around 100 battles where I could’ve been hit with Colossus, I only got hit with it 3 times, none of which were fatal. Regardless, it was important to keep in mind.


I’ve referenced “preventing a free Balefrost” previously. I’d like to explain that now before moving on. As a method to prevent pure minion strategies, the devs implemented a “cheat” for all monsters in the tower. If you don’t successfully cast a spell for 3 rounds in a row (even if you fizzle or are stunned), a global damage bubble for the monsters’ school will be casted. The first cast is a 35% global. Next, if you go 3 more rounds without successfully casting a spell, a 50% global is casted. I’m sure more globals are casted after, but I didn’t stick around any battle long enough to find out what the percentage boost is. This can be advantageous for the Balance Deckathalon (where the monsters boost to Balance damage), but for anything else, it’s very bad.

The obvious way to counter this, of course, is to cast spells. This can become a problem, however, when you’re trying to conserve cards. For example, will cast Vampire has a cooldown of 3 rounds. If I needed to pass for 4 rounds to cast 2 Vampires and win a battle in the Ice Deckathalon, a Balefrost would go up if I don’t cast anything myself. This would turn ugly if the opponent put up some other buffs in addition to the Balefrost.

As a more specific example, my plan for 3-2 was to pass until I could Fire Dragon (which killed the monsters through Tower Shields). I needed to wait 5 rounds to get pips to cast it. This meant that I was forced to cast a spell, or face what would likely be 3 rounds under a Balefrost against 2 monsters whose main attacks included Evil Snowman. Since putting up the Balefrost could have meant taking 400-500 damage that I would have otherwise avoided, I decided to sacrifice a shield. On round 2 or 3 (depending on context), I would use a Snow Shield in a battle where I wasn’t taking much damage regardless specifically to avoid the Balefrost going up.

For this reason, the concept of spacing out your passes becomes very important in marathon runs. I often would pass on round 1 on any stage even if I had a blade I knew I would need in my hand. I did this specifically to make sure I had room to pass later on if I needed to, say, wait for my will cast Vampire cooldown to break a Tower Shield or finish off an opponent. It’s definitely worth understanding this idea.

Specific Deckathalon Strategy

Now I’ll go over my specific game plan for each of the 27 floors I would take on per run at the end of the event. Note that I didn’t bother with stage 10, only the first 9 stages. Hopefully, this is interesting or insightful in some way. For calculation purposes, note that all enemies in the tower have a 15% boost Fire. This 15% boost to the opposite school is normal with all Deckathalons (Balance has a 5% boost to Spirit schools and Balance). Additionally, the monsters gain 15 flat damage boost per stage and roughly 5% pierce per two stages (pierce gain is staggered). See below:

Spells: Frost Beetle (main attack), Ice Trap, Snow Serpent

1-1 (Ice Elf Warrior, 240 health): Easy plan here, pass 3 times then will cast vampire. I can’t be damaged here since the enemy can only Frost Beetle or rarely Ice Trap.

1-2 (Ice Elf Warrior and Frostbite Colossus, 240 and 270 health): Same idea as above: pass 3 times, Vampire, pass 3 more times, Vampire the other mob. Balefrost went up after round 3, but at that point I kill the Colossus. The Elf couldn’t damage me even under Balefrost with my flat resist.

1-3 (Blizzard Troll, 300 health): Same idea of passing 3 times then Vampire. This mob could damage me with Snow Serpent and Ice Trap, but I would heal that off with the Vampire.

Spells: Frost Beetle, Ice Trap, Snow Serpent (main attack), Balanceblade (35%), Evil Snowman (rare attack)

2-1 (Malevolent Snowfella, 360 health): Here I would pass 3 times, Vampire, then pass 3 times and Vampire again. Sometimes I’d take damage here (a Balefrost went up and the Snowfella had 3 turns to use it), but it wasn’t significant.

2-2 (Malevolent Snowfella and Malia Snowdrifter, 360 and 400 health): I would dig hard for a Meteor Strike and use it as soon as possible. If I didn’t pull it in 3 rounds, I could Vampire to gain some health back.

2-3 (Spirit of Iceolation, 440 health): I would Krokomummy round 1 and guarantee victory.

Spells: Ice Trap, Snow Serpent (main attack), Balanceblade (35%), Evil Snowman (main attack), Tower Shield

3-1 (Hailstone Golem, 530 health): I would pass twice, Krokomummy, then Vampire. I delayed the Krokomummy until round 3 to catch any early Tower Shields; I would’ve rather had the Krokomummy be shielded than the Vampire.

3-2 (Hailstone Golem and Polar Bear Officer, 530 and 590 health): Here, as I mentioned earlier, I’d Snow Shield either on round 2 or 3 to avoid giving them a Balefrost (round 2 if they buffed round 1, round 3 if not). I would then wait until I could Fire Dragon. I would also Vampire on the same round as the Fire Dragon as needed.

3-3 (Cold Hart, 650 health):  Round 1 Krokomummy would put this monster in range of a Vampire. If he shielded before I could Vampire, I would just have to wait out the cooldown to Vampire again.

Spells: Ice Trap, Snow Serpent, Balanceblade (35%), Evil Snowman (main attack), Tower Shield, Ice Wyvern, Frostbite (rare attack)

4-1 (Ice Cleaver, 780 health): Same goal as 3-3 – Krokomummy round 1 and Vampire (twice if needed).

4-2 (Ice Cleaver and Frostfell Raider, 780 and 860 health): Same concept as 3-2 – pass twice, Snow Shield to stop Balefrost from going up, then pass until I could Fire Dragon and Vampire.

4-3 (Abominable Snowbeast, 950 health): Here, I’d pass twice, Phoenix, and Vampire. If it shielded, I would use the Vampire to break the shield and then Phoenix and wait out the cooldown for a second Vampire if needed. I had flexibility to choose in this shield scenario if I wanted to waste a shield to stop Balefrost going up or if I wanted to save the card. Usually I’d use the shield.

Spells: Ice Trap, Balanceblade (35%), Evil Snowman (main attack), Tower Shield, Ice Wyvern, Frostbite, Steal Ward

5-1 (Tundra Spider, 1050 health): Same as 4-3 – Phoenix and Vampire. I would shield if it got a buff or two up.

5-2 (Tundra Spider and Numbskull Nibbler, 1050 and 1160 health): Here, I’d pass twice, Fireblade, then pass until I could Fire Dragon. Again, I had the option to shield if they buffed. I didn’t need to break Tower Shields with Vampire here, so I’d usually Vampire after the Dragon. This is the first battle where I needed two cards to defeat the monsters.

5-3 (Crevasse Moth, 1260 health): For this one, I’d Amplify, Phoenix, and Vampire. If it shielded, I could break it with a Vampire and then either Phoenix and wait for another Vampire or just Helephant and defeat it outright.

Spells: Ice Trap, Balanceblade (35%), Evil Snowman, Tower Shield, Ice Wyvern, Frostbite (main attack), Steal Ward

6-1 (Blizzard Dragon, 1410 health): Same idea as 5-3: Amplify+Phoenix+Vampire. The one key difference is that Phoenix here is a roll to kill. I would always use a Koto or 5 pip Krokenkahmen here to guarantee victory. If it shielded, I’d again break it with Vampire and then Koto or Helephant.

6-2 (Blizzard Dragon and Frozen Pharoah, 1410 and 1560 health): Fireblade, Amplify, and Fire Dragon, spaced out with passing depending on what power pips I got. Shields here still weren’t a huge issue, so I didn’t need to break them with Vampire. Instead, I’d Vampire after using the Dragon into shields to make up for the shielded damage.

6-3 (Itztlacoliuhqui, 1720 health): Feint-Helephant or Feint-Koto and Vampire would do the trick here (depending on availability). Against a shield, I would break it with Vampire. Even if I wasted a Feint by getting hit with a poorly timed Tower Shield, I had enough extras to use another one here.

Spells: Ice Trap, Balanceblade (35%), Evil Snowman, Tower Shield, Ice Wyvern, Frostbite (main attack), Steal Ward, Angry Snowpigs

7-1 (Frostwood Goblin, 1900 health): Same exact plan as 6-3, but I needed to use Ember Everburn instead of Helephant because Helephant was a roll to kill when I had deck 9. With deck 10, Helephant was no longer a roll.

7-2 (Frostwood Goblin and Frozen Peas, 1900 and 2090 health): The plan here was to Blade-Amplify-Dragon. I also tried my best to break Tower Shields with Vampire. If needed (ie, if the initial hit of my Fire Dragon was shielded on one or both of the monsters), I would use my extra copy of Krokomummy or Meteor Strike to finish the battle.

7-3 (Tim Burr, 2300 health): Here I had the option to either blade and Sun Serpent or blade and Heck Hound.

Typically I preferred Sun Serpent here if I got enough power pips. I needed 3 power pips to use it, and had 1 from my deck and a 30% power pip chance on stage 7. I therefore had a 58% chance to get enough power pips to cast Sun Serpent. If I could cast it, I would, and if not, Heck Hound would be good. Against shields, I could either break it with Vampire and Sun Serpent, or just choose to Heck Hound instead. Either way, I would definitely have to space out my passing to make sure Balefrost didn’t go up while I gained the pips needed to attack. I’d always pass rounds 1 and 2, and only then blade on round 3. This was to avoid a situation where I would have to pass 5 times in a row to wait for pips to attack; if I did that, Balefrost would go up in the process.

Spells: Ice Trap, Balanceblade (35%), Tower Shield, Ice Wyvern, Frostbite (main attack), Steal Ward, Angry Snowpigs (main attack), Freeze, Colossus (very rare attack)

8-1 (Snowball King, 2530 health): Here I’d do the same as 7-3 (with better odds to be able to cast Sun Serpent). If it shielded, I’d have to break it with Vampire and use an Amplify. Note that because they can stun on stage 8, I’d blade on round 2 instead of 3. This was in case they stunned me on round 3 after I passed on rounds 1 and 2 (which would force Balefrost to go up).

Another interesting issue with stuns: if you use a will cast while stunned and then use it again the next round, it will not cast the round you’re stunned, but will cast twice on the round after (both on the target chosen on the second round). This bypasses the need to use happiness for the second cast, the 3 round cooldown between casts, and the total limit of 3 casts per battle. This is certainly a bug.

8-2 (Snowball King and Frost Djinn, 2530 and 2720 health): The goal here was to blade, Wyldfire, Amplify, and Fire Dragon, with Meteor Strike as a backup hit if needed. I used Vampire to break shields if needed and to deal some chip damage and/or finish off the battle.

8-3 (Qhool Qhat, 3030 health): Blade, Amplify, and Heck Hound or Sun Serpent (depending on availability and how much it shielded). I could use Vampire to break shields. This was similar to 8-1 and 7-3.

Spells: Ice Trap, Balanceblade (35%), Tower Shield, Ice Wyvern, Frostbite (main attack), Steal Ward, Angry Snowpigs (main attack), Freeze, Icespear (12%), Colossus (very rare attack), Icespear, Frost Giant

9-1 (Coldhearted Troll, 3380 health): Typically I’d shield round 1 if I had low (700 or below) health at this point to avoid a round 1 Frost Giant. Then I would proceed as in 8-3, with a blade, Amplify, and Sun Serpent or Heck Hound. Usually, it would be Heck Hound. I would be able to heal up with Fairy and Vampire while the DoT ticked to prepare for 9-2.

9-2 (Coldhearted Troll and Hailstone Hornet, 3380 and 3720 health): This is the theoretically most difficult battle in the run. I again would consider a shield round 1 to block Frost Giant, depending on my resources and health. Here, I’d use a Brandon minion as soon as possible. This is needed because they do necessary chip damage while conserving deck space. Balance was my preferred minion for this fight, but I’ve used every school Brandon.

After getting Brandon up, I would set up with a blade, Amplify, and Wyldfire for a Fire Dragon. This would be plenty of damage given Brandon’s chip damage and shield breaking. I would often have Meteor Strike as a backup hit here, too. Balance Brandon specifically also would provide an extra buff for Fire Dragon (30% Balanceblade or 25% Bladestorm), which helped greatly. I typically also had extra Feints and heals if needed.

9-3 (Geirrod, 4100 health): For the last battle, I had allocated a Snow Shield, a Fireblade, a Heck Hound, and a Brandon minion. I often also had an extra shield, a Feint, and a heal. If I could, I would again consider a shield round 1 to block Frost Giant. I would then Brandon minion, the preferred school here being Fire. The goal was to have Brandon do around 1500 total damage. This would put the monster in range of a bladed high pip Heck Hound. 1500 damage is very doable for every Brandon, but a Fire Brandon could do it fast. Once Brandon did some damage, the monster would target it, giving me enough time to set up Heck Hound.

Closing Thoughts

So that’s how I accidentally won the Ice Deckathalon. It’s definitely a time investment to play optimally – it can take an hour or more to do a marathon run. I typically had to do runs sort of in the background while doing other things. Although it was fun to do once, I don’t think I would want to compete for every Deckathalon. Depending on if I have time, I might compete for some future Deckathalons. However, I would have to genuinely enjoy playing in the tower and/or want a deck for use outside of the Deckathalon.

Although I think the Deckathalon in general is alright where it is, I personally would prefer a few changes. First, I think the decks should be easier to craft in terms of basic runes. Keeping the exact same recipes and reducing the amount of basic runes needed by, say, 30% would go a long way in terms of allowing more players to compete at the highest level. Currently, it takes over 2100 basic runes to craft a Deckathlete’s Deck 10 when starting with deck 1; the decks are a huge barrier to entry.

Additionally, I think that players should automatically get all of their credits at the start of the event. This is definitely more of a quality of life change, as you currently are forced to log in every day to collect credits. This sort of credit change would mostly benefit people farming for runes to craft decks, which again would lower the barrier to entry for competing.

Finally, I think that credits should be less limited (i.e., more than 3 per day, but likely not totally unlimited), but that only the first 3 (maybe even 2 or 1) credits you use per day should count towards your event points. This would have two main benefits. The first of these would be allowing players to practice and optimize their strategies on their competing wizard without having to risk a sub-optimal run through the tower. The second would be to allow competitors and more casual players to farm up runes without being restricted by a credit limit that is necessary to make the competition less grindy.

If any of these changes are made to the Deckathalon, I feel that they would be huge quality of life changes that would allow more players to more quickly reach optimal levels of play. As it is now, the Deckathalon will always have a very high barrier to entry. I think that the Hands of Fate recipes can remain as is to provide a long term goal for consistent Deckathalon players, but the deck recipes and credit distribution could be changed.

Hopefully this article was insightful! Any thoughts?
Let us know in the comments!

Share your vote!

Do you like this post?
  • Fascinated
  • Happy
  • Sad
  • Angry
  • Bored
  • Afraid
Final Bastion