When Did YuGiOh! Jump the Shark?

YuGiOh! is a Japanese collectible card game in which players summon fantastical monsters which fight against each other. Players can buy physical cards or play the game digitally. YuGiOh! World Championship 2007 is my most-played DS game with over 500 hours invested and every single card unlocked.

I didn’t keep up with any new content after 2007. Coming back to the game, I almost couldn’t believe my eyes. Pendulum Monsters contained not 1, but 2 very long textboxes. Every new card had a description so long and convoluted that both my eyes and my brain hurt after getting to the end of one. What happened?

Power Creep

Power creep in games with collectible items refers to new items being introduced having more and more power, leading to a slow upward creep of average power, making old items ultimately useless. Making new powerful cards gives fans a reason to abandon their current deck and buy new cards.

The original YuGiOh! manga only featured what we now know as the card game in one episode, but the author received so much fan mail about it that he later on expanded upon the idea.

Most cards of this era are pretty weak even by 2007 standards. Most normal monsters were direct comparables (the one with the higher number is better). The only way to get strong monsters on the field was a Tribute Summon, which sacrifices up to 2 monsters to summon a stronger one. This was hardly worth it however, as a 2-sacrifice monster would need 3 turns to get on the field and could easily destroyed with the other aspect of gameplay, Spell and Trap cards, leaving you with no monsters.

Many Spell and Trap cards of this era contain pretty straightforward effects, some of them ended up proving game-breakingly strong (think Pot of Greed, Raigeki, Monster Reborn). If anything, the game was too simple, with direct comparables and not a lot of cool combinations.

There were some Ritual Monsters, but requiring you to hold the Ritual Monster plus a special Spell card makes all of them laughably bad, except for Relinquished, which was banned. Fusion monsters also required a special Spell card (Polymerization) and specific fusion material monsters, but often had no effect and weak stats, which made them equally unplayable in this era.

The original anime ended and its sequel, GX, saw the main character use an Elemental Hero deck based around fusioning weaker heroes into more powerful heroes with powerful effects. This takes the number of turns to summon a level 7+ strength monster down from 3 to 1 or 2. Fusion-based cards accelerated the pace of the game further, by granting additional fusions or loosening fusion material requirements (Fusion Sage, Fusion Gate, Future Fusion, Hex-Sealed Fusion, Overload Fusion, Chimeratech Overdragon).

This era also saw the introduction of archetypes, a set of strongly related cards designed to work together, rather than the semi-random mish-mash of cards previously released. Strong archetypes of this era include Six Samurai and Monarchs.

GX ended and its sequel 5D’s introduced a new monster type: Synchro. To summon a Synchro monster, typically one of a special type of monster called a “Tuner” is needed, and one or more mosters which are not “Tuner”s. The sum of their levels has to add up to the Synchro monster’s level.

The level requirement sounds restrictive, but it’s the exact opposite. You can just put monsters from all levels 4 to 9 into your Extra Deck and always have a very strong monster available for summon. The Tuner/non-Tuner requirement is much more relaxed than Fusion monsters, which require specific cards as material. While Fusion monsters can be summoned in 1 turn but require 3 cards, Tuner monsters often have special effects that summon extra monsters onto the field, which allows you to summon a powerful Synchro monster in just one turn from just a single card.

Xyz (pronounced “exceed”) monsters are summoned by overlaying monster cards of the same level. I don’t know too much about Xyz, but my general impression is they do not spike too much over the Synchro monster power curve: The Tuner requirement is dropped, but the level requirement is stricter, and the number of materials remains around the same.

Reddit agrees that Duelist Alliance, which introduced Pendulum monsters, was an extremely overpowered set. Pendulum monsters can either be played as a normal monster, or played as a spell card into a Pendulum zone (hence the 2 text boxes). After playing cards into both Pendulum zones, the player is allowed to summon any number of monsters between the two levels.

The sheer versatility and multitude of effects that a Pendulum has, in addition to each effect already being powerful in itself, making gameplay extremely overpowered and complicated. Adding insult to injury, some Pendulum monsters are even Fusion, Synchro, or XYZ monsters, an additional layer of complexity.

This analysis only looked at the newly introduced monster types, but when I look at some of the new effect monsters they basically read to me as “This card has at least 3 effects to completely destroy your opponent”. It’s ridiculous how many and how many powerful things a card can do now.

Card Length

I used the database over at https://db.ygoprodeck.com/, which seems intended for pricing data, but does contain a year filter and description field which is all I need to check my hunch that the length of card descriptions keeps increasing. A first version of my script did output:

2002 has avglen 106 ( 496 for Toon Summoned Skull)
2003 has avglen 173 ( 649 for Pitch-Dark Dragon)
2004 has avglen 153 ( 508 for The First Sarcophagus)
2005 has avglen 168 ( 427 for VWXYZ-Dragon Catapult Cannon)
2006 has avglen 187 ( 525 for Allure Queen LV5)
2007 has avglen 197 ( 660 for Vennominaga the Deity of Poisonous Snakes)
2008 has avglen 209 ( 627 for Arcana Force EX - The Light Ruler)
2009 has avglen 214 ( 589 for Majestic Star Dragon)
2010 has avglen 202 ( 650 for Shooting Star Dragon)
2011 has avglen 200 ( 681 for Meklord Emperor Skiel)
2012 has avglen 226 ( 588 for Noble Arms - Arfeudutyr)
2013 has avglen 266 ( 688 for Meklord Emperor Wisel)
2014 has avglen 310 ( 777 for Qliphort Shell)
2015 has avglen 350 ( 838 for Odd-Eyes Rebellion Dragon)
2016 has avglen 373 ( 916 for Nirvana High Paladin)
2017 has avglen 365 ( 921 for Astrograph Sorcerer)
2018 has avglen 375 ( 968 for Chaos Emperor, the Dragon of Armageddon)
2019 has avglen 374 (1048 for Endymion, the Mighty Master of Magic)
2020 has avglen 411 ( 922 for Performapal Celestial Magician)
2021 has avglen 414 ( 714 for Ancient Warriors - Rebellious Lu Feng)

which confirmed my intuitions

  • The average card length has rapidly risen from 200 to 400
  • This sharpest rise occurred from 2012 to 2015
  • The longest card in 2005 is as long as the average card in 2020/2021
  • (not shown) More and more cards are released each year, up from 300–400 a year before 2008 to 774 in 2019

I avoid using copyrighted card imagery here so feel free to google some of the example cards I mentioned.

Card Types

One possible objection is that the new monster types (Synchro, XYZ, Pendulum, and Link) contain a lot of boilerplate text (“1 Tuner + 1 or more non-Tuner WATER monsters”) that can mentally be ignored or replaced with a symbol. Especially Pendulum monsters are known for their complexity, contain 2 textboxes instead of the usual 1, and thus the most text on average:

However, this is not true because

  • only a small fraction of released cards are Pendulum monsters
  • “classical” Effect+Spell+Trap cards show the same pattern:

Source Code

The following Python code was used to analyze the data and produce figures:

import requests
from matplotlib import pyplot as plt
url = 'https://db.ygoprodeck.com/api/v7/cardinfo.php'for card_type, card_color in [
('all', 'black'),
# ('Normal Monster', '#f0da16'),
('Effect Monster', '#c4700a'),
('Synchro Monster', '#d3d3d3'),
('XYZ Monster', '#2e2e2e'),
('Pendulum Effect Monster', '#06bf6f'),
('Link Monster', '#1427cc'),
('Spell Card', '#07ba85'),
('Trap Card', '#bd0275'),
]:
print()
print(card_type)
for year in range(2002, 2022):
r = requests.get(f'{url}?startdate={year}-01-01&enddate={year}-12-31')
ds = r.json()['data']
if card_type != 'all':
ds = [d for d in ds if d['type'] == card_type]
if not ds:
continue
avglen = sum(len(d['desc']) for d in ds) / len(ds)
maxcard = max(ds, key=lambda d: len(d['desc']))
print(f'{len(ds):3d} | Year {year} has average description length {int(avglen)} '
f'({len(maxcard["desc"]):4d} for {maxcard["name"]})')
plt.bar(year, avglen, color=card_color)
if card_type != 'Pendulum Effect Monster':
plt.gca().set_ylim([0, 500])
plt.gca().set_xlim([2001, 2022])
plt.xticks(list(range(2002, 2022)), rotation=45)
plt.title(f"YuGiOh! average card description length by year ({card_type})")
plt.xlabel('year')
plt.ylabel('number of characters')
plt.savefig(f"yugioh_card_len_{card_type}.png")
# plt.show()
plt.clf()

Conclusion

YuGiOh! jumped the shark somewhere between 2012 and 2015. The ever-growing card texts and number of cards make it nearly-impossible for new players to join the game. Established players complain about growing prices, although I did not investigate that claim. Power creep has been a problem in every generation, but now complexity creep in the form of super long card descriptions has sucked the fun out of the game.

The situation seems unsalvageable. Until Konami does something about it, the only way to have fun with the game is to play with a year-restricted card set, for example picking up any of the Yu-Gi-Oh! World Championship games released between 2004 and 2011. Cheers.

23-year-old programmer from Germany!