MoneyBall MT Orgrimmar
Here’s a new MoneyBall that includes 220 unique lineups, and then uses the match data to explore possibilities.
Probably best to download and view offline, because it’s a lot of data for a spreadsheet.
Skip to the end for a link to an explanation of what this is, if you haven’t seen this before. Skip this entirely is you’re busy or don’t like details
The purpose of this is not to predict ‘the best lineup’ and claim that’s what should be played. In conquest, you’re almost always safe playing the best decks. But you also need to know what to ban and how to ‘Tech’ accordingly. The purpose of MoneyBall is to understand the matchups, and explore different strategies that are possible. One could find a low scoring lineup from here, understand the ban strategy, craft decks with specific targeting tech choices, and use this lower scoring lineup to actually outperform the more obvious lineups. We’re in a meta where this is more viable than other metas I’ve seen. For example, one could find a ban-Paladin lineup, and then tech in anti-secret, watch post, Ogremancer, maybe ooze for rogues, etc… and do better than this model would necessarily predict, without further matchup number changes. If it’s one thing we learned from the Specialist era is that minor tech card changes can make a huge difference in matchups.
All possible 220X220 matchups (48,400) are on the sheet, ‘Moneyball’. The lineup statistics are on the sheet called, ‘LineupResults’.
This is not a representative meta. The lineups are unique, so results would vary. A representative meta would probably have, say, 20 identical lineups for the ‘best lineup’.
This picture shows the top 30 lineups, sorted by average lineup winrates / count of favored.
It should be no surprise that the top lineups are full of the top classes we also see on ladder and at the qualifiers. However, due to using data from Bronze to Legend, many of the top lineups also have Token Druid, for example. This is okay. Anywhere you see Paladin, I found that Secret, Secret Libram, or Aggro all did well in that lineup. In general Secret Libram Paladin was usually the better choice.
Token Druid – Face Hunter OR Spell Mage – Paladin – Rush Warrior
Spell Mage – Paladin – Watch Rogue – Rush Warrior
Some top lineups that were somewhat unique:
Primordial Druid – Face Hunter – Aggro Paladin – Rush Warrior
Token Druid – Spell Mage – Paladin – Elemental Shaman
Leoroxx Hunter – Spell Mage – Paladin – Watch Rogue
Leoroxx Hunter – Token Druid – Spell Mage – Paladin
Soul DHunter – Face Hunter – Paladin – Rush Warrior
Inquisitor DHunter – Face Hunter – Paladin – Rush Warrior
Spell Mage – Paladin – Miracle Rogue – Rush Warrior
Token Druid – Spell Mage – Paladin – Control Warlock
Token Druid – Spell Mage – Paladin – Miracle Rogue
Spell Mage – Paladin – Watch Rogue – Control Warlock
Leoroxx Hunter – Spell Mage – Paladin – Control Warlock
Leoroxx Hunter – Spell Mage – Paladin – Secret Rogue
Some fairly unique lineups that have a shot at beating the top lineup(s), but are still okay against the field:
Leoroxx Hunter – Spell Mage – Paladin – Control Warlock (or Watch Rogue)
Spell Mage – Libram Paladin – Watch Rogue – Control Warlock
Spell Mage – Aggro Paladin – Secret Rogue – Control Priest
Lifesteal DHunter – Aggro Paladin – Watch Rogue – Rush Warrior
Spell Mage – Aggro Paladin – Spell Priest – Rush Warrior
Spell Mage – Paladin – Spell Priest – Rush Warrior
Spell Mage – Paladin – Rally Priest – Rush Warrior
Spell Mage – Paladin – Watch Rogue – Rush Warrior
You will notice I haven’t listed any lineups with Lifesteal OTK Demon Hunter. By the numbers, it’s not good in the lineup, but highly skilled players could do well to ignore the numbers here. Other decks may be like this too. But unless you’re super skilled, you’d be best off sticking to one of these lineups.
This model is deterministic and the result is based on assuming the favorite player wins and ladder stats are relevant in the tournament.
Keep in mind that a prediction of a player winning 56% of the time, for example, means they lose 44% of the time, and vice versa.
Disclaimers. This comes with the usual caveats about this data being based on average numbers across all users and decks. If we learned anything from the Specialist era, it’s that specific tech choices and player familiarity with decks can make a huge difference in performance. Open decklist formats also don’t allow you to have any surprise factor so stats from ladder will be different than stats in tournaments. This also ignores player skill and any bias in terms of favorites. Nevertheless, this can be used as a sort of ‘viewing guide’ to understand each match while your watching it or to consider lineup options.
If you don’t know what this is, please review my previous post below for an MoneyBall explanation and how to read the charts.
One of these days, I’ll do better in a Google doc. If anyone wants to volunteer… please do, I have a baby but not time.
Thanks for reading, and I’ll see you on ladder.