Is the arena wizard still OP?

[wtcannonjr]

The Frugal Fire Wizard does not win by overwhelming opponents with brute force. He is a trickster and master manipulator, like wizards are supposed to be. So either some card(s) introduced after the core set broke him, or he was already OP in a core set-only meta and people weren't skilled enough to take advantage of it, or both.

To test this, I propose we have a few core-only OCTGN tournaments and keep track of how well the wizards do in each one compared to other mages, as well as what their elements are. Then repeat the whole procedure with the core set and the first expansion that was released. Then repeat the procedure with the core set and the first two expansions that were released.

Continue like this until you've added all of the expansions.



-If the wizard is broken in a core only metagame, then either some card(s) in the core set broke him, the Arcane school itself was superior to all the other schools, or the problem was something in the wizard's ability card.

-If the wizard isn't broken until a certain expansion is introduced, then card(s) in that expansion broke him. His elemental training cannot be one of the main issues here because each wizard can only be trained in one element, and those elements are shared by other mages, namely the Druid, Warlord, and Warlock.

-If the wizard breaks in a particular metagame and then breaks even more in a later metagame, then the first break is caused by one of the reasons listed above, and the second break is caused either by his training in Arcane or by his elemental training. If it's the former, then the breaking points after the first breaking point should tend to be sets which introduced arcane spells. If it's the latter then the breaking points after the first breaking point should tend to be sets which introduced elemental spells, and mages that share those elements should gain a corresponding increase in power level.

We're not going to settle this debate by merely arguing theory. We've tried settling it that way several times already and it didn't work. If you all really care about this issue and want the best outcome for the game, then you guys should all just stop arguing and actually test your assertions.

There's no need to be aggressive. Please calm down and discuss this rationally everyone.

This seems like a systematic approach to testing the hypothesis. I like it.

Another approach that anyone can do in their local meta is to have one player always play the Wizard and each other player use other mages and modify these spellbooks in successive matches until they are able to beat the Wizard player consistently. By comparing pairs of spellbooks from individual matchups along with the results it will help the community understand if there actually is a problem and whether it is happening during spellbook design or during the battle.

[Brian VanAlstyne]

I don't think anyone is saying it's impossible to build a spellbook specifically designed to beat a wizard. The problem arises because that's not a fun way to build a book and that book won't be as good against other mages. There are just certain aspects of the wizard with regards to his inherent powers and his ability to put together a book that are just superior to all other mages. And that's not counting the absolute joke that is the Tower.

[Sailor Vulcan]

The Frugal Fire Wizard does not win by overwhelming opponents with brute force. He is a trickster and master manipulator, like wizards are supposed to be. So either some card(s) introduced after the core set broke him, or he was already OP in a core set-only meta and people weren't skilled enough to take advantage of it, or both.

To test this, I propose we have a few core-only OCTGN tournaments and keep track of how well the wizards do in each one compared to other mages, as well as what their elements are. Then repeat the whole procedure with the core set and the first expansion that was released. Then repeat the procedure with the core set and the first two expansions that were released.

Continue like this until you've added all of the expansions.



-If the wizard is broken in a core only metagame, then either some card(s) in the core set broke him, the Arcane school itself was superior to all the other schools, or the problem was something in the wizard's ability card.

-If the wizard isn't broken until a certain expansion is introduced, then card(s) in that expansion broke him. His elemental training cannot be one of the main issues here because each wizard can only be trained in one element, and those elements are shared by other mages, namely the Druid, Warlord, and Warlock.

-If the wizard breaks in a particular metagame and then breaks even more in a later metagame, then the first break is caused by one of the reasons listed above, and the second break is caused either by his training in Arcane or by his elemental training. If it's the former, then the breaking points after the first breaking point should tend to be sets which introduced arcane spells. If it's the latter then the breaking points after the first breaking point should tend to be sets which introduced elemental spells, and mages that share those elements should gain a corresponding increase in power level.

We're not going to settle this debate by merely arguing theory. We've tried settling it that way several times already and it didn't work. If you all really care about this issue and want the best outcome for the game, then you guys should all just stop arguing and actually test your assertions.

There's no need to be aggressive. Please calm down and discuss this rationally everyone.

This seems like a systematic approach to testing the hypothesis. I like it.

Another approach that anyone can do in their local meta is to have one player always play the Wizard and each other player use other mages and modify these spellbooks in successive matches until they are able to beat the Wizard player consistently. By comparing pairs of spellbooks from individual matchups along with the results it will help the community understand if there actually is a problem and whether it is happening during spellbook design or during the battle.

Thanks! Your procedure is pretty cool too. +1 to each of us for coming up with a way to apply the scientific method to a card game!

Also, @DevilsVendetta, you're completely misunderstanding Wtcannonjr's idea. He's not saying to just build hate books against the wizard. He's saying to use your regular books and to gradually modify them until they're doing well against the wizard consistently, and then to record how much the non-wizard spell books had to change in order to do that. Doing this should help make it more clear to everyone whether the wizard is OP or not, and why.

[iNano78]

wtcannonjr's suggestion has some important flaws:

Player skill generally trumps mage type.  We have a player who is undefeated and generally plays Necromancer.  None of us think Necromancer is particularly overpowered (because several of us have lost while playing Necromancer, and find that building and playing a Necromancer is a challenge) - but beating that player is very difficult.  He builds strong books and he seems to always be one step ahead in his strategy when playing.  This makes sense given he "wrote the book" on several strategies in the very early days of Mage Wars, publishing "how to" videos online for various openings and what makes them good.

How would you ensure that the winner of a given "X vs. Wizard" match isn't just the better player?  Perhaps randomize who plays Wizard and who plays X each time.

Similarly, how would you ensure the Wizard player's book is a strong Wizard book? 

Perhaps the best thing to do would be to have every player build a Wizard book plus whatever else they want, roll off to see who gets to play as the Wizard that time, and then track each player's stats and perform a significance test on the data to see if players are statistically better with Wizards than non-Wizards. 

But there's still the flaw that everyone knows that their non-Wizard is going to be paired up against a Wizard, and will build an "anti-Wizard" book, whereas each Wizard book doesn't know what it's going to be up against and has to build something that can beat the other 11 mage types - which still isn't a fair test.

As you can see, the problem in proving that Wizard is or isn't overpowered through an experiment like this isn't an easy one.

[iNano78]

Another way to look at it:

Let's assume that the Wizard isn't overpowered.  What would it take to make each of the perceived underpowered mages as good as the Wizard?

- Give the each a channeling of 10 (except perhaps the Druid since treebond accomplishes this).
- Remove triple-cost schools (personally I think this would be a terrible idea as it would make even less distinction between mage types, as all mages would become "jacks of all trades" - but this is simply demonstrating an alternative to nerfing the Wizard).
- Give each mage full training in a primary school and either an elemental school (e.g. Warlocks and Warlords are fine; Beastmasters and Priest(esse)s  and Necromancer (Air for poison clouds!!) need an element; Druid needs full Water training) or another primary school (presumably the Paladin will be Holy/War; Sorcerer might be Arcane/Dark; Forcemaster would like to be Mind/War - see spawn point discussion below; etc).
- Give each mage an intrinsic way to negate damage or heal or gain life or other some defensive ability (e.g. Forcemaster, Druid and Priestess already have this and are probably OK; perhaps one Warlock could have a "reflect damage" ability that, once per round at a cost of 1 mana, when suffering damage from an attack transfers 1 damage to the attacking enemy non-mage creature (similar to a damage barrier but also reduces the damage taken in that attack); The other warlock could, for a quick-spell costing 1 mana, transfer 1 damage from itself to a target enemy non-mage creature at range 0-2; Warlords could have built-in armor value of 2; Malakai Priest and both Beastmasters need something similar that's thematically relevant; Necromancer has built-in poison immunity, which might be enough (?)).
- Give each mage a built-in attack enhancement (e.g. Battle Skills (+1 melee) is fine for those that already have it; Necromancer's direct poison damage ability is OK; Druid's vines and stuck abilities might be enough since it indirectly assists damaging creatures by hindering/pinning them down; Johktari Beastmaster needs a built-in quick-spell with range 0-1 called "Hunting Darts" that costs 1 mana that deals 2 dice + chance to Bleed (further enhanced by her +1 ranged bonus) possibly replacing her built-in basic melee attack - which is about equivalent to the Wizard's Arcane Zap, since it could be used twice per round (quick-cast and regular action), just like the Wizard; I can understand why they might have decided not to give the Forcemaster the Battle Skills +1 melee (more incentive to cast Galvitar), but she could either use it or perhaps a "force choke" quick-spell costing 1 that deals 1 direct Psychic damage (useful when your Galvitar attack comes up 1 short of a kill and you need that extra 1 damage to finish off the target); something similar for any mage I might have missed (Priest/Priestess?)).
- Give each mage a broken mage-type-only Familiar or Spawnpoint on the power level of Wizard's Tower; e.g. cheap in mana and point cost, non-Epic, non-Zone Exclusive, doesn't cost an action to use, has built-in channeling, can bind spells (e.g. doesn't use them up - maybe gets creature from outside your spell book (token creatures?)) and can switch out its spell for free, etc, etc.

I think that's about it.  Feel free to add anything I might have missed.


*edit*
On the idea of familiar/spawn points, what if Battle Forge were War Mage only?  That would nerf all other equipment-heavy mages/builds. Forcemaster might need a similar equipment or enchantment spawn point (or training in War would be nice, instead of an elemental training). And Warlocks might cry foul and need something too (or realize that they're more about "burn" than "armor-up and go melee").

Net outcomes:

- Druid doesn't change much at all, only gaining full Water training. She even has a "broken" spawn point in Vine Tree, so she's totally good to go, and will benefit from higher level water spells as they become available in Siren expansion. Oh, and doesn't pay triple for Fire (unless Wizard starts paying triple for something).
- Necromancer gains Air training, but is otherwise unchanged.  Oh, and doesn't pay triple for Holy (unless Wizard starts paying triple for something).
- Forcemaster either gains Battle Skill (+1 melee) or a Force Choke ability (similar to Necreomancer's poison ability but Psychic), and gains War training (and thus retains access to Battle Forge), but still pays triple for non-Mind creatures (eg no cheap access to War creatures).
- Warlords gain built-in Armor 2 and Channeling 10, and retain access to Battle Forge.  Oh, and no longer pay triple for Arcane (unless Wizard starts paying triple for something).
- Priestess needs a built-in attack bonus of some kind. Perhaps when she attacks as her action, she may also quick-cast a holy spell (?).  And elemental training (Water?). Oh, and no longer pays triple for Dark (unless Wizard starts paying triple for something).
- Priest needs a built-in defense or damage negation or healing/regen or something. And Channeling 10. And elemental training (Fire). Oh, and no longer pays triple for Dark (unless Wizard starts paying triple for something).
- Straywood Beastmaster gains a built-in damage negation or regen or something, and an elemental training (Air for Galador?) and Channeling of 10.  Oh, and no longer pays triple for Fire (unless Wizard starts paying triple for something).
- Johktari Beastmaster gains a built-in 2-dice quickcast attack spell with chance to bleed, and Channeling of 10, and some sort of damage negation, and elemental training (Earth?). Oh, and no longer pays triple for Fire (unless Wizard starts paying triple for something).
- Arraxian Crown Warlock gains a damage reflection ability and Channeling of 10. Oh, and no longer pays triple for Holy (unless Wizard starts paying triple for something).
- Adramelech Warlock gets a direct damage transfer ability andChanneling of 10. Oh, and no longer pays triple for Holy (unless Wizard starts paying triple for something).
-

[Sailor Vulcan]

Another way to look at it:

Let's assume that the Wizard isn't overpowered.  What would it take to make each of the perceived underpowered mages as good as the Wizard?

- Give the each a channeling of 10 (except perhaps the Druid since treebond accomplishes this).
- Remove triple-cost schools (personally I think this would be a terrible idea as it would make even less distinction between mage types, as all mages would become "jacks of all trades" - but this is simply demonstrating an alternative to nerfing the Wizard).
- Give each mage full training in a primary school and either an elemental school (e.g. Warlocks and Warlords are fine; Beastmasters and Priest(esse)s  and Necromancer (Air for poison clouds!!) need an element; Druid needs full Water training) or another primary school (presumably the Paladin will be Holy/War; Sorcerer might be Arcane/Dark; Forcemaster would like to be Mind/War - see spawn point discussion below; etc).
- Give each mage an intrinsic way to negate damage or heal or gain life or other some defensive ability (e.g. Forcemaster, Druid and Priestess already have this and are probably OK; perhaps one Warlock could have a "reflect damage" ability that, once per round at a cost of 1 mana, when suffering damage from an attack transfers 1 damage to the attacking enemy non-mage creature (similar to a damage barrier but also reduces the damage taken in that attack); The other warlock could, for a quick-spell costing 1 mana, transfer 1 damage from itself to a target enemy non-mage creature at range 0-2; Warlords could have built-in armor value of 2; Malakai Priest and both Beastmasters need something similar that's thematically relevant; Necromancer has built-in poison immunity, which might be enough (?)).
- Give each mage a built-in attack enhancement (e.g. Battle Skills (+1 melee) is fine for those that already have it; Necromancer's direct poison damage ability is OK; Druid's vines and stuck abilities might be enough since it indirectly assists damaging creatures by hindering/pinning them down; Johktari Beastmaster needs a built-in quick-spell with range 0-1 called "Hunting Darts" that costs 1 mana that deals 2 dice + chance to Bleed (further enhanced by her +1 ranged bonus) possibly replacing her built-in basic melee attack - which is about equivalent to the Wizard's Arcane Zap, since it could be used twice per round (quick-cast and regular action), just like the Wizard; I can understand why they might have decided not to give the Forcemaster the Battle Skills +1 melee (more incentive to cast Galvitar), but she could either use it or perhaps a "force choke" quick-spell costing 1 that deals 1 direct Psychic damage (useful when your Galvitar attack comes up 1 short of a kill and you need that extra 1 damage to finish off the target); something similar for any mage I might have missed (Priest/Priestess?)).
- Give each mage a broken mage-type-only Familiar or Spawnpoint on the power level of Wizard's Tower; e.g. cheap in mana and point cost, non-Epic, non-Zone Exclusive, doesn't cost an action to use, has built-in channeling, can bind spells (e.g. doesn't use them up - maybe gets creature from outside your spell book (token creatures?)) and can switch out its spell for free, etc, etc.

I think that's about it.  Feel free to add anything I might have missed.


*edit*
On the idea of familiar/spawn points, what if Battle Forge were War Mage only?  That would nerf all other equipment-heavy mages/builds. Forcemaster might need a similar equipment or enchantment spawn point (or training in War would be nice, instead of an elemental training). And Warlocks might cry foul and need something too (or realize that they're more about "burn" than "armor-up and go melee").

Net outcomes:

- Druid doesn't change much at all, only gaining full Water training. She even has a "broken" spawn point in Vine Tree, so she's totally good to go, and will benefit from higher level water spells as they become available in Siren expansion. Oh, and doesn't pay triple for Fire (unless Wizard starts paying triple for something).
- Necromancer gains Air training, but is otherwise unchanged.  Oh, and doesn't pay triple for Holy (unless Wizard starts paying triple for something).
- Forcemaster either gains Battle Skill (+1 melee) or a Force Choke ability (similar to Necreomancer's poison ability but Psychic), and gains War training (and thus retains access to Battle Forge), but still pays triple for non-Mind creatures (eg no cheap access to War creatures).
- Warlords gain built-in Armor 2 and Channeling 10, and retain access to Battle Forge.  Oh, and no longer pay triple for Arcane (unless Wizard starts paying triple for something).
- Priestess needs a built-in attack bonus of some kind. Perhaps when she attacks as her action, she may also quick-cast a holy spell (?).  And elemental training (Water?). Oh, and no longer pays triple for Dark (unless Wizard starts paying triple for something).
- Priest needs a built-in defense or damage negation or healing/regen or something. And Channeling 10. And elemental training (Fire). Oh, and no longer pays triple for Dark (unless Wizard starts paying triple for something).
- Straywood Beastmaster gains a built-in damage negation or regen or something, and an elemental training (Air for Galador?) and Channeling of 10.  Oh, and no longer pays triple for Fire (unless Wizard starts paying triple for something).
- Johktari Beastmaster gains a built-in 2-dice quickcast attack spell with chance to bleed, and Channeling of 10, and some sort of damage negation, and elemental training (Earth?). Oh, and no longer pays triple for Fire (unless Wizard starts paying triple for something).
- Arraxian Crown Warlock gains a damage reflection ability and Channeling of 10. Oh, and no longer pays triple for Holy (unless Wizard starts paying triple for something).
- Adramelech Warlock gets a direct damage transfer ability andChanneling of 10. Oh, and no longer pays triple for Holy (unless Wizard starts paying triple for something).
-

All those tests you just proposed can come later. The first thing we need to do is to determine whether the wiz is OP or not, and if he is then by how much and when exactly he became that way. Now that I think of it, it probably would make sense to measure power level of any spell and ability in the game in terms of change in win-frequency per spellbook point.
The way the game was originally going to be, players would start with nothing but points and spend those points to make their spells and abilities before the fight. That was too complicated of course so Mage Wars instead became what we have today. But the principle still holds that everything can be worth spellbook points at least in theory.

Another way to say this is that, given a sufficiently large sample of players of approximately equal skill level (both at book-building and playing) playing a sufficiently large sample of games:

the wizard's power level is the number of spellbook points that must be added to a non-wizard in order to change his win-frequency in that matchup to approximately 50%.

This value can vary between matchups of course.

As for the issue of ensuring that skill levels of the participants are approximately equal, that could definitely be a weakness to such a study. But I don't think it's a fatal weakness. Usually the difference in player skill levels can be determined by how easy it is for the winning player to win against the losing player, or simply by doing some preliminary rounds to weed out the less skilled players from the study. Or just having a sufficiently large sample of players might suffice too.

[Biblofilter]

I'm not a mathematician, but I'm pretty sure you need a pretty big sample of matches if you really want to test this.

That said you could play 2x10 matches (which hopefully will take care of difference in player skill)

Hanma´s Wizard
Alexander West's Beastmaster
Tarkin 84´s Priestess
and piousflea´s Warlock all won an event.

So that would make an easy starting point for the 4 core set mages.

You might allow updates to the 3 non-wizard books and allow all current tourney legal cards. A lot of us assumes there's something wrong with Wizard Tower - but we might be wrong. I'm saying there's no reason to test core set first and then move forward if there's nothing wrong with the current card pool.

You also need to find a number that confirms the hypotheses: "Wizard is OP".

Id say superman would win at least 90% of all matches vs. ordinary men, but you might settle for 8-2 or 7-3.
Nobody is disagreeing with Wizard is the strongest or one of the strongest mages I'm assuming?

It might still show something's "wrong" if players A beats player B 7-3 with Wizard vs. lets say Beastmaster and when the roles are switch the result would be 5-5.

Playing a large number of matches like that will most likely take lace at OTCGN and then it won´t show much about live tourney mages with the 75 minutes timelimit.

You can sign me up for playing at least 20 matches in a study like this, but it is a lot of time so it might not be easy to get enough people for something like this.

[Sailor Vulcan]

I'm not a mathematician, but I'm pretty sure you need a pretty big sample of matches if you really want to test this.

That said you could play 2x10 matches (which hopefully will take care of difference in player skill)

Hanma´s Wizard
Alexander West's Beastmaster
Tarkin 84´s Priestess
and piousflea´s Warlock all won an event.

So that would make an easy starting point for the 4 core set mages.

You might allow updates to the 3 non-wizard books and allow all current tourney legal cards. A lot of us assumes there's something wrong with Wizard Tower - but we might be wrong. I'm saying there's no reason to test core set first and then move forward if there's nothing wrong with the current card pool.

You also need to find a number that confirms the hypotheses: "Wizard is OP".

Id say superman would win at least 90% of all matches vs. ordinary men, but you might settle for 8-2 or 7-3.
Nobody is disagreeing with Wizard is the strongest or one of the strongest mages I'm assuming?

It might still show something's "wrong" if players A beats player B 7-3 with Wizard vs. lets say Beastmaster and when the roles are switch the result would be 5-5.

Playing a large number of matches like that will most likely take lace at OTCGN and then it won´t show much about live tourney mages with the 75 minutes timelimit.

You can sign me up for playing at least 20 matches in a study like this, but it is a lot of time so it might not be easy to get enough people for something like this.

I kinda figured that the wizard being overpowered could be determined by whether his power level is an outlier or not compared to the other mages.

You think 20 games per participant, each one gets assigned to be a wizard or a non-wizard?

Yeah. To be honest I'm not entirely sure exactly how big it needs to be. Maybe 30 times the number of non-wizard arena mages? So like 360 games?

[iNano78]

To put things in perspective, there's a game that's become rather infamous for a balance issue.

A Few Acres of Snow is a war game by Martin Wallace that uses a deck building mechanic. It's an assymmetric game where the English battle the French for control of North America. A prominent member of the board gaming community, the late Tim Seitz (aka "out4blood") determined that the game was broken. He proposed that if the British performed a strategy he named the "Halifax Hammer" and executed this strategy flawlessly, that the British would always win. He described and demonstrated the strategy a few dozen times, but the designer refused to admit that the game was broken, claiming the strategy was not infallible and that the French had options that could win. Tim believed otherwise, and went on to prove that the English could always win by  going undefeated as the English after several hundred online plays against the (next) best players from all over the world. After about a year of this, the designer conceded the fact that the game was indeed broken and that the English could always win if they executed this strategy perfectly. Some rules were errratad and some cards were removed from the starting decks to prevent this strategy from being infallible, but the damage was done. The game got a lot of negative attention for being "broken".

What's less well known is that, even after publishing this strategy, Tim had a 100% win rate as the English but also had something like a 90% win rate as the French, even against players who tried to execute the Halifax Hammer strategy (but were unable to do it perfectly). A superior player could generally win even when taking the "inferior" side, even against someone trying to use the "broken" strategy.

It's similar to Go and Chess. The player who goes first in either of those games has an advantage over the player who goes second. Elite players at these games will add a disadvantage to the player going first (or advantage to the player going second) to account for this imbalance. However, it only matters at the extreme Elite level, as generally a superior player will win regardless of who goes first. But inherently, the game is imbalanced favouring the player who goes first.

So we're back to Mage Wars, where there's enough luck and plenty of deck building variety and counters to make it difficult to prove that one Mage is significantly better than others. A superior player with a well-built book will generally defeat an inferior player with a similarly strong book, or a similarly skilled player with a poorer book. But that doesn't mean that there isn't an inherent advantage to playing the Wizard (or inherent disadvantage to playing another Mage).

We could run a big experiment to prove through simulation and statistics that the Wizard is overpowered (or not), but it would take a large sample size to show that it's significant (like Tim Seitz did for A Few Acres of Snow). Or you can look at the problem analytically and see that the Wizard has superior training (Arcane + Element and no triple cost), superior stats (eg 10 channeling), and superior abilities (built-in damage negation; built-in powerful ranged attack; no conditional abilities) as I've already done.

Yet another way to look at it:

Let's strip all mages of their abilities and just look at stats: those with 10 channeling are at an obvious advantage over those with 9, even accounting for differences in life total.

Then let's strip all mages of abilities and stats and look at training: those with 2 schools are at an advantage over those with 1 - and having no triple school penalty puts the Wizard above all others. And anyone who suggests that Training in Arcane isn't superior to training in any other single primary school is delusional or simply playing devils advocate.

Then let's strip all mages of stats and training (eg everyone pays double for everything; same life and channeling) and look at Mage abilities. Having a way to reduce damage / regen / life gain is ALWAYS useful in a game about killing the other Mage. Having an ability that increases damage dealt or grants another way to deal damage to the other Mage is almost always useful (unless you exclusively use other creatures/spells to deal damage). Conditional abilities that require another creature with a specific subtype is less useful but might be OK - unless it only does something under very specific circumstances (like "Holy Avenger" which is ridiculous; and Wounded Prey doesn't work against the opposing Mage so is completely useless against a solo Mage) in which case it's very weak.

The Wizard is the only Mage that has (1) best stats (channels 10) AND (2) double training AND (3) (a) damage negation ability with no drawbacks or conditions AND (3) (b) attack enhancement ability (Arcane Zap is at least as good as Battle Skill) with no drawbacks or conditions, which can be used for both/either main action and/or quickcast (!). Druid is close (partial training in second school; gets Channeling 10 and built-in Regen as long as her tree is around and barring Deathlock/Poisoned Blood; has Vine markers to mess with the opponent, which is like an indirect damage enhancement through spell range and hindering of non-flying enemies). But the Wizard also gets training in the most important school for toolbox spells (Teleport, Dispel, Seeking Dispel, Nullify, Jinx, Mordok's Obelisk, Harmonize/Mana Crystal/Amulets, Purge Magic, Mage Wand) and no triple cost school so he can have unlimited options for the most diverse selection of tools (holy healing, dark curses, mind incantations, nature enchantments, war equipment, fire/water/earth/air utility and attack spells... the best of all schools!). And he also gets some of the best spawn points and familiars (including Wizard's Tower) - but his advantages go way beyond any single card. You can see that analytically; no need for simulations and statistics.

[Sailor Vulcan]

To put things in perspective, there's a game that's become rather infamous for a balance issue.

A Few Acres of Snow is a war game by Martin Wallace that uses a deck building mechanic. It's an assymmetric game where the English battle the French for control of North America. A prominent member of the board gaming community, the late Tim Seitz (aka "out4blood") determined that the game was broken. He proposed that if the British performed a strategy he named the "Halifax Hammer" and executed this strategy flawlessly, that the British would always win. He described and demonstrated the strategy a few dozen times, but the designer refused to admit that the game was broken, claiming the strategy was not infallible and that the French had options that could win. Tim believed otherwise, and went on to prove that the English could always win by  going undefeated as the English after several hundred online plays against the (next) best players from all over the world. After about a year of this, the designer conceded the fact that the game was indeed broken and that the English could always win if they executed this strategy perfectly. Some rules were errratad and some cards were removed from the starting decks to prevent this strategy from being infallible, but the damage was done. The game got a lot of negative attention for being "broken".

What's less well known is that, even after publishing this strategy, Tim had a 100% win rate as the English but also had something like a 90% win rate as the French, even against players who tried to execute the Halifax Hammer strategy (but were unable to do it perfectly). A superior player could generally win even when taking the "inferior" side, even against someone trying to use the "broken" strategy.

It's similar to Go and Chess. The player who goes first in either of those games has an advantage over the player who goes second. Elite players at these games will add a disadvantage to the player going first (or advantage to the player going second) to account for this imbalance. However, it only matters at the extreme Elite level, as generally a superior player will win regardless of who goes first. But inherently, the game is imbalanced favouring the player who goes first.

So we're back to Mage Wars, where there's enough luck and plenty of deck building variety and counters to make it difficult to prove that one Mage is significantly better than others. A superior player with a well-built book will generally defeat an inferior player with a similarly strong book, or a similarly skilled player with a poorer book. But that doesn't mean that there isn't an inherent advantage to playing the Wizard (or inherent disadvantage to playing another Mage).

We could run a big experiment to prove through simulation and statistics that the Wizard is overpowered (or not), but it would take a large sample size to show that it's significant (like Tim Seitz did for A Few Acres of Snow). Or you can look at the problem analytically and see that the Wizard has superior training (Arcane + Element and no triple cost), superior stats (eg 10 channeling), and superior abilities (built-in damage negation; built-in powerful ranged attack; no conditional abilities) as I've already done.

Yet another way to look at it:

Let's strip all mages of their abilities and just look at stats: those with 10 channeling are at an obvious advantage over those with 9, even accounting for differences in life total.

Then let's strip all mages of abilities and stats and look at training: those with 2 schools are at an advantage over those with 1 - and having no triple school penalty puts the Wizard above all others. And anyone who suggests that Training in Arcane isn't superior to training in any other single primary school is delusional or simply playing devils advocate.

Then let's strip all mages of stats and training (eg everyone pays double for everything; same life and channeling) and look at Mage abilities. Having a way to reduce damage / regen / life gain is ALWAYS useful in a game about killing the other Mage. Having an ability that increases damage dealt or grants another way to deal damage to the other Mage is almost always useful (unless you exclusively use other creatures/spells to deal damage). Conditional abilities that require another creature with a specific subtype is less useful but might be OK - unless it only does something under very specific circumstances (like "Holy Avenger" which is ridiculous; and Wounded Prey doesn't work against the opposing Mage so is completely useless against a solo Mage) in which case it's very weak.

The Wizard is the only Mage that has (1) best stats (channels 10) AND (2) double training AND (3) (a) damage negation ability with no drawbacks or conditions AND (3) (b) attack enhancement ability (Arcane Zap is at least as good as Battle Skill) with no drawbacks or conditions, which can be used for both/either main action and/or quickcast (!). Druid is close (partial training in second school; gets Channeling 10 and built-in Regen as long as her tree is around and barring Deathlock/Poisoned Blood; has Vine markers to mess with the opponent, which is like an indirect damage enhancement through spell range and hindering of non-flying enemies). But the Wizard also gets training in the most important school for toolbox spells (Teleport, Dispel, Seeking Dispel, Nullify, Jinx, Mordok's Obelisk, Harmonize/Mana Crystal/Amulets, Purge Magic, Mage Wand) and no triple cost school so he can have unlimited options for the most diverse selection of tools (holy healing, dark curses, mind incantations, nature enchantments, war equipment, fire/water/earth/air utility and attack spells... the best of all schools!). And he also gets some of the best spawn points and familiars (including Wizard's Tower) - but his advantages go way beyond any single card. You can see that analytically; no need for simulations and statistics.

That sounds a bit like Aristotelian-style deduction. Pure probabilistic reasoning from
priors can only take you so far. If we don't get enough actual, you know, evidence that can be observed rather than just theorized, then this debate will never be resolved.

[SharkBait]

...then this debate will never be resolved.

Quoted for truth. I don't see this being resolved, regardless of any of the above suggestions.

[Ravepig]

...then this debate will never be resolved.

Quoted for truth. I don't see this being resolved, regardless of any of the above suggestions.

All the testing required to just prove a point. Whatever happened to just having fun?  Playing 20+ structured games to prove a point does not sound like fun at all.

[Sailor Vulcan]

...then this debate will never be resolved.

Quoted for truth. I don't see this being resolved, regardless of any of the above suggestions.

All the testing required to just prove a point. Whatever happened to just having fun?  Playing 20+ structured games to prove a point does not sound like fun at all.

It's not just to prove a point. I actually am not entirely certain whether the wizard is still OP or not and would like to know. And you mean to tell me that playing 20+ games of Mage Wars isn't fun?

[bigfatchef]

...then this debate will never be resolved.

Quoted for truth. I don't see this being resolved, regardless of any of the above suggestions.

All the testing required to just prove a point. Whatever happened to just having fun?  Playing 20+ structured games to prove a point does not sound like fun at all.

It's not just to prove a point. I actually am not entirely certain whether the wizard is still OP or not and would like to know. And you mean to tell me that playing 20+ games of Mage Wars isn't fun?

So it's about proving the obvious that is obviously hard to prove

[Sailor Vulcan]

...then this debate will never be resolved.

Quoted for truth. I don't see this being resolved, regardless of any of the above suggestions.

All the testing required to just prove a point. Whatever happened to just having fun?  Playing 20+ structured games to prove a point does not sound like fun at all.

It's not just to prove a point. I actually am not entirely certain whether the wizard is still OP or not and would like to know. And you mean to tell me that playing 20+ games of Mage Wars isn't fun?

So it's about proving the obvious that is obviously hard to prove

No it's really not. My current best hypothesis is that the wizard is overpowered and that his tower is the sole cause, but I'd only give that maybe a 60-65% probability and it's based only on my own experience/intuition and the opinions of other competitive Arena players, plus a little bit of background knowledge. That's not the same as being "obvious". And if it were obvious there wouldn't be so much disagreement about it.