Your ideas for a tournament tie-breaker

[DeckBuilder]

[Significant edits on original post]

I shall assume everyone reading this knows the GenCon tie-breaker rule change implemented. If not, it's on the main site front page. This is an attempt to get everyone to contribute ideas on how to make Mage Wars into a tournament game which generates fair results.

So... what to do with a game of variable length based on strategy match-up?

I don't profess to have a definitive answer. I'm hoping this post will provoke a brainstorm that will generate the best idea. However, I will set the ball rolling with my (current) take on this predicament.

My qualifications to start this thread are poor: a tournament organiser (way back in the last millennium) and a pseudo-statistician (the heavily lifting done by stats programs) so I am in no way qualified to come up with the breakthrough concept, I am hoping however that the loyal fan base will jump at the opportunity to put their own stamp on the game by coming up with a better one to mine.



A Starting Idea

So Mage Wars is to be played in a Swiss tournament.
The number of rounds in the tournament = N (based on Con timetable scheduling)
If players > 2^N, we may get more than 2 players in the final round with 100% wins.
So there needs to be tie-breakers not just for draws but for all scores to decide who is in the final.

The result of a game is either
(a) a win = 2 points
(b) a draw = 1 point
(c) a loss = 0 points

Why no win premium? Because to get to "The Final", you have to win all your matches anyway.
And to get in the top half, you will have to win more than you lose.
Tournament players are not newbies turtling; they enter the tournament to play to win.
Half points for a draw is to salve the sting of only getting a draw due to defensive play.

We should strive for the same elegant single tie-breaker used for any match result.
There really is no need to have 3 levels of tie-breakers like in Magic tournaments.
This game does not warrant complicated mechanisms against a "Swiss gambit" etc.
After all, Magic has a $1 million Pro Tour; that is why it has algorithmic tie-breakers.

Let us assume every mage (package of spell points + life + channel + training + powers) is made equal.
(The nerfed Priestess may moan that some are made more equal than others, to paraphrase Orwell.)
In which case a lower life mage has traded "loss resilience" for other benefits (often greater channel).
Therefore any tie breaker that looks at "remaining life left" (life - damage) would be intrinsically flawed.

Let us assume the spell level of every object (any permanent in play) is indicative of its power.
Is this fair? Probably not. But it's elegantly simple and investment is definitely correlated to level.
And we can see that investment is also correlated to a spell's power, benefit or utility (exc. HoB).



The Tie-Breaker Formula

At the end of every game, win, lose or draw, the following calculation must be made.

My positional score = 2x total of the levels of all objects I own (not control) in play at game end
My aggression score = damage inflicted on opponent at game end (may not exceed total life)

My tie-breaker score for this game =
my positional score + my aggression score - opponent's positional score - opponent's aggression score

Check: my tie-breaker score = my opponent's tie-breaker score x -1

My tournament tie-breaker score is the average of every round where I have an opponent who plays legally.
You do not count rounds where I have a bye or where my opponent does not show or is DQ'd.

One final rule: "The Final" has double the normal time limit to ensure it rarely ends in a draw.



Some Objections (preemptively discussed)

Why 2x levels for positional score?
Well, for a start, 1x level would be far too little to make any difference in most game ends.
Attack and healing is roughly equal to 4x level as a one-shot benefit to aggression scores.
But objects have persistence; their existence helps gain more board control, removing enemy objects.
I wish I can demo a half-life formula or binomial infinite series to prove that 2x is a good approximation.
I confess it's based on intuition and some scenario testing of hypothetical board positions and damage.

So what about damage or conditions on objects?
Well, remember that if you win with 1 life left, you still win = 2 points, kerchingg!
So having objects with 1 hit point left follows the same threshold include/exclude approach.
As for weakness, rot etc. shame on you for not finishing off those objects that can hinder etc.
It would be needlessly complicated defining non-temporary conditions, applying minuses etc.

So what about upkeep costs?
Control spells with upkeep costs (because of the points swing) are nerfed by the "own" rule.
For enchantments with upkeep costs (for either player), its owner gets points for its benefits.
We assume that any owner upkeep cost is a persistent part of the cost of casting that spell.

What about channel mana advantage in a drawn game?
Ah, remember the mage package means you traded loss resilience for more channel etc.
If you built up greater channel, those conjurations, enchantments and equipment are objects,
So they are contributing to your positional score over your opponent.

What about healing being more efficient than attack spells?
Ah, the big one. Healing does not need to penetrate armour.
Group Heal is selective while Firestorm, Ring of Fire etc isn't.
I contend a tournament format where draws don't win you tournaments solves this,
Healing may not need to penetrate armour but it heals no conditions.
Damage spells often impose advantageous effects for that game state.
Both have an opportunity cost and are very short-term burst plays.
But healing only maintains the status quo, excess healing is wasted.
Excess damage is also wasted but that object is no longer in existence.
As for Ring of Asrya and Divine Reward, these are "Priestess package" benefits.
Because we don't look at remaining life, Divine Reward will only help avoid loss.
Also every mana spent on healing yourself is not improving your position
Meantime the opponent is improving board position and has a chance to win.
Why would a good player over-commit to a healing strategy if it only avoids a loss?
These are not newbies turtling, these are tournament players wanting to win.

The bottom line is: to win a tournament, you will need to win every round.
Healing does not help you do that. It just helps you potentially avoid a loss, at a cost.
If you want to aim for mid-table mediocrity with good tie-breaks, go healing for draws.
Healing will be used for board control by saving assets, only at a pinch to avoid a loss.

The reasons behind what I tentatively propose should be obvious,
It is designed so a mage who trades life for board control, who should win if it continued, is rewarded,
It is designed for transparency, simple addition and subtraction maths, not some mysterious algorithm.
But most of all, it is designed so that the margin of victory of all matches becomes cumulatively relevant.
It does add a small bit of simple maths at game end, both players checking to see they reconcile totals.
The current horrible solution is extra admin throughout a game of keeping track of total damage taken!

[See post below on 80min game length and Planning timers for more on this approach]



So there you have it: a "thought-starter" in Brainstorm Facilitator-speak,
It's by no means a fully-formed idea, It's just out there to be shot down.
However, I'd really like to start a discussion on possible tie breaker mechanisms.
The game designers are reading this so why not put YOUR IDEA out there too?
Who knows, you may get the thrill of boasting "that was my idea, you know"...

So come on, people, what tournament tie-breaker would you use for Mage Wars?

[ringkichard]

Of all the point systems that evaluate game state, this is my favorite I've seen so far because it's pretty simple. But the system it's based on assumes that a level 1 permanent is worth 1 damage. That exchange rate is at the heart of the question posed by Mage Wars. Is a level one permanent worth one damage? Is it a good idea to cast this Bitterwood Fox, or not? I'm concerned that this tiebreaker system would encourage distorted play.

Also, it doesn't do anything to reduce the likelihood of drawn games. I hate draws. And I think 60 minute rounds are going to cause a lot of tie games for control books, which biases book selection toward agro books like Forcemaster and Warlock. Your tiebreaker is for determining strength of schedule, right? The GenCon tiebreaker is to eliminate drawn games. It's considerably more aggressive.

So in the interest of reducing draws, I'm going to propose something (re-post from another thread) a bit more radical:


tl;dr: record damage done for a three round tiebreaker instead of for the whole game, and use that to break tied games.

Speaking of tiebreakers, I've given a lot of thought to an alternate tiebreaker for tournament games that go to time. I share Piousflea's and others' concerns that the current "Who did the most damage this game, total?"  tiebreaker system heavily favors aggressive decks like Forcemaster and Warlock, and adds an additional level of bookkeeping to tracking damage, which slows normal play.

If a control book starts the game by taking 22 damage in the first forty minutes, but thereafter kills all the opponent's creatures, takes no damage, gains 1 health and is does 2 points of damage to the enemy each turn, it's clearly winning. And it would be very likely to win if given enough time. But it's going to have a hard time proving it with the current tiebreaker system.

At the same time, I don't want to try to judge the game-state with a points system that awards victory points based on the levels of creatures and conjurations in play and tries to compare that to life totals (because who can say, exactly, how many points something should be worth?). I also don't particularly like the, "Everyone starts losing life each turn," elimination methods that try to impose a game shortening condition on play, because winning under those circumstances is very different from winning during normal play, and can distort the metagame badly. Warmachine's experience with "Victory Point Sniping" comes to mind, where one player does a little damage, then retreats for the rest of the game, and thus wins on tiebreakers.

But really, what I want to know isn't, "Who has done the most damage?" I want to know, "Who has come closest to winning". To use a mathematical analogy, I want to know the slope of the direction the game is heading, not the sum of where the game has been. If one player has an overwhelming creature advantage, I want to give them a chance to use it and prove their strength, while at the same time giving their opponent a last opportunity to win the game on the back of the damage they've already done.

Here's what I propose:

• The final portion of every tournament game is reserved for tiebreaker play. In a 90 minute tournament game, the final 15 minutes is reserved for tiebreaker play. In a 60 minute game, the final 10 minutes is reserved. Prompt play during tiebreaker game rounds is strictly mandatory.
• When the expiration of normal time is announced, the current game round is finished and the final three rounds of play begin. The game will end when the third game round finishes, or when final time is called.
• The three rounds of Mage Wars Tiebreaker play proceed normally, with the exception that each player must record the damage done to his or her opponent. Each round, the player who does the most damage to the opponent's Mage (including loss of life, tainted, etc) wins the round and one of three possible tiebreaker points.
• After three tiebreaker rounds, the player with the most tiebreaker points--two out of three--is the winner.
• If, during tiebreaker rounds, one player takes damage in excess of his or her remaining life, that player loses as normal.
• If a tiebreaker round is tied because neither player did any damage, or because the players did equal amounts of damage, no point is awarded.
• If, at the end of tiebreaker rounds, the players are tied in round points awarded, the winner of the game is the player who did the most total damage summed over all three rounds
• If both players did the same amount of damage totaled over all three rounds, the game is a draw.
• If tiebreaker time expires without the completion of all three tiebreaker rounds, end the current tiebreaker round as it stands, and score it as above. The player who has done the most damage in the portion of the round that has been played scores the point. The player with the most tiebreaker points is the winner, as usual. If the points are tied, check total damage done during tiebreaker play. If that is tied, the game is a draw, as normal

My thinking is that while 50 minutes may not actually be long enough for a control book to win, it should certainly be long enough for the control book to stabilize the game and be on its way to winning... if that's what it's going to do. If the control book can limit the damage from the agro book's last desperate gasp to kill, for two out of three rounds, it probably has the game well enough in hand that it would eventually win.

The agro book, on the other hand, is given three more rounds to kill the opponent, or at least to prove 2-out-of-3 that it is still capable of putting up a a superior fight and is not controlled. If it can do that, it wins.

Of course, either book can win the game outright--by killing the opponent-- during the overtime turns, and may very well do so now that both books must concentrate on doing at least some damage immediately, and cannot win by defense alone.

I settled on three and a half rounds of overtime (finish the final round, then take three more rounds) because it is a relatively short amount of play that two skilled and motivated players will be able to finish in the time available, and because it prevents a single surprise unrepeatable nova-damage turn (e.g. Wall of Thorns + Forcepush) from deciding the game. Whatever a player does to win, they must be able to do it twice (or do it well enough to kill the other mage). 3.5 rounds is also provides a fair distribution of initiative between the players. Each player will have the benefit of initiative twice during the overtime period.

Any thoughts or improvements?

[DeckBuilder]

Hey, I have now edited my formula to 2x level with an extra paragraph about it.

Your 3-round tie-breaker is indeed a radical idea!
I like the thinking behind it: end game threat level is all that matters.

I don't understand why you don't go straight to a sum though.
Unless there is some tactical significance.
As I am adopting an escalating damage approach.
You are adopting a flat damage approach,
Whose approach proved superior for that crucial middle round?

Your great idea still only deals with drawn games.
I highlighted that you are going to need tie-breakers for all matches.
What happens when you have 3 players with 100% wins in the final round?

Anyone else got ideas?

Remember: the First Rule of Brainstorm Club is "no idea is a bad idea"...

 

[DeckBuilder]

Hi kingrichard

After sleeping on it, I've concluded whatever mechanic you apply to make a tournament game fit 60-90 minutes, this will distort play if the game looks to be ending in a draw.

You have a great idea there. I love the concept of "it's actually your most recent damage output" in the delayed control game vs. "last chance to finish off mage" in the diminishing aggro game.

However, would this not lead to saving your 3 Battle Furies (and 3 Retaliate vs. Aggro Melee Mage) until this End Game phase? Not to mention the Control mage accumulating mana to Banish threats for most of this window. Although both distortions come at a price of opportunity and tempo loss.

Also, I think I've demonstrated you need a tie-breaker for all games in a large tourney, not just a decider to turn draws into win or lose.

You approached this as trying to find a winner in a draw because Control does not have enough time to win. How I approached it is with making the game length long enough for Control to win. Then players will always gain 2, 1 or 0 points. All the position/aggression tie-breaker does is rank you among equal cumulative scored opponents in the Swiss tournament.

So how long does it take for Control to win? I suggest your Control strategy should be possible to win in 80 minutes (10 minutes between rounds ) as long as Planning is timed and slow play is monitored (as with Magic).

Planning on timers is how we play using phone stopwatches: 90 seconds to prevent AP. If you have insufficient cards picked, tough. With this rule, you find players looking through spell books (not allowed to pick) during the opponent actions. "Unforced" errors creep in. It's the only way to fit in Control games in a reasonable time limit.

Having 90 minute rounds means less rounds in an all-day tournament which means tie-breakers become more important (assuming the game goes ballistic after its Origins award win hopefully).

I think the secondary nature of my tie-breakers (win 2, draw 1 or lose 0 is primary result) will be less distorting. Effectively, all I advocate is longer game times and a tie-breaker system for ranking equal tournament points.

The key difference between our approaches is I accept a draw as a valid result (my Magic background), try to accommodate Control via game length and Planning timing and only aim to provide a method for ranking players on equal scores. You are trying to adjudicate a draw game into a win game to either player. That's dangerous as it may foster ill-will.

I am a big fan of your method. It is so much better than the horribly inelegant GenCon method, even with the burst spells end play distortion.

I just think a paradigm shift into 3 results possible, with draws less common than the other 2 but prevalent (as in Magic) with max. game length extended and Planning AP appropriately regimented, may be the way forward.

Yes, it means Aggro players who win quickly will end up playing a short other game between 90 minute round starts This happens in Magic tournaments all the time.

Whatever Arcane Wonders chooses, they have to plan for large tournaments where tie-breakers decide who is in the showcase "The Final".



So, anybody else got a solution to the tournament tie-breaker issue? C'mon, I'm sure there are great ideas out there...

[ringkichard]

I don't understand why you don't go straight to a sum though.
Unless there is some tactical significance.
As I am adopting an escalating damage approach.
You are adopting a flat damage approach,
Whose approach proved superior for that crucial middle round?

One reason I'm going with best two out of three, rather than flat sum, is that I don't want the player who starts the Tiebreakers with initiative to go full nova (e.g. 2x Hurl Boulder) on the first turn and then play total defense now that he's used up all his direct damage. I'd much prefer the incentive to maximize damage over the course of three turns, which gives the other player a chance to respond when he gets initiative, and also increases the chance that one of the two players will win normally without resorting to the tiebreaker.

Your great idea still only deals with drawn games.
I highlighted that you are going to need tie-breakers for all matches.
What happens when you have 3 players with 100% wins in the final round?

Honestly, I think tournaments should never have so many players that log2(players) is greater than the number of rounds. If that can't be helped, I think the final round should be played as a cut, with the rest of the Swiss players going home. If that fails, just use strength of schedule.

You're also probably right that control players will probably start planning their Big Push for the start of overtime. I think the real culprit here is the 65 minute games that are going to be required to fit Deck Check and 5 rounds into 6 hours. Any tiebreaker strategy is going to cause some distortion, because the ultimate cause of the distortion is a failure to finish the game in the allowed time. I know this method isn't perfect, but I'm proposing this tiebreaker method because Arcane Wonders clearly will not be allowing drawn games, and I think this will distort things much less.

[DeckBuilder]

You misunderstand, I like your method, it's a big improvement. I think it may favour Control more but that may be a good bias to address the current imbalance of being able to win tournaments with quick win Aggro decks whilst Control will often have to run the gauntlet of your End Game method to scrape a win.

As for the current clunky extra housekeeping method, I just wonder why they didn't instead rule thus:

"In Tournament play, if a mage heals from any source, a mage increases his life by the healed amount instead of reducing damage. If a game ends unresolved, the mage who has suffered more damage loses and his opponent wins. If tied on damage, the mage with higher life wins."

Now that executes the current tie-breaker (which I don't agree with on so many levels) in a far more elegant fashion that does not slow play with extra housekeeping and possibility of honest error. Of course, it doesn't solve its shameless promotion of Aggro spell books. Both of us are trying to address this imbalance by evaluating the Control "tipping point" in different ways.

I think it's the inelegance of their tie breaker that irks me most: it's a crime against the intuitive elegance of this game!

[ringkichard]

I'm sorry if my tone came off as hostile 
I focused more on your constructive criticism than your compliments because my ego is big enough already.

I think you're right that if control makes it to the end of the timed round, that my tiebreaker method is more likely to award the game to the control player than the agro player, but I think that's because if a game goes for an hour, it's more likely that a control book will beat an agro book overall. Agro books tend to win pretty quick, or not at all.

I do think that there may be some pro-agro bias in tournament play on the part of Arcane Wonders just because playing 5 rounds of control at two hours each round doesn't really make for a compelling tournament experience. At least, not the way tournaments for MtG are run, which are clearly the model here.

[DeckBuilder]

I can't help but feel a bit of a fraud: we are both so obviously fans of control (although I totally respect aggro, I just suck at rolling dice). So we are hardly unbiased in our opinions!

I am bit disappointed nobody else has ventured any better ideas. However, I think I will proudly plant my flag on a principle that you touched on...

"If an aggro book has failed to beat a control book within c.65mins, because aggro diminishes in potency with time while control increases with time, the aggro book must be prepared to demonstrate that he is still a lethal threat to claim the win, Else by definition of control, the aggro book should not be surprised if tournament guidelines favour the control game which by definition over-takes the aggro book going forward."

I do not think this principle is anyway biased, just common sense definitions of Aggro and Control.

If we assume persistent threats early to mid game are far more threat efficient than burst plays, then healing incantations (even if more efficient than attack spells) is just an inefficient counter because it is a burst play, not persistent like a Regrowth.

Maybe the issue is Regen 2 on your high life mage: is it too good? Because the mathematician in me just can't see how spending your mana on burst heal spells can be better than swarming or buff then fury on champions.

Or am I very wrong here? Is burst healing really that distorting to tournament play?

[ringkichard]

Heh, actually, I have the following three books built out of my set, right now: Beastmaster, Warlord, Forcemaster. And I'm taking the Forcemaster apart to build a combo book, maybe (I'm struggling with it). I'm not a control player. This is because my regular opponent loves control and is better at it than I am.

Not to say that I don't enjoy control, (I do tend to play mid-range Argo Control with disruption) but really I just don't want tournament play to wander down a blind alley where all the books are hyper-agressive agro because that'll be its own ecosystem and not really relevant to the game the rest of us are playing. Which would be a shame, because the game the rest of us are playing is really fun!

And no, I'm pretty sure that burst healing isn't actually any good at the tournament level, except for healing buffed/big creatures when you don't have the tempo to wait a turn while you summon another one.

[NorthernMage]

On a slightly different note.  Swiss is a great system, but what I've seen at Orgins a double elimination tourney would be better. After some players lose twice they just leave and that tends to leave open games in the later rounds. Making it less fun and watering down later rounds for some players.  It is also very easy to track and build a bracket for double elimination.  I know it Way to late for Gen Con, but any thoughts of this in the future?

[Tacullu64]

I'm going to cover my thoughts on several aspects of tie breakers so this might be a bit lengthy when I'm done. If I start to ramble off on tangents as I am wont to do, it will definately get lengthy.

Current Tie Breaker
My first topic is the current tie breaker. My gut feelings are to dislike it. On the surface it appears to give aggro builds and mages who are more efficient at dealing damage an advantage. The biggest beneficiary is the warlock. He has his battle skill, great weapons only he can use, in school access to burn effects (shared only with the fire wizard), in school access to efficient damage dealing creatures, and the cherry on top, in school access to direct damage curses. I would say the warlocks closest rival is the beastmaster of straywood. The beastmaster boast his own battle skill, the best buffs in the game, in school access to efficient damage dealing creatures, in school access to the bleed condition, and the quick summon ability. Now I realize that any mage can go the aggro route, my point is some are significantly better at it. I'm looking at you warlock. It would seem that a tie breaker that so strongly favors a particular style of play and those mages that are best at it would unbalance tournaments.

So, what does the current tie breaker have going for it? If you could forget about the in game bookkeeping it would be simple. I really, really, really, favor simple tie breakers. The only other reason I can think of to keep the current system is that so far all we have is supposition that it is a bad system. There has not been enough tournaments to generate the kind of data we need to prove it is a bad tie breaker system. This alone may be enough to warrant keeping it for a while longer.

My Former Preferred Tie Breaker
My original favorite tie breaker was life remaining (max life - damage on mage). However two cards that I really like a lot from CoK ruin remaining life as a tie breaker.

The first is Sunfire Amulet. Sunfire Amulet is a cool card but if remaining life was the tie breaker it would be in every spell book built and could become a focal point of tournament games and this would be unhealthy. If AW is going to promote tournament play I wonder about the wisdom of creating cards that are designed to prolong the game. The time limit is already a topic of much debate.

The second card is Drain Soul, another card that may only be used by the warlock at the moment (is a warlock secretly in control of AW, he is currently getting a lot of love). I thought that the priestess was supposed to be the best mage at life gain, not anymore, with sole access to Drain Soul the warlock is king. The warlock starts with the most life at 38 and with quick six from DS he is at 44. The priestess will never catch him in life total. This tie breaker system favors the warlock even more than the current one. Best damage dealer combined with most life equals guaranteed win if it goes to the tie breaker.

Tie Breaker I Currently Favor
The tie breaker I currently like best is damage on the mage. The main reason I like it is that its simple. Another reason I like it is that it doesn't rule out healing. I don't understand why people are hellbent on removing healing as a factor in tie breakers, it is a part of the game too. Finally, it is fairer than the current tie breaker.

Boxing and Mage Wars
In my youth I watched quite a bit of boxing. It wasn't all that uncommon for one fighter to win more rounds but be on the ropes in the final round, close to getting knocked out. All he had to do was make it out of the round and he would probably win. Sometimes he would survive to take it to the judges score cards and win a decision and sometimes he would get knocked out and lose. In any case the judges score cards reflected what happened during the fight not who they thought would win if the fight continued. So what does this have to do with Mage Wars your wondering?

There has been some discussion about a tie breaker that would award the game to the mage who was going to win if the game continued. While I like the concept, the application is flawed short of a machine that can peer into the future to watch the demise of the eventual loser. The tie breaker should determine the winner based on what has already occurred. The goal of the game is to make the opposing mage's damage marker meet his life marker, not cast the most creatures, or have the least spell points left in your spell book, or whatever other condition can be thought up. Going in to the match both players know how much time they have to knock out the opposing mage. The winner should be determined based on what occurred during the time allotted. That is why I liked the life remaining tie breaker because that would tell you which mage was closer to victory when the match ended. However that is no longer viable due to the reasons I mentioned above. That left me believing the damage on the mage when time is called is the fairest to all mages and styles of play. Is it perfect? No, but your goal should be to make sure the game doesn't go to the judges scorecard.

That I believe is everything I have at the moment.

[ringkichard]

Tacullu64, even though I disagree with your conclusions, I really like the Boxing metaphor. And I can totally see how you'd prefer a simpler tie-breaker method; I'd prefer if my proposal were simpler, too. Basically, what I'm saying is that even though your post is largely a criticism of my proposal, I think you're mostly right, and you've clearly given this a lot of thought. That's really the best reception anyone can ask for an idea; thank you.

I think that your comparison of the current Mage Wars tiebreaker system to a boxing judge's recorded notes of the whole fight is right. And I think you're right that boxing's method of scoring each round is very similar to tracking all damage done throughout the game. I think it's the right metaphor.

I just don't want tournament Mage Wars to be boxing, and I think that boxing is exactly what we'll get under the cumulative damage systems. If we score the tiebreakers with an emphasis on counting every hit, then that's what we'll get: spellbooks designed to hit all game long. It'll be a Warlock party.

Outside of tournament time limits, though, high level play does include investment cards, and then a large part of the tactics and strategy of the game is about the conflict over those investments. How many Mana Crystals can I conjure without dying? Will a Wall pay for itself before it's destroyed? Can I keep a Wand in play long enough to justify the cost? If a game lasts 2 hours instead of 1 hour, control books can actually be pretty good, but if all games have to be over in 60 minutes we're going to see a whole lot of agro books, and not a whole lot of Mana Crystals.

I think our biggest point of disagreement is well summed up in your last line "but your goal should be to make sure the game doesn't go to the judges scorecard." I emphatically disagree!

First, I disagree because that is not what will happen. Tournament play is about winning, and tournament players will do what is rewarded, not what is in the spirit of the game. If it's possible to win the game by being ahead in life when the time runs out, then some players are going to play with that victory condition as an objective. If we give players a choice of routes to victory, some are going to chose the one we don't like, but it's our fault for giving them the choice.

The solution to this is to make the tiebreakers as much like the rest of the game as possible, so that gaming the tiebreakers requires actually playing the game.
 
Second, I think goal of tournament Mage Wars  should be to win under the rules of the game, and the game as published doesn't have a time limit. Now--for practical reasons--tournament games do need to end sooner than 120 minutes each, but as much as possible players should be playing the same game at home and at tournaments. They should be playing to win the game, not to beat the clock. We need a fair way to evaluate what the state of play is when time is called, not to create a situation that favors agro because it's more convenient for tournament organizers.

Having control books in the meta promotes a sustainable ecosystem because it keeps agro books honest: the reason push-wall isn't a dominating strategy is that sometimes your opponent is going to put on a Nullify, a pair of Eagle Claw Boots, or some armor. If all your opponent ever did was charge in and cast Bear Strength, conjuring up a Wall of Thorns + Force Push combo would be an easy-button.

I've ended some good games in 25 minutes. But I've also played games that lasted two and a half hours, and with the same spellbooks and players. When I hear people say that they're ending all their games in under an hour, what I hear is that both players are deliberately not playing any of the slow but successful strategies like walls, guarding, attrition, intercept creatures, mana denial, etc, etc.

Those are all part of the game, and they should be present at tournament level play. We shouldn't drive them out just because we want the games to be over fast.

[Tacullu64]

Thank you for taking the time to respond ringkichard. I can tell your ok with the tone I took, but I'd like to say that the point of my post was not to criticize your idea. The post as a whole contained my rambling thoughts on tie breakers for mage wars. The last paragraph while critical was all that was directed at your tie breaker system. I really do prefer your concept to any other. At my core I'm a guy that just wants to see justice done. I want to see the player who deserves to win actually win. I have a couple of issues with systems like you are proposing.

The first which I believe you see my point on is that it is very complex. Generally speaking simpler is better. On a lesser side note how much time will a system like this take to adjudicate? Enough to effect the number of rounds in a tournament? I don't know? If a system like this is implemented would it encourage stalling in order to garner a win on points? Again, I don't know but I would be concerned about the latter.

My second concern on this type of system is its ability to actually predict who would be the eventual winner based on an arbitrary weighting system.  There is a lot of ebb and flow in mage wars, if the game was played out maybe the player with the weaker position when time was called came up with a brilliant idea and turns the game around or the person with the stronger position makes a critical mistake and ends up losing. We just can't know in many situations what would happen unless the game is played out. What if the scoring system is not an accurate predictor and continually rewards the game to the wrong person. These are the thoughts that nag me when I think about the viability of actually choosing the correct winner.

My third concern is more philosophical. It involves who deserves to win. Is it the guy who did enough during the time allotted or is it the guy who came on strong at the end and would probably win if given more time. Ultimately, since there is a time limit, I would say it is the guy who is closer to victory when time is called, more damage on the opposing mage, as opposed to total damage dealt ( should be life remaining but that is not possible anymore).

I don't want the topic of the Sunfire Amulet to get lost in the larger points of the discussion. I like the concept and I like the card for non-tournament play. If there weren't tournaments I would leave it just like it is. Since there are tournaments I would make cards like Sunfire Amulet epic. Not because it is overpowered because its not, but so that it only has to be dealt with once during tournaments. Right now time is a big issue in MW tournaments. I don't think most people would play it in tournament builds at the moment, but I do believe it helped make my favorite tie breaker unusable. Magic went through this exact learning curve with card design once tournaments were introduced. If memory serves me correctly they even banned a card because it made the game last too long not because it was OP.

[Tacullu64]

I was rushed for time on my last post and forgot to address one of your points.

First I would like to say that it seems we have the same goals and agree about almost everything except how to achieve those goals. I very much want to see control builds be a part of a healthy environment.

It has been my experience that it is not uncommon for aggro builds to dominate early in a games life and after significant expansions are released. This usually results in a natural self correction that brings up the level of control builds to near or maybe even a little bit better than that of aggro builds as control players fine tune their craft. If this self correction doesn't occur our concerns will be justified. I don't think we have reached the time yet in MW for that self correction to happen. If it doesn't happen I would prefer that Arcane Wonders correct the imbalance gradually by releasing cards to strengthen control builds as opposed to adjudicating it through a complex tie breaker system.

I do think the current tie breaker probably needs some tweaking to something more fair, just not a complete overhaul.

[Amadseer]

I'm thinking that tiebreakers could be determined based on board position and superiority.

Here's the thing -- as someone commented earlier in the chain, tournament players are playing to win. Whether that's through control or aggro, they have to be doing something, or else they will just be chewed up and destroyed, long before the time limit.

However, since the ultimate win condition is vanquishing the enemy mage (through points of damage), this favors an aggressive play style. We all know that's not the only way to win, just the most clear path.

But, even without the damage, the game is still about owning the board. It's just that owning the board is the launchpost to allowing a Mage, its creatures, and its conjurations to dealing more damage.

That's why I think zone superiority can help break ties.

Here's what I propose.
• Each controlled zone is worth 3 points. A mage controls a zone if the zone has
  -- A zone exclusive conjuration
  -- A friendly creature (and there are no enemy creatures)
  -- A revealed enchantment (and there are no other conjurations or creatures in the zone)

• Each contested zone is worth 2 points to each player. A zone is contested if:
  -- There are both friendly creatures and enemy creatures in the zone
  -- There is an enemy creature and a friendly zone exclusive conjuration)
  -- There are enemy creatures and friendly zone enchantments

• Each creature you control is worth 1 pt.

• Empty zones are worth 0 points.

Each players adds these up, and whoever has the highest point total wins the tie.

Thoughts?