any heroes vs battlesites head to head

[breadmaster]

so i've been looking at past tournament results, testing the generally accepted theory that sites vs anys are 'more or less 50-50'

so far i've looked at the results from 4 live tournaments

may 2011-toronto: anys 9-sites 9, with a playoffs record of A:1, S:3
june 2012-toronto: anys 14-sites 4, no playoffs (round robin format)
oct 2012-toronto: anys 12-sites 5, with a playoffs record of A:3, S:0
march 2013-buffalo: anys 15-sites 9, with a playoffs record of A:3, S:1

cumulatively, this gives an overall record of A:50 to S:27, for a winning percentage in favour of anys by 65%.  the playoff numbers were 7-4 anys, for a percentage of 64%.

these initial numbers don't support the 50-50 theory

however, there are a few tournaments missing. if anyone has the results round by round for the following, i'd love to see them

august 2011-toronto
summer 2012-california (i think...this one was hosted by judgeone)

these were both won by battlesite decks, so the numbers may equalize

there are also 4 different online tournaments i can check out, but there are some information gaps here.

opo 1: i don't have all the builds, but only 8 teams played in single elimination, so there are only 7 matches
opo 2: this was the big one where we had 10 players in a round robin.  all the information is there; i just have to go through the topic to dig it up
opo 3: this had a really uneven amount of matches played, and the decks are incomplete.  i may be able to cobble it together, since i know i played everyone at least once
opo 4: this was the random character one.  i'm not going to use this data
diceKs draft tournament:  this is a tricky one too.  it never finished, but there were alot of games played.  the problem is, the build was alowed to change between weeks, so i'm not sure if people recall what they used

if anyone has the data for the 2 live tournaments, or any better recollections from the online ones (i think you can even resume to see what teams were used), i'd love to see it

[Torch]

Of the 'Anys' decks, how many had DoW? 

[Onslaught]

Battlesites are significantly stronger than Any Heroes, roughly 70% of 30+ person tournaments in the post X-Men environment were won by battlesite decks. The advantage becomes even more marked when the top cut is double elimination. Sample size from the tournaments you listed is too small to draw any general conclusions (no event larger than 20 people, some of them used house rules (?), only a handful of tournaments to choose from), but it does give good insight into your specific metagame. If you are playing an Any Hero deck that is tuned to beat other Any Hero decks, you would have a significant advantage in your tournament scene since battlesites are so underrepresented.

At the same time, making a deck (regardless of Any Hero or battlesite) that preys on the inherent weaknesses of an Any Hero deck also seems like a good choice. It is extremely difficult for Any Hero decks to defend level 10 teamwork followups, and it is outright impossible outside of Guardian if you aren't running one of the EE characters. Spawn, H4H, Prof X, or any other dual stat 8 character that can allow for easy level 10 followups should be dominant personalities in your environment.

[Jack]

If the data isn't in the wiki, I don't have a record of it.

http://overpower.ca/wiki/OverPower_Toronto#Tournaments_2

[BigBadHarve]

I tend to agree that Sites have the power here overall. Ultimately, the variety of specials combined with the ability to play them sooner is a significant advantage. The only major strike against sites is that they are significantly easier to shut down by an opposing player. There are more counter-site strategies than there are count-Anyhero techniques, making it a bit of a turkeyshoot for someone who wants to build an anti-site team. Of course, if you gear your deck to taking down a site then you are probably leaving yourself vulnerable to another strategy.

The main trouble with DoW is that it's a crapshoot. Yes, it can and does cripple teams, but I've noticed the majority of the time a team that can weather and get rid of DoW will usually come back and take the win.

This variety is what makes the game so compelling even after all these years.

[breadmaster]

anecdotal evidence from the past has no place here

i think you guys are really basing your thoughts on what happened 15 years ago.  since i never played seriously then, i couldn't understand why y'all thought that they were even, when my perception was that a good any team seems to DOMINATE a good site team.  so i looked up the head to head record in documented tournaments since the revival.

the numbers really support this.  77 matches is by no means definitive, but it's quite a start. and 2/3 is about as far as you can get from 50-50 before it becomes insane to use a battlesite.

as far as not using the data because of 'house rules'...nice try.  the only tournament listed that used them was may 2011, which is also battlesites best showing so far head to head at 9-9.  if i did eliminate those games, the winning rate goes to 69% anys, and a whopping 86% in the elimation rounds

[Onslaught]

anecdotal evidence from the past has no place here

By definition, all data is "from the past." At least 8 of the last 12 regionals were won by battlesite decks. After that final round of regionals, there was a year or so of sanctioned tournaments hosted by dedicated players with a very high level of competition. The cancelled nationals at Comic Con was won by a battlesite deck (I think Morlock Tunnels). The OP CUP 1 was won by Ty using Citadel. The OP CUP 2 was won by Matthew Puckett (I may have his lineup somewhere but I don't know offhand if it was a site or Any Heroes), but the other top 4 finishers all had battlesite decks (one of which was AoA). A sizable Buffalo tournament hosted by the Bertrands featured  top 4 of all battlesite decks. An event in Chicago was won by AJ Kitzmiller, going 8-0 on the day with a battlesite deck (and I think Four Freedoms homebase). The Scarlet Witch Invitational was won by a battlesite deck. A Western Viginia tournament hosted by GDMJR featured Doug Simms vs Bluefire in the finals, both of them using battlesites. From five midlevel events in Ohio and Pittsburgh: 10 finalists, 9 battlesite decks, 1 Any Hero deck. These all took place during a time period where the playerbase was much larger, more active, incentivized by big prizes (including cash), and more informed than the tiny remainder of players that exist now.

Also, look up what "anecdotal" means since you don't seem to understand the meaning of it.

i couldn't understand why y'all thought that they were even

I don't think they are even, and objective analysis of tournament data shows that battlesites are strictly superior to Any Heroes. They are only "even" in the sense that Any Heroes are a legitimate counter to certain types of battlesite decks, giving the illusion of parity (at least in terms of prevalence). Battlesites are applicable to every type of deck: offensive, defensive, gimmick based, etc. Meanwhile, Any Heroes (at high level play) are restricted to two extremes: super redundant decks (something that loses only 4-5 unique cards when a character dies, generally very offensie) and super defensive decks (with criss cross pure ADs or an EE). In any given playerbase, there is going to be a majority of players drawn to one of the latter two archetypes because they are easier to play, easier to build, and generally consistent. The Any Hero decks that did manage to win regionals mostly followed these rules (an X-Man/Spawn team, a H4H/Brute/Juggs team, a Sentinels team though this shouldn't count since it won through collusion, and a Zealot/Serp team). When I see lineups that are begging for some battlesite defense yet run Any Heroes due to fear of being DOWed or something, I really wonder how such lineups are able to keep a character alive for more than a turn or so. That's fine if your deck is built with a low special count, but generally just tossing Any Heroes into any lineup that isn't catering to their advantages is a bad decision.

the numbers really support this.  77 matches is by no means definitive, but it's quite a start. and 2/3 is about as far as you can get from 50-50 before it becomes insane to use a battlesite.

Citing the incidence of Any Hero decks doesn't really tell you anything about their usefulness, it just gives an indication of what people in your tournaments prefer to use. Descriptive statistics from such a small sample size are completely irrelevant, and if you are trying to make qualitative analysis from your tiny amount of data then you should be getting at the one thing that actually matters: which deck (whether it be playing against Any Heroes or Battlesites) has the highest overall winning percentage? I think Jack lists the round by round results from your more recent events, so if you wanted to go through and track individual results of each player it would be a good foundation to start gathering detailed analytics for your tournaments.

which is also battlesites best showing so far head to head at 9-9.  if i did eliminate those games, the winning rate goes to 69% anys, and a whopping 86% in the elimation rounds

Again, this doesn't tell you anything about the actual usefulness of battlesites or Any Heroes. If only 5 battlesite decks enter a 17 person tournament and none of them win the elimination rounds, you don't really learn anything from it.

[breadmaster]

all irrelevant.  the lady doth protest too much methinks

i don't care about what decks won or lost tournaments.  i care about head to head matchups between anys and sites.  if all that data isn't available, i don't use it.  if 75% of the decks are sites, of course sites are going to win the majority of tournaments.  same with anys; more decks=more overall wins.  i don't care about that.  only head to head 

all data head to head is useful.  there are good/average/bad site teams, and good/average/bad any teams.  over 77 games that all smooths out, and we're left with an unmistakeable trend.  of course you don't want to acknowledge that, you always crap on the data when results don't go as you expect (which is quite often)

as i said, there are 2 other live tournaments that i'm hoping to get results from, so we'll see if the numbers change

[gameplan.exe]

so, breadmaster, are you saying the 77 games you pulled are games featuring a Battlesite vs. AnyHero deck?

also, Onslaught, isn't it accurate to say that the vast majority of tournament decks were using Battlesites? if that's true, shouldn't we expect that a Battlesite team is more likely to represent in the finals? this would basically just be the opposite of what we're seeing now (when AnyHero decks seem more prolific).

if this winning percentage is truly based on 77 games that featured a B.S. vs A.H. deck, then that seems the most telling. I mean, sure, in the grand scheme of things it seems like a small sample size. but if you were doing controlled tests of games, "playtesting" 2 decks against each other, and 77 games into it you had a 67% win pct. for one of the decks, wouldn't we all call that pretty conclusive?

[Demacus]

The biggest problem with the sample size is the fact that of 77 games, a 67% win ratio comes out to between 51-52 wins...  in a sample of 100 games, the remaining games COULD all go to the other side, which would in fact bring the win percentages to 51-49 or 52-48, which is much closer to what was originally being tested, a 50-50 win ratio of one over the other.

Even in that scenario, with Any's coming out on top 2-4% more of the time isn't enough of a ratio to justify stating that they are stronger then Battlesites.  You would need a few thousand match ups for real, definitive evidence that one build type is stronger then the other, truly.  If you are really not willing to look at past records, and are only going by the data created by this community, your data will be incomplete for some time, and you really shouldn't be drawing any solid conclusions just yet.

[HotDogCart]

I never understood where the idea of battlesites being equally as powerful as any heroes originally came from? This must have been before DOW was regularly played, right? With any heroes you throw the top 10 powerful any heroes in your deck and almost any deck is more powerful then a battlesite is going to be. Battlesites take a lot of time and planning to make, it is like having to make two decks, to play one, and then remember all the cards you are playing with. Of course it is all a matter of who you are playing with, but generally speaking I have always though that any heroes are overwhelmingly more powerful then a battlesite is going to be.

[breadmaster]

nic: yes, anys vs sites were the only games looked at

dem: i shouldn't draw any conclusions from 77 games?  that is by no means an insignificant sample size.  these are also all from tournament games, where we should assume people are building decks that give them the best chance to win.  of course sites COULD go on a 25ish-0 run in the last sample (not sure why 100 games is some 'magic' number though).  i don't think it's likely

also, i'm willing to look at battlesite vs any data from any tournament, as long as all the data is there

htg: exactly my thoughts.  this is a trend i've observed, and everyone has disputed the last couple years.  well, i decided to gather the data and see what that indicates.  rather than cause people to re-evaluate their notions, they come up with reasons why the data is flawed.  human nature i guess

there's a couple reasons i feel this way. 

DoW: enough has been said about this.  it's an unstoppable card that is worth at least 2 missions (ultra-conservative), probably around 3-4, and often the game

entropy field: people say this is balanced by gods of stone-i disagree.  most any teams will use team overpower, which allows them to place 1 any in the round this comes up.  they can also place new universe on a character.  so while gods of stone will hurt the round it comes, they can carry some advantage to the next round.  also, site decks traditionally have 2 specials/character and a beyonder (and some characters with 3).  that's 13-14 cards, vs 9-10 anys.  therefore, one should expect more discards when enropy field comes, vs when gods of stone comes. what that difference is worth in mission cards...not exactly sure.  i know personally i've taken some real beatings with gods.  lets say 1 mission

so using this dirty math, we have at the minimum a 3 mission advantage (and i personally think it's more).  of course sites give you more versatility (as stated multiple times), but is that versatility enough to overcome those 2 cards?  data so far indicates not

[Onslaught]

You have roughly a 40% chance to draw DoW in the first three turns. Meanwhile, playing x3 of an activator like Colossus for Siberian Strength or Kingpin for his HQ gives an 82% chance of drawing a trump card in the first three turns. The average game lasts 3.5 turns. DoW, while unblockable, is mitigated in many ways over these first three turns (a KL, a GL, a concession while your placing spot on Team OP is taken, Gods of Stone, Down But Not Out being setup, etc). Also, since DoW is a game lasting effect, you obviously are not maximizing its utility if it is played on the final turn of the game. If you play it on a decisive turn 4 or 5, it is most likely a 2 for 1. If I went first on that turn and opened with an activator (a wise decision for a battlesite deck that has bet for the win with DoW still lurking), you get a 1 for 1. If you are venturing properly (aka aggressively) with a battlesite deck vs an Any Hero deck, you should be at a huge advantage. Are there games where you get randomed out by a first turn DoW and can't recover? Of course, but that's the price you pay for the tremendous advantages that battlesites have in every other regard.

Besides the head to head comparison of the best card from a site (HQ or whatever) vs. the best Any Hero card vs sites (DoW), the battlesite deck also benefits from the superior defense, flexibility, and customization that Any Heroes cannot match. Leech should never successfully land against a battlesite deck due to having 7+ avoids floating around in your deck at any given time, in addition to your regular team defense. Meanwhile, Leech hitting another Any Hero deck is fairly common, especially in the very early or very late stages of the match. Having access to unique effects like Obfuscate or a JW is a privilege that only battlesites have, which keeps Any Hero decks firmly in the "consistent/resilient" style of play. Most importantly of all, the huge increase in defense creates an insane amount of virtual card advantage. All things being equal, a battlesite deck can keep its characters alive much longer than an Any Hero deck, which is the purest form of card advantage in the game.

[Demacus]

@Bread - I was not attacking you or your project, just stating that with such a small sample size, each individual game carries too much weight to be reliable information.  If you read my second paragraph in the above post, I noted that you would need considerably more then 100 games to draw definitive conclusions as to which may or may not be the stronger set up.  I didn't say anything about 100 being a "magic number," it was just an example based on a 1 to 1 ratio. 

If I were to sample 4 games, each game would carry a weight of 25%.  With 50 games, each game is worth 2%. At 100 games, each game is worth 1%. A sample of 77 games, gives each game a weight of 1.29%. 

It's too easy to sway the percentages if you are below a sample of 1000, which would require 10 games to change a single percentage point.  10,000 is even better, as it would take 100 games to alter a single percent.  Within 77 games, someone's just plain old bad luck could affect your findings.   At least with a 1000 game sample, winning and losing streaks would have to be higher then 10 games to change the results of the findings.

[breadmaster]

siberian strength and HQs are negatable.  dow is not.  advantage dow.  also, nice try downplaying dow by telling of it's disadvantages at the end of the game.  even if it is drawn on hand 7 (which should be the last before power pack), it should give an advantage.  with 13 activators, i expect 1.9 activators/hand.  yeah, IF you go first, you can play an activator before it's out, but half the time you DON'T go first.  and the 86% of the time you draw it before hand 7, that advantage carries over into every subsequent round

dem:  agreed, more data would be better;  it's just not there currently.  but certainly trends can be observed from what is there