NTBB2.0/2014

[plasmoid]

The discussion with Dode and VoodooMike has certainly opened my eyes to the fact that a lot more data about BB is available now, than it was 3-4 years ago. I'm thinking working with this data could seriously strengthen the NTBB project.

This is my thinking so far on the matter:
IMO, there are 3 popular and distinct environments in BB. Tournament (low TV, a few skills, resurrection); TableTop (short term, season/trophy) and Survival (Long term/perpetual). I think it would be great if NTBB could take all 3 environments into account, making roster changes to teams that are overly strong/weak in one environment, as long as it would not push the same team outside of tier 1 in another environment.

I don't have enough data on TT league play, and I doubt that I'd be able to get it.
But there is plenty of data in FOL and especially Box for the Survival mode, and Ian has collected the NAFstats for the Tournament scene.

The objective of NTBB is and would still be to make more teams and tactics viable.
The CRP+ rules would still stand - this is about my NTBB roster changes. The CRP+ rules adress the long term anyway, in an attempt to change the metagame, while the NTBB roster changes are focused on the shorter term.
It is probably worth clarifying that:
*NTBB aims to improve the tier 2 and 3 teams. The goal is to make tier 2 teams close to the weakest tier 1 teams, while tier 3 teams are to be improved without putting them in tier 1. (For Tournament play, I can accept further increased parity, as that's what tournaments is about anyway).
*NTBB abandons the lifetime performance adopted by the BBRC. If a team is significantly overpowered or underpowered in any part of it's lifetime, then I consider it ill-balanced.
*NTBB does not use mirror matches for the calculation, even though the BBRC did. Mirror matches are meaningless data that will pull teams towards the 50% mean. Especially in Tournament (swiss) play this is apparant. For the most part this will make the NTBB definition of the tiers slightly narrower than the official ones.

Looking at Box+FOL and NAF stats, this is what I found
1) Tournaments:
Looking at the combined stats for resurrection tournaments using LRB6/CRP I get:
Wood Elf: 55.06 - 57.58 (5978 games)
Undead: 55.30 - 57.72 (6401 games)

2) Survival:
Amazon, TV 0-1500: 59.05 - 60.97 (9960 games)
Undead, TV 0-1500: 57.60 - 59.24 (13823 games)
Lizardmen, TV 0-1500: 55.17 - 56.89 (12720 games)

It might be worth adding that all the Elf sides move into tier 0 territory as TV climbs (Wood and Dark TV1600+, High 1700+, Elf 1800+). Orcs and Dwarfs drop below tier 1 in long term play (1600+ and 1800+, respectively), while other non-CPOMB-heavy bashy teams struggle, without being decisively outside tier 1 (Norse 1800+, Khemri 1800+, Undead 2000+).
While this is not the focus of the NTBB roster changes, I believe that the CRP+ will change the metagame in a way that will have a positive effect on this: CPOMB-bash will be weakened, allowing non-claw teams to better survive a scrum - while the diminished need for survival skills will allow teams to take more elf-fighting skills. (That, and SE/Bank will bother the elven sides). But admittedly that's all speculation.

So, NTBB would be looking to nerf Undead, Amazon, Wood Elfs and Lizardmen. (Orcs and Dwarfs would then go free). And Halflings, Gobbos, Ogres, Vampire and Underworld would get buffed, (the Slann change would then have to be abandoned, while Human and Khemri would not - being part of CRP+)

Cheers
Martin

I just might go looking for TT data for Dwarfs. The original LRB6 data had them suspiciously high, and I wish I had the means to look into that.

[Shteve0]

Good stuff.

One thing I'd raise:

I don't question the logic of looking at low TV / low # games etc for short term metas, and using that as a basis for short term nerfs/buffs.

I see that there's a brief discussion around high performing teams at increasing TV in your intro, and that's interesting; but what defines Survival as a format is that it covers a lot of games (not equal to high tv).

My observation would be that, in order to cover that format, an idea of which teams under or over perform in that format's defining range would seem appropriate. For example, if a team (WEs) overperform both in TT and Surv you will take a very different approach to 'fixing' them than, say, a TourneyT0 but SurvivalT2 team. You'll agree that it's no good applying a straight short term nerf to the latter if they collapse in CRP after 20 games, for instance.

That's all. Applaud the greater use of stats. Would also be keen to see some discussion around what yardstick range (%) you are aiming to bring each team's win ratio into at each format - you won't be able to measure it, of course, but I'm quite the fan of clarity of direction.

[spubbbba]

If survival is long term play why are your stats only for TV under 1500?

Lizardmen are only just over 55% and if this drops above 1500 then they would be fine, I’ve always seen them as a pretty balanced team. They only do well in tournaments where you can assign skills. Zons are even more vulnerable to the differences of high to low TV play, but I’ve always wanted them changed mostly due to them being the dullest roster in the game.

B is a pretty outlandish environment, part of the reason zons are so successful there is the vast amount of bash that love claw and lack tackle.

[Hitonagashi]

I wouldn't necessarily look to nerf Lizardmen.

They are strong at low TV (and the Box stats represent the sweetspotting that everyone does at that point), but they are one of the "Forlorn" FUMBBL races, which is in the 10 or so years that the site has been running, no Lizardman coach, however much they pimp their team has ever won a major tournament (of which there are 8 per year).

They fall apart very easily, and are a *very* hard race to modify.

[garion]

agree with everything hito has said. Lizardmen are a good team no doubt. But in a box environment they dont play elves often who can run rings round them. Also looking at data from that sort of environment teams generally are either CPOMB or built solely to play against CPOMB, with little regard for other races.

Lizardmen more than most teams struggle at being able to build an effective roster to take on all types of race. e.g if you build to take on cpomb you often weaken your self to elves and dwarves in a TV based match making environment, build to face dwarves or elves and you make your self weaker against CPOMB teams.

As for Zons, Spubbbbba is right on the money. Change the roster to make it interesting and not the dullest roster ever. Lack of tackle in blackbox is ridiculous which makes them seem better than they are. Show my a perpetual league where zons have been successful and I will be amazed. They just need to be made more interesting and removal of blanket dodge is a must to stop them being sooo good at low tv vs no block/tackle teams, and so terrible vs teams with lots of block and tackle.

[plasmoid]

Hi all,
I want to reply to all of you, and I will, but it's gonna take a while.

@Shteve0

but what defines Survival as a format is that it covers a lot of games (not equal to high tv).

Indeed. What defines Box and Cyanide MM is not high TV, but the high number of games allowing coaches to refine their team and build towards their "sweet spot". This is further compounded by MM preventing you from meeting overdogs, removing the incentive to grow out of your sweet spot. So, to look at Survival, you have to pay attention to the sweet spots.

My observation would be that, in order to cover that format, an idea of which teams under or over perform in that format's defining range would seem appropriate.

Agreed. To provide a bit of detail: Lacking Cyanide data, I looked at the box data. I first looked at all teams in 10(0)TV increments, making note of any team with a mean win% below 45 or above 55. Teams that had connected increments of over- or underperformance made me break out the calculator. So I hope to have located all problematic teams/ranges.

You'll agree that it's no good applying a straight short term nerf to the latter if they collapse in CRP after 20 games, for instance.

Absolutely.
Though I'm a bit tied up here by lack of league data in significant ammounts.
And then there's the wildcard of the new CRP+ metagame/environment.

Would also be keen to see some discussion around what yardstick range (%) you are aiming to bring each team's win ratio into at each format - you won't be able to measure it, of course, but I'm quite the fan of clarity of direction.

Apart from the non-mirror-match win-percentages, tier 1 is still tier 1 (45-55). I'd hope to put any nerfed team in the top half of that.
I'd hope to put tier2 teams as close to 45 as possible.
And tier3 teams as close to 40 as possible.
That's the average anyway.
And as mentioned for tournaments, I wouldn't mind of tier2 teams "accidentally" climbed a bit further into proper tier1.

Cheers
Martin

[plasmoid]

Hi Spubbbba,

If survival is long term play why are your stats only for TV under 1500?

As I explained above, those breaking points came from the data. And in Survival, teams that are strong in any range will just sweetspot - like lizardmen and amazons can.
And as stated, unlike the BBRC, I consider a team 'broken' if it is broken in any part of it's existence, rather than a lifetime average.

B is a pretty outlandish environment, part of the reason zons are so successful there is the vast amount of bash that love claw and lack tackle.

I think B works a lot like Cyanide MM. So I think it is a legitimate environment, which NTBB would try to encompas. There are certainly a lot of games played in Box and Cy-MM.

Cheers
Martin

[Shteve0]

Great, I think we're aligned.

FWIW, here's a suggestion that outlines how I would approach NTBB 2.0 in your shoes. Feel free to take what you like and ignore what you don't; it's simply a reflection of what I foresee as the most transparent and firewalled approach to a discussion of re-tiering.

1) Deal with and specify core assumptions (if necessary, redefine the tiers). They currently overlap and offer little by way of true classification. That's fine, but my approach would differ.

First, a disclaimer that a team may be in different tiers in different categories (eg of which may be <20gp, >50gp and NAF), rightly or wrongly. Here's a strawman reclassification.

T0 - Over 55% in the specified format
T1 - 50-55% in a given format
T1.5 - 45-50% in a given format
T2 - 40-45% in a given format
T3 - Under 40%

2) Grid out your findings, and state sources.

Here's another strawman.

Untitled.png

By clarifying where the problem areas appear to lie and clarifying how far you intend to nerf or buff in different formats, the process of selecting roster adjustments becomes one of identifying how closely any suggested tweaks adhere to the target map. It might not change anything to what you decide to do, but it should help to focus any feedback from playtesters or observers.

A couple of simple steps, but I think they'd be really valuable.

Best

Steve

[garion]

B is a pretty outlandish environment, part of the reason zons are so successful there is the vast amount of bash that love claw and lack tackle.

I think B works a lot like Cyanide MM. So I think it is a legitimate environment, which NTBB would try to encompas. There are certainly a lot of games played in Box and Cy-MM.

Cheers
Martin

The thing is in those environments the biggest problem is how TV works. It just isn't a fair indication of how strong teams are, hence the ridiculous lengths teams go to min max.

Look at this team for example - https://fumbbl.com/FUMBBL.php?page=team ... _id=728419

They are winning with ease at the moment because they have min maxed to the max. In a league setting this team would really struggle but in match maker environment they are slaughtering teams that are not min maxing.

Other good examples are zons, who can get a linewoman to blodge guard for 100k which is just 10k more than a blitzer with no skills. so you get teams with one blitzer with MB, PO, Tackle and then the rest of the team is linewomen. that are blodged up with a few guard dotted about. At stupidly low TV. Another example is Necromantic roster where you don't take Golems because they are so expensive. Just max out on werewolf skills so they are CPOMB players and blodged up scorers. Then give your zombies block and guard, at a cost of 90k, the same price as a rookie wight.

Of course the zons thing could be fixed by a complete overhaul of the roster, though I believe a change this big is not something you are interested in. Secondly if you want to fix the problems in a matchmaker environment you need to address all the problems caused by inducements and TV which would be a big change.

[dode74]

The thing is in those environments the biggest problem is how TV works. It just isn't a fair indication of how strong teams are, hence the ridiculous lengths teams go to min max.

Which is why TV should not be used to matchmake.

[garion]

agreed. But the point is that is how the games are matched, and making changes solely based on that environment is questionable.

[spubbbba]

B is a pretty outlandish environment, part of the reason zons are so successful there is the vast amount of bash that love claw and lack tackle.

I think B works a lot like Cyanide MM. So I think it is a legitimate environment, which NTBB would try to encompas. There are certainly a lot of games played in Box and Cy-MM.

Cheers
Martin

The thing is in those environments the biggest problem is how TV works. It just isn't a fair indication of how strong teams are, hence the ridiculous lengths teams go to min max.

That’s true and I don’t think Undead or Lizardmen are particularly used to game the system in TV based matching. So changing those teams would not really impact the environment, whilst no on would play zons if they got a low TV nerf.

Weakening those 3 teams (none of which can take claw) would reduce the variety still further in B and MM, by all accounts low TV is where you get a reasonable amount of variety. One reason oft quoted for keeping teams artificially low (being TV efficient rather than min-maxing) is that once you get above a certain point all you’ll play is the same 1 dimensional bashers over and over.

I’m also curious as to how these changes would be tested since NTBBL is a challenge league.

[Hitonagashi]

Let's be fair here Spub, people don't min-max to win in Box, they min-max to build and destroy teams/legends.

Jimmy is getting a lot of attention with his Klaw team, but he did the same with Wood Elves and Lizardmen, and nobody was crying then. Tarabaralla picked up 37 wins in a row with Lizardmen I think, by abusing low TV, and nobody cried.

Smallman and Tuamadre make ultra-death killing machines, and everyone hates them, even if they don't actually win as much as the min-maxed winners. I've played both a few times, and last time I played Smallman, he actually gave me the ball on his drive in order to try and get me to cage up and stop fouling his legend (didn't work ).

I think most of us play Bloodbowl to some extent to team build (chess/settlers are both much more satisfying games over a single game), and the ultimate expression of team building is running killer teams with decent AV at a range where other players haven't got killers that can dispose of your players yet.

[garion]

Let's be fair here Spub, people don't min-max to win in Box, they min-max to build and destroy teams/legends.

Jimmy is getting a lot of attention with his Klaw team, but he did the same with Wood Elves and Lizardmen, and nobody was crying then. Tarabaralla picked up 37 wins in a row with Lizardmen I think, by abusing low TV, and nobody cried.

To be fair I hated everyone of those teams you mentioned. I couldnt care less if they are a killer min max team or an elf one, bottom line is that is not how the game should be played and I personally think those coaches who are already good at the game should lead a better example. Its really a sorry state of affairs.

But then CRP is a massive pile of poo on the whole anyway.

[Shteve0]

Garion mate

All of this discussion about min maxing being an optimal strategy for this or that team, about such-and-such team needing a nerf or buff; it's all speculation. If you read the OP, the NTBB 2 is purportedly going to take, as a basis for its changes, a slice of the vast amounts of data that exist out there; not you waving your arms about which incarnations of particular teams you hate, or why CRP is a pile of crap.

Step 1 is defining the extent of the problem. Demanding a rewrite of amazons on the basis that you saw a team of them win some times and that you don't like how dodge and tackle interact is a pretty irrelevant contribution.

Smiley face

S