NTBB2.0/2014

[VoodooMike]

Regardless, we are thread-jacking again. I would like to hammer out our differences, because you are clearly intelligent (and most of the time, correct), but this thread is not the time or place for it. I'm also pretty sure that even offering to discuss will be met out of hand with a scornful insult, probably comparing talking to me to negotiating with a terrorist, but hey, I offered.

I don't think it's threadjacking - the topic is NTBB2.0 and we're talking about things that it is meant to address. Now, that said, there's never going to be a compromise on the concept of anecdote as data... at least as far as I'm concerned. There's a reason that no serious discipline in the entire world favours anecdote over data. It also doesn't matter how trivial the application is.. from games to sports to medicine.. there's NO REASON not to do things the smart, right way.

I (and I imagine Garion) don't spend hours carefully crafting the ideal retort/response.

I'm flattered you think my postings take a long time to craft. They don't - I'm just verbose, eloquent, handsome, and charming.

If you are only looking at short term, cool! You could be right about CRP+ being a long term boost which negates the short term nerf. The games I mostly care about are high (1600+) TV, so the team charts look different at that size. I'd like to see NTBB attempt to narrow the tiers at all ranges, rather than just short term, but that's a much harder proposition.

Maybe yes, maybe no. Long term outcomes are based on skill access. Looking into the utility of skills in each of the skill categories and trying to balance the categories to have some semblance of equal utility... would help a ton. The same is true of giving different skills different values in terms of TV.

The fact that will make everyone cringe is that bash teams don't win very much compared to agility teams which means that balancing the games around TV levels and win% would involve boosting bash or nerfing agility. Certainly in Box type environments people would scream about that being the opposite of what is needed... which suggests that trying to balance things to work in all environments would require some form of robustness balance as well... so that bash teams win more, but agility teams get wrecked less.

[Hitonagashi]

I won't argue with you on Bash teams not winning enough. I would hypothesis that the variance of their results due to coaching skill is more...limited than an agility team (you get the players off the pitch, you win, else, you lose/draw), but that's anecdotal again. Not sure exactly how to test that, and more importantly, what useful information could be gained by testing it.

The three main 'skill groups' that would need to balanced are General, Strength, Agility. The Agile have Agility/General, the Bash have Strength/General and the Hybrids have players with both. Your skill accesses pretty much define your "Agile/Bash/Hybrid" status as a race.

Similarly, the strength of A/S is in combination with G, not alone. Pure S access is a handicap (see Big Guys), as is pure A (see Stunties).

Logically, if you are trying to boost winning percentages of B, while increasing survivability of A, then you want to add a 'winning' skill to S, and either nerf the S attacking skills, or put a counter skill in A.

The other way of looking at it is that S access is about teamwork, whereas A access is about individual greatness. No S skills can make a player a game-winner singlehandedly. Even a single claw/mb/po/tackle/block player cannot reliably decide a game. On the flip side, a Block, Dodge, Leap, Strip Ball, Wrestle, Tackle player (say, a Wardancer...) will make a decisive difference the majority of the time.

This I believe to be part of the reason why bash is so weak at low TV. The most successful bash teams at low level are those who have an absolutely huge 'natural' advantage in terms of ST and skills compared to Elves (orcs and dwarves), and yet, even then, they are not overpowered.

As you skill up, the bash teams get better at team work, whereas the agility teams get more individualists (meaning that they have redundancy for when one or more get removed). At high TV, a bash team played extremely carefully is therefore very dangerous....but they still run into the "bad luck" problem! If the elf team can pull off one combination, their careful positioning and teamwork suddenly works against them, as the ball is now on the other side of the pitch, and they cannot just move one player to counter, as their team relies on working as a unit.

As to what the point of all this is..I don't know . Mainly just a thought-dump about what is involved in the categories at the moment. I have an idea of how to fix this, but it doesn't belong here, as it's somewhat controversial.

[VoodooMike]

I won't argue with you on Bash teams not winning enough. I would hypothesis that the variance of their results due to coaching skill is more...limited than an agility team (you get the players off the pitch, you win, else, you lose/draw), but that's anecdotal again. Not sure exactly how to test that, and more importantly, what useful information could be gained by testing it.

Regardless of what causes the variance, the teams win less overall. When using win% as the primary metric things like robustness are simply not taken into account... should they? Opinions likely vary on that. You could deal with the problem in MM/Box, for example, by making it a "resurrection" environment or something similar, and that would probably shift the focus to winning matches instead of preserving players. If that's the only environment where it is a major issue then you solve a lot of issues going that route rather than trying to change the entire balance of the game to deal with anything BUT win%.

This I believe to be part of the reason why bash is so weak at low TV. The most successful bash teams at low level are those who have an absolutely huge 'natural' advantage in terms of ST and skills compared to Elves (orcs and dwarves), and yet, even then, they are not overpowered.

Bash is not weak at low TVs: Chaos is weak at low TVs. Chaos Dwarf and Dwarf, for example, are very strong at low TVs, even as far as win% goes. Certainly the teams that have a large starting advantage in terms of stats and skills have a good record in early play... those are almost synonymous statements. If you want to change low TV success you rebalance starting skills/stats. If you want to change success across higher TV ranges, you change skill access.

As you skill up, the bash teams get better at team work, whereas the agility teams get more individualists (meaning that they have redundancy for when one or more get removed). At high TV, a bash team played extremely carefully is therefore very dangerous....but they still run into the "bad luck" problem! If the elf team can pull off one combination, their careful positioning and teamwork suddenly works against them, as the ball is now on the other side of the pitch, and they cannot just move one player to counter, as their team relies on working as a unit.

Again, this seems more like you're using "Bash" and "Chaos" interchangeably. Most bash teams actually seem to do worse as TV increases, with Chaos being the major exception.

[Hitonagashi]

Really? (not being sarcastic here!)

I'd have thought that the bash teams performance would get better at higher TV. I always found it easier to win with Orcs at 1800 than 1300. Does the data show non Chaos/Nurgle bash teams suffering as they get higher? I suspect part of this comes from my own experience with Lizardmen....they are naturally very strong against the 'bash' at low TV, and weaker against elves, which probably colours my perceptions.

I wasn't intending for bash to be synonymous with Chaos. For me, Orcs and Dwarves pick up more guard as they get higher, and since it's a stack skill, it gets much better the more you have.

I wouldn't have expected the 'stack' of blodge across more players to have more of an impact on team performance than the 'stack' of MB/Guard, but if the bash teams do spiral off as TV increases, then that would seem to be what the data shows.

It's actually a fairly common ask on FUMBBL to be able to set up a L tournament which can run resurrection events, and I think it's on Christer's todo list somewhere, along with a much larger re-vamp with far more customization options.

[plasmoid]

Very busy ATM unfortunately.
Just popping in to say that in Box, all Bashy teams except Chaos, Nurgle and CDs take a sharp performance drop at high TVs.

One might hypothesize that at high TV, these teams are facing mostly other Bashy teams, and in Bash vs Bash, Chaos, Nurgle and CDs prevail. One reason for this could be the obvious one. Claw.

Cheers
Martin

[Hitonagashi]

Very busy ATM unfortunately.
Just popping in to say that in Box, all Bashy teams except Chaos, Nurgle and CDs take a sharp performance drop at high TVs.

One might hypothesize that at high TV, these teams are facing mostly other Bashy teams, and in Bash vs Bash, Chaos, Nurgle and CDs prevail. One reason for this could be the obvious one. Claw.

Cheers
Martin

I think your reasoning might be correct. How useful a metric might it be to calculate the win percentage as the mean of each racial win percentage for that team? The idea would be that a 100 game old orc team that has played 60 of it's games against Chaos (where at high TV, it's probably the weaker opponent) can still have a more accurate picture of performance against the other 20 races in the game that don't use claw.

[VoodooMike]

I'd have thought that the bash teams performance would get better at higher TV. I always found it easier to win with Orcs at 1800 than 1300. Does the data show non Chaos/Nurgle bash teams suffering as they get higher? I suspect part of this comes from my own experience with Lizardmen....they are naturally very strong against the 'bash' at low TV, and weaker against elves, which probably colours my perceptions.

Yes, most bash teams don't ramp up as TV goes up... only Chaos, really. Chaos Dwarf stars strong and gets weaker (though never *terrible*), Orcs get decidedly weaker, Nurgle mostly just recovers its groove at higher TVs, but is closer to flat performance across the various levels.

I wasn't intending for bash to be synonymous with Chaos. For me, Orcs and Dwarves pick up more guard as they get higher, and since it's a stack skill, it gets much better the more you have.

Dwarf and Orc both get worse as TV increases, at least in Box. Maybe the skill "stack" makes the player feel like it is succeeding more during a match, but it doesn't seem to translate into more wins.

One might hypothesize that at high TV, these teams are facing mostly other Bashy teams, and in Bash vs Bash, Chaos, Nurgle and CDs prevail. One reason for this could be the obvious one. Claw.

You and your CPOMB fetish. Somehow everything relates back to it. CPOMB: Blood Bowl's God-of-the-gaps.

How useful a metric might it be to calculate the win percentage as the mean of each racial win percentage for that team? The idea would be that a 100 game old orc team that has played 60 of it's games against Chaos (where at high TV, it's probably the weaker opponent) can still have a more accurate picture of performance against the other 20 races in the game that don't use claw.

We've already looked at win% while controlling for racial composition in the environment and as I recall it doesn't shift much around.

[plasmoid]

Hi Hito and Mike,

How useful a metric might it be to calculate the win percentage as the mean of each racial win percentage for that team?

AFAIK one of the problems with this approach is how to deal with the tier 2 and 3 teams.
Do they count fully towards the average? In most/all environments they aren't as numerous as the tier 1 teams - but the suggested metric intends to eliminate that.
Some teams have a win percentage around 70+% against the three tier 3 teams, so counting them fully would certainly push up the numbers.

With NTBB2 I'm taking a pragmatist approach.
IMO there are 3 huge and legitimate environments, and rather than speculate about the hows and whys, I want to look at what is actually happening in those environments. (Unfortunately, I won't be able to get sufficient stats for TT-league play).

Mike said:

You and your CPOMB fetish. Somehow everything relates back to it. CPOMB: Blood Bowl's God-of-the-gaps.

Well, not everything
And I didn't say CPOMB. Just C.
I know I'm just eye-balling, but I do think it is interesting that looking at Box stats for TV1800+ Chaos has a better cas-count than their opponent against everyone, Nurgle has the best cas-count against everyone except chaos, (and CDs have a better cas-count than everyone except chaos and Nurgle - though admittedly some are by a very narrow margin (potentially reversing the result of you include the CI, no doubt).

So Chaos and Nurgle certainly beat up the other bashers at high TV. And they both have a high win percentage against Orcs and Dwarfs at high TV. Against Khemri too. (Again ignoring CIs).
No causation guaranteed. Surely Mike can work these numbers way better than me.

Finally, as an aside, at high TV (1800+), The top 5 most numerous opponnets that Chaos is facing is:
Chaos, Nurgle CDs, Dwarfs, Orcs (in that order) - making up 56.5% of their games (in that range).
Nurgle is also facing: Chaos, Nurgle, CDs, Dwarfs and Orcs (in that order) - making up 57.9 of theirs.

Being able to defeat your most numerous opponnets is obviously a good thing. Being able to beat them up probably helps too.
Though, obviously, how this cas differential comes about remains to be investigated.

Edit: It stands to reason that a positive cas differential is generated either by getting more knockdowns than your opponent, or having better damage modifiers.
Luck can generate both more knockdowns and more armor breaks - but we hope to eliminate luck by the ammount of data.
Coaching skill (good positioning) will generate more knockdowns and thereby more armor breaks, but again we hope to eliminate this factor by the ammount of data.
Remaining variables (as far as I can tell) are skills that cause damage modification (C-PO-MB) and skills that generate more knockdowns (Block - Guard - Frenzy; did I miss any? Stand Firm perhaps...)

Edit2: Naturally, you can also get a positive cas differential by reducing the number of cas' you're taking.

Cheers
Martin

[CyberedElf]

skills that generate more knockdowns (Block - Guard - Frenzy; did I miss any? Stand Firm perhaps...)

Tackle, Dauntless, Diving Tackle, Horns, Juggernaut, Prehinsile Tail, Multiple Block, Pro all can help generate more opponents on the ground.

Not to mention tons of Extraordinary skills.

Dirty Player also affects casualty differentials. I would include Sneaky Git. It does not generate more opponents on the ground or modify rolls, but it causes coaches to chose to make those rolls more often.

One of the goals I perceive in NTBB/NTBB2 is to encourage more diversity in the environment(s). That is the metric I would like to see evened out, not necessarily win %.
If Bash does not win more, but it is played significantly more, there is still a problem. In my experience (obviously anecdotal) the median online BB player is just a five year old at heart that wants to pull wings off insects. In my TT league, Bash is discouraged because it is a free format and people can just refuse to play against it.

[Jas1279]

Just been wandering thru the NTBB 2014 and as a primarily high elf coach it looks pretty alarming stuff.

The 'safe' long range ballistic platform high elf throwers provide is kinda a key part of the high elf offence and hitting them hard with the loss of safe throw is no small blow to the race.

Given the 6th skill at 176 is so far off the most useful/common build I aim for and see in others is the classic (pass, safe throw) +accurate +strong arm +block +dodge +fend. Recycling them for the double if you don't get it early for SA. Fend being pretty critical for long term survival given the prevalence of PO at higher levels.

If we now need to make room separately for safe throw in the build (critical for those long passes) we now need to explicitly choose between making long passes *very* risky or the survivability of my throwers at high TV losing either of which *will* hurt a lot.

If the high elves need a nerf could you consider hitting them in a less critical part of their core playstyle?

[MattDakka]

Passing game is not indispensable for any of the four Elf teams, actually passing is a risky option that good coaches avoid most of times, developing a super HE thrower is not so TV-efficient.
If you need to move the ball quickly a catcher-to-catcher hand off or a quick/short pass is all that you need, passing is not such a common action to invest a lot of TV on it.

[adhansa]

Just been wandering thru the NTBB 2014 and as a primarily high elf coach it looks pretty alarming stuff.

The 'safe' long range ballistic platform high elf throwers provide is kinda a key part of the high elf offence and hitting them hard with the loss of safe throw is no small blow to the race.

Given the 6th skill at 176 is so far off the most useful/common build I aim for and see in others is the classic (pass, safe throw) +accurate +strong arm +block +dodge +fend. Recycling them for the double if you don't get it early for SA. Fend being pretty critical for long term survival given the prevalence of PO at higher levels.

If we now need to make room separately for safe throw in the build (critical for those long passes) we now need to explicitly choose between making long passes *very* risky or the survivability of my throwers at high TV losing either of which *will* hurt a lot.

If the high elves need a nerf could you consider hitting them in a less critical part of their core playstyle?

Don't think it is ment as a nerf, sure hands are usually rated a superior skill to safe throw. Haven't got enough experience with High elf to argue, but with pro elf i choose sure hands over safe throw on my bomber every day of the week as i would rather worry about wasting team rerolls on a pickup than having to make a pass that might be intercepted.
On a tournament team i would certainly prefer the sure hands roster.

[dode74]

I think that's kind of the point. You'd choose sure hands but wouldn't choose safe throw. Giving them what is actually a very useful skill for their playstyle, but not one you would take, changes the playstyle.

My impression (and it may be only me) is that these rules are now tweaks to tweaks to tweaks and have kind of got a bit lost for the tinkering.

[Jas1279]

sure hands are usually rated a superior skill to safe throw. Haven't got enough experience with High elf to argue, but with pro elf i choose sure hands over safe throw on my bomber every day of the week as i would rather worry about wasting team rerolls on a pickup than having to make a pass that might be intercepted.
On a tournament team i would certainly prefer the sure hands roster.

Really? I'm very curious how you build your pro elf bombers then... With 4 rostered catchers the long pass is a key part of the Helf playbook and without safe throw at least 1 in 6 of your long passes that go anywhere near the opposition will fail no matter how well you mark them.

The other big down mark for sure hands is with the long bomber you never usually pick up the ball the same turn you throw it on offence (and arguably on defence too as that can just as easily be a handoff). So sure hands saves you maybe a reroll on a non critical turn (which is often a bare 2+ anyway), but you lose safe throw which is needed to prevent 1/6 of your *critical* ball moving plays going horribly wrong and giving the ball to the opposition.

Worse it by swapping a critical skill out for a junk skill it makes it no longer possible to build the standard effective h-elf thrower, so you're significantly hitting a key part of the race's play style (at higher tv) for a minimal benefit on a 'rarely important' 2+ roll.

Usual h-elf thrower being:

6-3-4-8 <safe throw> <pass> strong arm, accurate, block, dodge, +1 spare slot at level 5*

(Level 5 slot usually being fend in my book given the importance of the reliable bomber to the high elf team and prevalence of PO at higher TVs)

If you dump sure hands on him you need to take safe throw and will have zero spare skill slots before level 6

So no slot for any +stat improvements if you want to build a reliable thrower before level 6

No slot for a protective skill before level 6 on your key playmaker

No possibility of having a protective skill on a +stat thrower as you've got a junk sure hands pick clogging him up.

The fact it's seen as an alleged benefit makes me wonder if plasm plays high elves much, you're giving them a tiny benefit at low TV in exchange for completely killing their key play at higher TV.

And as AV 8 elves they do tend to be the elf team picked by those wanting to stay the distance to higher tvs in longer leagues

...makes me wonder what role high elves are intended to fill in NTBB after their long passing game has been heavily nerfed, weak dark elves with less positional options?

[Jas1279]

@Dode

....I suppose the other thing that doesn't help is safe throw still being bugged at failing to prevent non-natural 1 turnovers in the cyanide client

-DarkStalker
UKBBL