LRB6 stats pulled from fumbbl's box div:

[Carnis]

The stats actually describe games which have a max of 15% TV difference, this still allows for 2000 TV team to play versus 2300 TV etc. I really have hard time understanding how having less than maximum inducements makes the stats any worse, for me it seems to make them better, if anything.

[mattgslater]

Some teams are dependent upon specific inducements, which are generally not cheap.

Some teams are held in check by either specific inducements that they tend to face over and over, or by certain other team races that don't tend to hold the same rough TV. For instance, Amazons are really cool when not facing many Dwarfs. But Dwarfs' superior skill memory will cause their TV to rise to a higher ceiling than Amazons. Similarly, in a format where one tends not to play the teams that tend to have a lot of Dodge, one is free to take less Tackle, and thereby get more skills that are useful against the mix in question. Another way of saying this is that optimum BB strategy and relative BB team power are driven by format. If you don't correct for format (and the Black Box data can't correct for format), you can only make claims that are specific to the sampled format.

Contrary to what Joe says, sample size will not correct this. It will fit into all samples, and will skew all your numbers.

[spubbbba]

Happy to have more data. But as I said with Cyanide's MM ... the rules were not meant to be used on leagues that did only TV matchings for match-ups. Its not organic and it ignores that certain inducements were added in to help weed down those higher TV teams.

That doesn't mean its not fun and a great way to play BB ... but its not a statement on balance in my opinion.

Tom

The only stats I've seen from tabletop games just showed straight win/draw/loss, they gave no indication of what races played each other or what their respective TV's were.

I'd say that stats showing how teams do against each other at roughly equal TV broken down by TV range and opposition are more useful than those we have seen. Especially as there will be a higher number of games played.

Of course the best source of data will be when the FUMBBL leagues go FBB and hopefully that is not too far off.

[dines]

It seems logical that high/wood/dark elves (avg. 60 games/team) do well in >1800 TV in Black Box as they are so few compared to orcs, dwarf, chaos and chaos dwarf teams (avg. 226 games/team). Therefore nobody has built their teams to stop elves, but instead loaded up on guard to bash the other bashers. How long they can stay above 1800 TV in there is another question...

[Greyhound]

As a note, my TV 2300 orc did really well (won all my games) all season against everyone except against the humans who came at TV1300. I couldn't manage Morg and Grif at the same time, and the wizard would have sealed the deal had the human coach not forgotten to use it.

Replaying the game over and over again there was little I could have done to prevent the draw I suffered. I think that high inducements are actually challenging and I'm now considering taking dauntless on a blitzer/lino for these times where I meet Morg and the other SP big guys.

[Joemanji]

I think that high inducements are actually challenging and I'm now considering taking dauntless on a blitzer/lino for these times where I meet Morg and the other SP big guys.

Which is great IMO. Encouraging coaches to manage their TV rather than just being a race to the top is awesome.

[mattgslater]

Or even if it is a race to the top, managing your team with relative TV in mind is still something to consider. That all-MB roster, you'll be sorry if you don't have the positioning skills to survive the Wizard, you know? Down here, we don't like to do more than incidental TV trimming, and that works fine, but those guys with the runaway TV, they have to be ready for what that brings....

[GalakStarscraper]

I really have hard time understanding how having less than maximum inducements makes the stats any worse, for me it seems to make them better, if anything.

The problem is that the rules were specifically designed so that lower TV teams (much more than the gap allowed by MM and FUMBBL black box) had access to inducements that could specifically cause permanent damage to the higher end TV team as a way of bringing them in check.

Other teams like Goblins and Halflings have as part of their design the ability for the coach to manage TV so that his normal match up gives him some brutal inducements to bring to bear.

So like it or not ... it IS how the game was designed. So when you make a league that doesn't allow it ... it is still fun ... but is an invalid source for testing data to state anything about overall game balance except for how the rules work in that different environment where the rules were not designed to be used for.

Tom

[mattgslater]

We have a Skaven coach down here who is 13-7, and except for three games against my High Elves all his losses have come as a 400k+ overdog, including one horrible butt-whupping against a rookie Nurgle squad with Grashnak and Max. A lot of it is his Block/Claw/Guard/SF Rat Ogre who tends to go down suddenly in the face of chainsaws and ST6 stars, sometimes taking the Apothecary with him.

[Smeborg]

We have a Skaven coach down here who is 13-7, and except for three games against my High Elves all his losses have come as a 400k+ overdog, including one horrible butt-whupping against a rookie Nurgle squad with Grashnak and Max. A lot of it is his Block/Claw/Guard/SF Rat Ogre who tends to go down suddenly in the face of chainsaws and ST6 stars, sometimes taking the Apothecary with him.

I once beat a well-coached Skaven team 3-0 with rookie Humans, by taking all the damage-causing inducements (Wizard, Saw, Stabber, with Bribes and Babes to keep the SWs on the pitch). Just took down GR after GR.

[Joemanji]

So like it or not ... it IS how the game was designed. So when you make a league that doesn't allow it ... it is still fun ... but is an invalid source for testing data to state anything about overall game balance except for how the rules work in that different environment where the rules were not designed to be used for.

Invalid? C'mon. The data Plasmoid collected has flaws too, just different ones. For example cherry-picking, or the fact that the data was collected from short leagues starting from scratch, wildly skewing the data to TVs of 100-130 ish. These leagues could quite easily be made up of one good coach and some disinterested mates he roped in, again leading to heavy bias to the race the good coach is using. Also there is no way to verify that games were even played.

Black Box has no way for coaches to hide from particular races or just other good coaches. For that reason I think this data will eventually be much more telling (and not just because of its size). If the purpose of a good set of stats is to test 'balance', then one (soon-to-be) massive sample is better than a small collection of tiny samples IMO. Because in the latter there is no real 'balance' of races - strength of coaching is far more important. The fact that stunties can't minimise TV to compete is a small weakness of the data IMO.

[Greyhound]

I thin we'll agre to disagree

Next time someone comes up with a black box system based on (match played/ coach level etc...) we will start to see some accurate data.

Another important skewing from TV matching system is the inherent issue with MNG not damaging your team and coaches who stick to a sweet spot.

If you start a dwarf team and early on get +ST twice and + 2 MV on the runners like I onc had you just stick there and farm teams who are struggling with still getting block.

[Carnis]

I thin we'll agre to disagree

Next time someone comes up with a black box system based on (match played/ coach level etc...) we will start to see some accurate data.

Another important skewing from TV matching system is the inherent issue with MNG not damaging your team and coaches who stick to a sweet spot.

If you start a dwarf team and early on get +ST twice and + 2 MV on the runners like I onc had you just stick there and farm teams who are struggling with still getting block.

That's really an issue of a closed league, not an issue of the B stats. In a B environment getting 160k of team value will get compensated by opponent getting 8 skills, a worthy match for 2x +ST early.

[mattgslater]

That's kind of the point. Team W/L records don't exist independently of the format, because team development is format-dependent.

There are a lot of ways to keep games competitive. Some leagues use ladders to ensure that the more successful coaches face each other disproportionately. Others use schedules to guarantee that everybody gets a similar competitive mix. Others have an honor system which may or may not get abused to one degree or another. I do think that these Black Box numbers can be represented as part of a good, inclusive sample, even if these numbers themselves may not correspond to the realities of a league with a different format. I also think it's kind of silly to say "team A does well against opponents at similar value, and team B does not, so team A must be better than team B" without then looking at how TV disparities play in, or at how development may vary from one format to the next.

I've seen a lot of Black Box team builds, and y'all take development strategies that would get no end of grief in our format. I'm not saying that these strategies are suboptimal, because the opposition can be expected to hover around the same TV level. For example, if you plan on building a killer Chaos team, you can focus on Claw because you know that when you hit peak value you will be facing a lot of Armour teams that can maintain high TV perpetually. While if you're playing the same team in a scheduled format, you might know in advance that your last four games are Skaven, Amazons, Undead and Halflings, and instead you may ignore Claw and go for positioning and/or tackling skills. Drop either team into the other format, and they'd be looking at a lot of dead TV.

[plasmoid]

I think the data is good.
Certainly no worse than what I collected.

Yes, the TV-matching is a house rule that may slant the data. On the other hand, since the inducement system is designed to give big overdogs a considerable advantage, then any set of statistics from a truly open league could be accused of not showing anything other than 'who is the overdog'.

Yes, in small leagues there are certain development dynamics. In huge leagues the connection between individual teans are almost non-existant. Face one team, and you may never face them again. I don't think the rules were designed with only small leagues in mind. So I think the data is interesting. I'd just like more.

IMO, a redeeming feature about "my" data is that it was collected from so many sources that no super-coach and no league-set-up affected the data pool in a major way.

Cheers
Martin