I've posted this on FUMBBL. I've re-posted here for anyone interested.
>>>>>>>>>>>>>>>>>Hi all,
I'm a rare guest to these forums, but I've been following this bash-stack thread pretty closely, as well as the TFF one. I've even scraped the surface of the monster-threads on Cyanide. I figured I'd give you an eye-witness account of how the stack turned out the way it did. Mind you, my memory may be a bit off. So who am I anyway? Well I'm not on the BBRC, but I was part of the vault development process from the start and stuck with it to the bitter end. My post count was also fairly high all the way through, much to the chagrin of my wife
Some of you may know me as a Galak disciple. That's not entirely accurate though. I do think he did one heck of a job with the resources at hand, but we don't agree on everything. Most notably, Galak has a lot of trust in the finished LRB6/CRP. I do too, but I'm also open to the thought that some things might be off. We were imperfect people working from imperfect data, discussing and extrapolating as best we could. But in the end, we're just people. That's not to say that I'm certain that something is wrong. Sorry. I'm kind of on the fence on that one. I hope that keeps my post nuanced.
One thing - I hoipe this long post doesn't come across as condescending or arrogant. I'm shooting for factual.
Phew.
Right:
The Vault Process
First thing worth noting is that Galak is absolutely right when he says, that we had all the numbers in advance. We had done all the math for the stack. That's not exactly the same as saying that we knew exactly how it would turn out in actual play. After all, we only had playtest feedback from the leagues that played along. On that note - and this is not me being bitter - 20/20 hindsight is easy. The BBRC and the community developing the rules were under a lot of fire for making LRB6 Bloodless Blood Bowl. I remember very well that on fumbbl the BBRC was described as sitting around with wood elf teams kissing their feet. Funny how that turned out. The other major contribution by fumbblers were Zombie and some likeminded coaches, who said that not only would they not help develop the new rules - they'd happily sabotage them. Maybe that's the bed we're all lying in now? That, and the piles of vicious personal attacks meant that of all the BBRC, only Galak would read fumbbl. So here we are.
LRB4
Anyway, when we developed the bash rules, we didn't want it to be anything like LRB4.
That's right.
I don't remeber anyone even suggesting it!
LRB4 is the least bloody of all the 3rd ed. incarnations of Blood Bowl, and the sentiment of the people involved was that it was simply a nerf too far. Too far or not, that's the environment you've all become accustomed to after playing there for quite a while longer than everyone else.
Since original 3rd ed. bash had deteriorated a lot.
*IGMEOY was added.
*DP became bonus to armor OR injury. It used to be both.
*Same thing with Mighty Blow.
*Sigurds Casualty Roll meant that injury modifiers didn't just add up to a kill (10 + Fang = Dead)
*Rerolls were removed from armor and injury rolls. Both team rerolls and Pro.
(I remember fouling with +2/+2 dirty players with a team reroll and pro. Max damage, rarely sent off. Crazy!)
*Frenzy become just 1 extra block. (No more guaranteed knockdown).
*And finally, with LRB4, Piling on went from adding ST to the armor roll - to reroll armor. No way near as potent.
At this point you might argue that all these steps were taken with good reason - that they were necessary nerfs.
LRB4-piling on is the exception to that rule.
LRB3 bashing was not widely critisized. It was bloody, but there was no cry of "broken" at that point.
If anything, the ageing rules were testament to the fact bash could not curb team growth.
The LRB4-piling on change was something else entirely.
The skill use rules had recently been rewritten, so that modifiers were only added after seeing a dice-roll.
Piling On used to be a big modifier to the armor roll - so with these new rules you would no longer go prone for no gain: You could see the armor roll, and only pile on if you knew you would break armor. And that was just absurd.
So piling on was changed to no longer be a modifier. The only other option was being a reroll!
But not all was good with the new piling on. For one, it was the mathematically weakest of all the bash skills, so it seemed rather strange that it was the only bash skill that would come with a penalty. Either way, bash had been severely weakened.
So, when re-jigging the math - we all wanted to make sure that BB would remain bloody.
We based the math on LRB3.
We wanted to curb fouling, weaken certain parts of bash, while at the same time make it harsh enough to make up for the loss of the ageing rules. And we wanted to make it contingent on doubles skills.
The main part of the tone-down was that while it was felt that hard bash was OK against other bashy teams, hard bash against AV7 was pure slaughter. I guess that's one of the reasons that new claw was invented.
The Math
Back in LRB3 (and before) Piling On added your ST the armor roll.
It was especially popular on ST4 players. Less so on ST3 players. And Big Guys with their +5 (!) could also cause a lot of carnage.
We decided to base the math on ST4 players.
Sure, not everybody has those. But that was the compromize.
In LRB3, a ST4 player with POMB who required a knockdown would have the following chance to remove (KO+) an opponent:
AV7: 29.71%
AV8: 27.01%
AV9: 23.41%
With hitting skills like block and frenzy, these stats would go up.
In CRP, the stats are reasonably similar for a POMB player:
AV7: 32.47%
AV8: 24.97%
AV9: 17.48%
You'll notice that it has become slightly rougher being AV7, but less rough being AV8 and AV9.
Now we add the special case of Mutations-only "Claw". This should be compared to the old Mutations-only "Fang".
With Fang, the LRB3 stats were (Piling On + Fang)
AV7: 36.78%
AV8: 33.34%
AV9: 28.98%
Compare this to CRP (Piling On + Mighty Blow + Claw)
AV7: 32.47%
AV8: 32.47%
AV9: 32.47%
Meaning that for AV7 and AV8 players, mutated killers are less deadly.
For AV9 players. mutated killers are a bit more deadly - but then you don't have to worry about the old DP anymore.
It this point, it should be noted that Piling On + Fang required a doubles.
That is true. And the math is kind of hard to do on that.
But consider that PO+Fang was just a 2 skill combo - So it was actually even more effective than what you see here!!
When you compare it to CRP, you should actually be comparing per skill so:
3 skills vs AV8: PO+Fang Block (45.14%) - CPOMB (32.47%)
4 skills vs AV8: PO+Fang Blockzy (55.17%) - CPOMB Block (43.84%)
At 5th skill an LRB3 would have nothing to take (unless you consider pro), but he'd still be better than his CRP counterpart, who'd be 53.58%)
Now add on top of that, that we/BBRC built in a counter to Piling On, namely Fend, and there is just no way to claim that the blocking game has gotten rougher. At least not compared to LRB3.
Compared to LRB4, sure.
How about casulaties?
Igvy has implied that I've only posted stats for KO+ as part of a cover-up, and that the difference in casualties between LRB3 and CRP are actually much larger.
Well - here are the stats for a CAS then.
First the non-mutant version (LRB4 ==> CRP):
POMB vs AV7: 14.15% ==> 17.23%
POMB vs AV8: 12.86% ==> 12.99%
POMB vs AV9: 11.15% ==> 8.92%
So while life is a bit rougher at AV7, at is a bit easier at AV9.
Then the mutated version (old Fang; new Claw):
CPOMB vs AV7: 21.22% ==> 17.23%
CPOMB vs AV8: 19.29% ==> 17.23%
CPOMB vs AV9: 16.72% ==> 17.23%
Here things have become significantly better for AV7 and AV8. And a tiny bit harsher (we're talking half a percentile here) for AV9.
In short, I maintain that LRB3 and CRP are very similar.
By all means consider the impact of no more ageing, and the big fouling nerf. People have certainly used these very stats to claim that CRP is watered down sissy-bowl.
The Evidence
In a game as complex as Blood Bowl it impossible to actually prove anything. There are simply too many variables to take into account. So with no smoking gun in sight, all we can do is discuss.
Garion and others have pointed to 2 situations, that at the very least deserve to be commented on.
1) Garions elf teams generally beat the basher teams, but have to run away all of 2nd half.
I'm unable to see how this is not evidence against CPOMB being broken.
If anything, it could be argued that finesse teams are broken, since CPOMB don't beat them.
Either way - it sounds like the only possible outcome.
You have one team which doesn't bother with protecting the ball.
And another one which doesn't bother with protecting it's players against bash-stack.
It is the logical conclusion that one gets both the win and a bloody nose.
2) RandomOracles impressive 38-1-1 record.
In the end, I'm personally open to a small nerf to the stack. And not just in fumbbl/cyanide.
I'm just not sure this is actual proof.
RandomOracle is obviously an above average coach. And a way above average chaos coach.
But those numbers are just a tad too good.
However, RO has metagamed exceptionally well.
If you check the teams record, he has played predominantly against heavy bashers - excactly what his team is built to take on. AFAIK he has played only 2 games against elfs.
Furthermore, many of his opponents were grandfathered teams - i.e. LRB4 teams that didn't know what hit them.
And finally, the format is new, and change is slow.
He can't have met many teams adapted to the new environment.
Imagine for a second that the BBRC had (re)invented the dodge skill.
Taking a blodge heavy elf team against opponents who had either not figured out (or downright refused) to take tackle would probably yield an impressive record also.
Meta-gaming
So in the end, what is the point of all this?
Galak and Ian, the only 2 BBRC memebers available for comments, have suggested a - house rule - tweak to Piling On.
There is absolutely no way that they can do anything other than that.
The BBRC have been disbanded by GW, there will be no more official annual Rules Reviews, and anything suggested will never make it onto the GW site (or the Cyanide game) for newcomers to see.
In short: By definition anything they have to say will be a house rule.
Christer has stated that he doesn't want any house rules on fumbbl.
I guess it's time to stop petitioning the BBRC and start petitioning Christer.
With no official rule and no house rule, all there is left is to meta-game.
Basically, there are two things you can do:
1) Take Fend.
No - not on everyone. But certainly on 3 players for the LOS.
If you only lose 0 or 1 player from the LOS, then you have a much better chance of surviving the ensuing fight.
On the other hand, if you lose your LOS, you'll often lose the drive.
And once you've survived the initial onslaught, then putting Fenders in the places likely to be blitzed, or next to killers prone from piling on, will further limit the damage they can cause.
2) Take teams that handle CPOMB well.
Blodge heavy elf teams are an obvious choice.
But so is any team that Fends well.
Might as well go with AV7, to reap the benefits of starting with it, instead of being reduced to 7 by Claw.
I recently did some math on this.
It's somewhat situational, so just notice the general picture rather than the specifics:
If you're getting hit by a CPOMB+Blockzy player, you'll have 53.58% chance of leaving the pitch.
Fend cuts that to 31.75%
Fend + Block/Wrestle cuts it to 17.36%
Fend, Wres, Thick Skull reduces it to 12.22%
While Fend, Wres, Dodge reduces it to 9.55% (Assuming no tackle)
Foul Appearance will also have an effect, though less than Thick Skull and Dodge.
Any team that can easily muster linemen with 2-3 of those skills should stand a pretty good chance.
That's Norse, Amazon, Dwarfs, Chaos Dwarfs, Undead and Khemri (and Bretonnians
)Then be prepared to meta-game again, when these teams start growing in numbers
The House Rule
As you may know, Ian & Galak has suggested a house rule to fix the kill-stack in fumbbl:
Simply put if you pile on, then no modifiers apply to the reroll.
I got the math wrong the first time around, which prompted me to post my own alternative rule. However, the real math looks pretty good (to me at least) so there really is no need for my own suggestion.
Never the less I've kept it in the post - so that some of your replies will still make sense.
I've done the math, and it goes like this:
POMB vs AV7: 32.47% ==> 29.26%
POMB vs AV8: 24.97% ==> 22.02%
POMB vs AV9: 17.48% ==> 14.98%
CPOMB vs AV7: 32.47% ==> 29.26%
CPOMB vs AV8: 32.47% ==> 27.38%
CPOMB vs AV9: 32.47% ==> 25.88%
The shift is something like 3%, which sounds like a reasonable small nerf. AV9 players get a considerably better deal when Claw isn't involved, but I guess that won't hurt either.
The numbers may be a tad more polarized though: Will you pile on to break AV7? Probably. Will you put your player prone for a 1-in-6 shot at breaking AV9? Probably not!
But all in all, it looks good!
So my suggestion below isn't really neccessary - even if I like what it does
My suggestion was:
Add to piling on: "In order to Pile On you must have gained proper momentum by following up."
The effects would be:
*CPOMB and POMB would no longer combo with block, it's most potent combo skill.
(And if (C)POMBers stop taking block, they'll be easier to retailate against).
*Players starting with block are the most potent (C)POMBers. They would be weakened.
*Grab would no longer counter Fend. Only Juggernaut would (and only on a blitz action).
*Big guys (and the teams relying on big guys for protection) could take Stand Firm (S) rather than Fend (G)
So, the effect isn't just in the math.
But against all AVs, the power of CPOMB with Block goes from: 43.84% ==> 38.26 (or roughly 5.5 percentiles).
Phew.
That's all.
I hope you've enjoyed the thesis
Cheers
Martin
. It's not quite going back to Gold edition though! I just thought of a mechanism that would mean AV9 teams were not as exposed as they currently are.