Balrog's Guide to Chaos

[Zombie]

Marcus, you seem to have taken a course at Chet's school in statistics computation. The increase in chance to get an AG8 player off the pitch is actually 100%. I just calculated it because your 10% didn't feel right at all.

[Marcus]

Sorry, you're right, from 11% to 22% is an increase of 100%. My poor use of mathematical language. I assume this is what you mean?

My way of thinking is that you go from 11% chance of removing them to 22% chance of removing them but have a 16% chance of not doing a damn thing. Works out to about a 18% chance of getting them off. Feel free to pounce on that if I'm wrong. I'm a BA and was always crap at maths

I see the point there. I still would not want to give up the field position and initiative that comes with keeping the troll covered and possibly losing his tacklezones. Not with a starting team anyway.

Marcus

[Zombie]

Yes, but try to think about the other team as well. "Around the survivors, a perimeter forms." That's what usually happens with a troll around. He can clear a big part of the field with his mere presence. If not, then he can pin down 3 opposing players, giving you a numerical superiority. Think about what goes on in the opposing coach's mind when playing against 10 beastmen and one troll. I assure you that staying away from the troll will be a top priority for him, and the troll will have more impact to him than a chaos warrior would.

[Pink Horror]

Okay Marcus, I'll step in with some numbers for you.

Mighty Blow should knock twice as many players off the pitch, but really stupid should keep you from blocking at least one-sixth of the time. That lowers your net gain to two-thirds, because you lose a sixth from the extra knock-offs, and a sixth from the original knock-offs. I don't know how you got your 18%. 67% would have made more sense.



Pink Horror

[Acerak]

"100% increase" might be accurate, but "percentages of percentages" are often misleading. That's been proven in the real world time and again. It makes the following jumps look the same:

* 1% to 2%
* 4% to 8%
* 16% to 32%

I think you can understand the point, but in case you missed it: those jumps are not the same. I don't understand the "school" that teaches that you should describe them in identical terms. They're clearly not identical.

My personal preference is to say that you go from 11% to 22%; be done with it there.

-----

Ok, boring stuff is over! As for the Troll, I'm starting with one on my Chaos Pact squad. I'll take Mighty Blow and Regenerate when I'm starting without an Apothecary. It's easier to deal with the crippling injury on the Troll than it is on the CW.

-Chet

[Pink Horror]

But, Acerak, those jumps are the same.




Pink Horror

[Zombie]

Yes, they're exactly the same.

And thanks for the precision Pinky, i was gonna say that but forgot to. 67% is correct.

[Lucien Swift]

see, this is why i hate all math that can't be done on my fingers... to me, 4% to 8%, well, that's only 4% more common.... 16% to 32%, 16 times more common! that's 4 times the increase in my book, and a whole more signfigant... if you're gonna go and tell me its the same jump, i'm gonna hit ya...

[Pink Horror]

Let me try to explain why it's the same, Lucien Swift.

Percents are all about multiplication. The normal function of a percent is to be multiplied by something. So, looking at the differences between percents in terms of addition and subtraction isn't the best way to go, because when you apply percents, its multiplication and division that count.

If some event in Blood Bowl occured on 1% of every block, and then the game designers tweaked that up to 2%, you'd see the event happen twice as much, right? It's the same as if it was 4% before and 8% after, or 16% before and 32% after. I don't know what Chet means when he says this, in my opinion the clearest way of viewing percents, has "been proven in the real world time and again" to be misleading. The natural way to think of a 16% increase, to me, is that I get 16% more of it. 100% increase means 100% more. If the old chance was 16% for knockouts, and now its 32%, I'd expect twice as many knockouts. If somone had to roll a d100 every square, tripping on a 1, and you said that the chance to trip increased by only 1% when you made a 2 also fail, you're going to have a lot of angry people saying that it seems like twice as many players are falling over as before, and it doesn't make sense to say the chance went up by only 1%.

If people talking about percents stuck with Chet's preference, we'd have less problems. The before and after numbers both deserve to be mentioned whenever you're writing about a change. If you don't know exactly what you're talking about, don't throw around terms like up 10% or down 4%. You'll degenerate your topic into another thread about percents, like what happened to this one.



Pink Horror

[Lucien Swift]

no, i get that 4&8 and 16 and 32 are 100% increases, double the odds, whaddever... i just think that you can't say that something going from 4 to 8 is the same impact as something going from 16 to 32 because the range of increase, if not the multiple of it, is dramatically larger... yes, it's twice as often, but one is a lot more probable than the other...

...this is why statistics have a bad reputation

[Pink Horror]

Well, the only reason that the 16 to 32 change might have more "impact" is because you started with a bigger number, it's not because the change was any different.

The is a correct way and several incorrect ways to express a percent increase. The correct and natural way is in terms of the size of the first number. The usual incorrect way does not factor in the size of the first number. That's the definition. If you think it should be expressed in some other way, than do so. Chet's way, with the before and after, is fine. Saying 16% of your rolls will be affected is fine. But saying the chance went up 16% is not fine. Can we agree on that?



Pink Horror

[Marcus]

Yes, let's ignore the fact that Marcus is full of analgesics and is talking garbage. It's not important.

I'm still good at working out odds when it involves multiples of 52....

On the troll thing I couldn't bring myself to use something that unpredictable on a starting team. IMO you're banking on casualties and not rolling too many boneheads. That's a little too much in Nuffle's hands for my taste.

Marcus

[Norse]

You guys are all very sad..

go and play a game and put the calculators down..

PS I appreciate the benefit of playing the odds more than anyone, as I am a championship Bridge player.. the difference is you cannot legislate for dice rolls in the same way as cards. Argue all you like, it won't matter one iota when you roll a double skull.....

[Balrog]

I disagree with your point number 5. Under normal circumstances I find
it better to receive the ball, score in 7-8 turns in the first half and do the
pressure play in 2nd.

1) You have first blood by hitting the opponent's LoS on turn 1. It can
change the tide of the game should you get quick casualties.
2) You take the lead and are not forced to play your weaknesses to tie
the game (pass the ball as opposed to cage play) to tie the game.
3) Remember, the opponent can also play time control. You don't want
this to happen.

Daguun

1) I start 11 beastmen, how much damage to you think they can inflict on the first turn? Not much. Beastmen are at their best when mobile, put 3 on the LoS, the rest back, then place your beastmen in such a way as to isolate opposing players. Then beat on those while he scores.

2) You don't have to pass, in fact the only times I ever do (barring unforseen circumstances) is deep in my own end right after the kick, and I make sure it's a Quick Pass. An opportunity for 1 SPP should not be passed up lightly. And with 11 beastmen your cage isn't going to be that strong, so unless you've had the time to soften up his team before going on the offensive you might find yourself short on men.

3) It is much easier to play the clock when you get control of the ball from the moment your opponent scores until the end of the game. I frequently possess the ball for the last 13 turns of the game. If you have to count on your opponent scoring quickly in the second half you may be in for some nasty surprises.

Although a lot of this has to do with personal mindset and view of the game. My belief is that kicking first nets you the greater ball control advantage in exchange for the small loss of not hitting first. You'll get to hit first for the other 2 kickoffs, that's good enough for me.

-Balrog

[Balrog]

At least people seem to aggree that the starting team should be:
10 Beastman
1 Chaos Warrior OR Troll
3 Re-Roll
9 Fan Factor

Its annoying when you end up with some new coach 'Front Loading' his Orc Team with a Big Guy, 4 Black Orcs and 4 Blitzers cos he doesnt know any better but succeds in damaging your team
---Korhil

Well, the difference between 3 and 4 rerolls is HUGE on a team with no skills, so the decision isn't obvious. And forget about getting that 4th reroll, at 140k (with many Chaos Warriors to buy) it'll be a long time before you get it. And forget about Leader trait; why waste a double when you can get all those nice and nasty mutations?

Yeah, those front loading orc teams are nasty at first, but they lose a lot of steam quickly.

-Balrog