SFR, Inc.

cliffwiggs · Posted: Mon Mar 01, 2010 10:52 am GMT Post subject: How to use Statistics

This is a problem that is as old as the concept of an SAI. How do you properly compare dice/armies?

So my question is how do YOU do it or worded differently, what would you like to see in a tool to do it?

From my experience, I do not compare armies, I only compare dice to each other. If I make the best decision at the die level, then the army is generally the best decision as well. (i.e. an army is just the sum of the average and variances of its individual components)

There are two main ways that I compare dice: Faces that apply and average normal results.

The first is easy and works well for SAI's. There are only 5 different types of results (melee,missile,magic,maneuver,save). A face either works or doesn't.

The second is a bit harder. A die doesn't always produce normal results depending on if it gets an SAI and depending on racial abilities and other effects even normal results might differ.

The old DCM made some assumptions about SAI's giving some 'normal' results and some 'special' and ignored terrains giving you a worst case view of each die and thus each army.

For those who haven't guessed, my eventual goal is a tool that allows you to pick a group of dice and analyze that 'army'. What I can't decide is what type of analysis is proper and correct.

you can't put too much data on a chart, but you can't put too little either.

you can't make a chart for every terrain... or do I have too?

The army vs 8thface/terraintype/resulttype... thats 60 possible data points, many of which will be identical.
_________________
Multiplayer Champion Gencon11/Battlefest Champion(tied) GenCon10/World Champion GenCon07/National Champion Origins05/Intermediate Champion GenCon02/Novice Champion Origins99

chuckpint · Posted: Mon Mar 01, 2010 11:33 am GMT Post subject:

Bruce did a lot of that with DDAssist. You can compare with any of the four different terrain colors or with no terrain. It shows charts of all the possible outcomes of that die. It does not really take into account specials that don't directly cause damage too well.
_________________
You can never have too many dice.
First Place at the first ever Daemon Dice sealed starter tournament.
Battlefest tied for first GenCon 2012
Single Race Champion GenCon 2008-2009, Sealed Box Champion GenCon 2007,2009,
My collection is 21,500 Dragon Dice, 20,000 Daemon Dice, and others (too many to count).

cliffwiggs · Posted: Mon Mar 01, 2010 11:59 am GMT Post subject:

I'm trying to duplicate and improve on that type of analysis.

The way he did it would ignore most of the SAI's, which I'd prefer not to do, but am not sure how to handle that. (there may be no solution)
_________________
Multiplayer Champion Gencon11/Battlefest Champion(tied) GenCon10/World Champion GenCon07/National Champion Origins05/Intermediate Champion GenCon02/Novice Champion Origins99

cliffwiggs · Posted: Mon Mar 01, 2010 1:50 pm GMT Post subject:

FYI - for those who can't read my mind.

I'm calculating an average and variance for each die and then you just sum those up for an army.

you can make nice graphs based on overlapping these bell curves, but outside of engineers and mathematicians most people don't know how to read such a probability distribution graph.
_________________
Multiplayer Champion Gencon11/Battlefest Champion(tied) GenCon10/World Champion GenCon07/National Champion Origins05/Intermediate Champion GenCon02/Novice Champion Origins99

cliffwiggs · Posted: Thu Mar 04, 2010 7:41 am GMT Post subject:

for anyone who cares.

I am making two stats.

1) does it work during this roll type. I'm not doing individual saves, I'm not doing dragon attacks. I'm not making a distinction between saves against anything, if it works for one save I'm counting it. I'm not making a distinction between types of melee, if it works for one, I'm counting it.

2)How many results does it give. This is where I'm being controversial. If it targets health, I'm counting it as that many health reguardless of what you require that health to do. If it targets a unit and does damage, I'm counting that much damage.

I'm not doing any racial advantages. If you want to know double maneuvers, then double it. If you want to know 'count as' abilities, then add the stats.

back to the grind stone.
_________________
Multiplayer Champion Gencon11/Battlefest Champion(tied) GenCon10/World Champion GenCon07/National Champion Origins05/Intermediate Champion GenCon02/Novice Champion Origins99

cliffwiggs · Posted: Thu Mar 04, 2010 4:25 pm GMT Post subject:

BTW, I almost forgot...

I assume magic ID's are always doubled and that dragonkin always get their autosaves unless they roll a belly (thus getting double on fly and ID's)
_________________
Multiplayer Champion Gencon11/Battlefest Champion(tied) GenCon10/World Champion GenCon07/National Champion Origins05/Intermediate Champion GenCon02/Novice Champion Origins99

cliffwiggs · Posted: Fri Mar 05, 2010 8:10 am GMT Post subject:

I'm screaming in an empty room, but just FYI - white dragons are assumed to do 10 breath damage and all hybrid dragons do 10 breath damage as well.
_________________
Multiplayer Champion Gencon11/Battlefest Champion(tied) GenCon10/World Champion GenCon07/National Champion Origins05/Intermediate Champion GenCon02/Novice Champion Origins99

DEEPBLUEB2 · Posted: Fri Mar 05, 2010 9:20 am GMT Post subject:

Ok look...
I hear you...
let me say this...

I consider group monsters as units that can double.

So a common mage can double magic,
because it has a 2 icon face...
well most of them do anyways...
so I extend this to the group monsters when rating.

So a Sprite Swarm...
even though it is rated as a 3 health monster in magic...
I extend this to double magic...
on the ID.

So a evoker and a Sprite Swarm have equal double magic strength, built in the die.

Officially no...
but that's how I do it anyways.