Sunday, December 11, 2011

Melta-guns, Bell Curves, and YOU!

Ok, so today I saw a dude's blog where he was trying to do some unit-vs-unit analysis. There were several things that I thought were really wrong with his analysis, but the glaring mistake was when he calculated the "expected number of pens" from meltaguns.

This guy had these huge tables of numbers, and was relying upon them whole-heartedly in his decision-making/analytical process.

But he assumed that every meltagun that hit a target penetrated it.

Every. Single. One.

His final conclusion was the Stealth Suits are the bomb, and Crisis Suits are terrible. So we know SOMETHING went wrong, amirite? ;D

All joking aside, he had a table that looked like this:

Stealth Cloud with Markerlight help against:
AV 10:  3.61 glances, 5.28 Penetrating
AV 11:  1.94 glances, 1.67 Penetrating
AV 12: 1.67 Penetrating
AV 13: 1.67 Penetrating
AV 14: 1.67 Penetrating


So... this is badly, obviously wrong. We aren't going to check his AV 10 and AV 11 numbers from the Strength 5 shots because frankly, I don't care, and he probably did those correctly.

And no, I'm not going to link to him or call him out. That's not the point.

The only reason I'm writing about this is that I realized this guy probably just didn't KNOW how to calculate the penetration likelihood for a Meltagun.

It's understandable. If no one's ever told you how to do it... how would you know?

Well, you could make a table with every possible combination of 2 die 6, and then tell Excel to inform you how many times each you would roll a "2, 3, 4, 5, 6, 7, 8..." etc.

But who the fuck does that shit? Amirite?


<><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><>

123456
1234567
2345678
3456789
45678910
567891011
6789101112

Ok, so the above is just a simple table with every possible combination of 2d6. You can imagine rolling the dice and getting a "3" and a "5," which as you can see on the table is one of FIVE possible ways to roll a total of an 8.

When you tally the number of ways there are to make each possible value (two is the lowest you can roll, twelve is the highest), you get these results:


<><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><> <><>
Total Dice RollNumber of Occurances
21
32
43
54
65
76
85
94
103
112
121

Interesting, yes? Well, with this information, you know all you need to know to calculate penetrating results from a meltagun. But let's get to that in a second. First, another picture:



No, I didn't label my axes. Yes, it's off by one (should read two through twelve instead of one through eleven).

Fuck off, it's late.

The vertical is the number of occurrences out of a possible 36, the horizontal axis is the dice roll. So you can only roll a "2" one possible way (a "one" and another "one"). So your odds of rolling a two are ONE in THIRTY-SIX. Same with a 12. You will notice as you get closer to the middle, the number of ways you can roll that number increases. The most likely number you will roll on 2d6 is a seven. It is also the average on 2d6.

So, how do you translate that into chances to pen?

Ok, well let's look at AV 14. We need to roll at least an "7" to penetrate AV 14 with a meltagun in range (base Strength 8 plus 7 equals 15, penetrating). So we need to find out how many combinations of dice rolls will get us AT LEAST a seven.

Looking at the handy chart again:


Total Dice Roll Number of Occurances
2 1
3 2
4 3
5 4
6 5
7 6
8 5
9 4
10 3
11 2
12 1

I can see that I can roll a "seven" six possible ways. I also add in the number of ways I can roll anything MORE than a seven, and I get 21. So 21 out of 36 possible combinations will result in a "penetrate" on AV 14.

Well, 21/36 is .58, SO I have a 58% chance to penetrate AV 14.

Now, I can use this in a standard formula to determine my chances to kill a Land Raider with a (single) meltagun (edit: this only includes the penetrating hit).

2/3 to hit, 58% chance to pen, 50% chance to destroy (AP 1!). That's a total of 19.44% chance to destroy a Land Raider from a single meltagun.

Whew. That's enough from me. Bed time.