Analysis by Parts
First, I feel that I should point out that one of part of my analysis (Part 1 actually) was my gut instinct. However much you do or do not trust me, I doubt that anyone but me should be relying on my gut. So, I won't be offended if you pretty much just ignore that part when making your own conclusions.
Now, on with my analysis.
Part 4: Individual Voting:
Obviously, an innocent person can vote in a way that looks suspicious. It's not plausible that only the wolves would vote 'suspiciously' because the innocents don't really know for sure what they are doing( me included

). That said, certain voting will still
look suspicious, whether the person is innocent or not. So, this is how suspicious a vote looks, and I'll have to do it for each person seperately, so it will be long. I will be putting my own opinions in
italics, so they can be easily seen as my own thoughts.
Caranlondien - Day 1: Votes for
Anguirel.
Gives decent reasoning. (
0) Day 2: Votes for
Lhuna. Still wants double lynch. (
0) Day 3: Votes for
Lhuna. Widens gap between
Kath and
Lhuna.
In light of the planned double-lynch, I really don't like this vote. (
-1) Day 4: No vote.
tar-ancalime - Day 1: Votes for
Farael.
Hit the lover, but that doesn't really tell us much for Day 1. (
0) Day 2: Votes for
Eomer. Quotes reasons.
Despite voting for an innocent, her extensive reasoning makes it less suspicious. (
+1) Day 3: Votes for
Kath. Narrows gap between
Lhuna and
Kath.
Too easy for a wolf to vote for Lhuna, it wouldn't make sense for one to vote for Kath. (
+1) Day 4: Votes for
Glirdan.
Can't say yea or nea about it. I probably would have done the same. (
0)
Cailín - Day 1: Votes "randomly" for
Lalaith.
Can't know anything for sure about this one. (
0) Day 2: Votes for
Eonwe. Only current suspect with a vote already.
Doesn't seem suspicious. (
0) Day 3: No vote. This allows double lynch. (
+1) Day 4: No vote.
Naria - Day 1: Votes for
TGWBS.
Her reasoning was strange, but typical Day 1 vote. (
0) Day 2: Votes for
Eonwe. "...after I jump on this bandwagon!!" (
-1) Day 3: Votes for
Kath. Ties
Kath and
Lhuna with 3 voters left.
Very unlikely for a wolf. (
+1) Day 4: No vote.
Valier - Day 1: Votes for
Glirdan.
Reasoning was occupation mostly, but Day 1 is random. (
0) Day 2: Votes for
Eonwe. Votes for leader for being 'most annoying'. (
-1) Day 3: Votes for
Kath. Narrows gap that ended in double-lynch.
Wouldn't make sense for a wolf to vote this way, one Lhuna vote pretty much stops the double and doesn't look suspicious. (
+1) Day 4: No vote.
Lalaith - Day 1: Votes for
TGWBS. "Following his advice."
Seems mildly suspicious, but not enough for a negative. (
0) Day 2: Votes for
Eonwe. Because of
Eonwe's random vote on Day 2.
Could be a safe vote for a wolf, but just as easily could be enough reason for an innocent. (
0) Day 3: Votes for
Lhuna.
This widened the gap between Kath and Lhuna (which was bad), so could be a wolf trying to save Kath. (
-1) Day 4: Votes for
Glirdan. Ties
Glirdan and
Guy for doulbe-lynch.
Can't tell anything from this, there were still many voters left. (
0)
Thinlómien - Day 1: Votes for
Glirdan. Seems most guilty to her. (
0) Day 2: Votes for
Naria. Votes for an unknown in a six way
2 unknown-2 wolf-2 innocent tie.
Drawn between Lhuna and Naria, saw Naria as more suspect; I can't blame her. (
0) Day 3: Votes for
Lhuna. First vote for
Lhuna.
Could be setting up lover-wolf to look innocent. (
0) Day 4: Votes for
TGWBS.
Can't really say anything, as I also was mildly suspicious of him. (
0)
Formendacil - Day 1: Votes for
TGWBS. Ties
Garin and
Guy.
He put up a tie between two innocents, but with all the support for double lynchings... (
0) Day 2: Votes for
Kath.
Ensured Eonwe's death. It seems far to much like a wolf voting for a fellow when someone else already has the noose around their neck. (
-1) Day 3: Votes for
Kath. Ties
Lhuna-wolf with
Kath-wolf.
Extremely unlikely that a wolf would bring his true fellow up beside a Lover. (
+1) Day 4: Votes for
Naria.
Votes for an unknown rather than bandwagoning, but could be wolf not breaking the tie. (
0)
littlemanpoet - Day 1: Votes for
Eomer.
Seems retaliatory; although I, on Day 1, was also sure there was something going on between these two. Eomer was acting slightly odd, so no negative. (
0) Day 2: Votes for
Lhuna.
Gives reasons through several posts. Extensive reasons and quotes seem unlikely, as it would be wolf building a case against wolf. (
+1) Day 3: Votes for
Lhuna.
After this, a double lynch would have been much more difficult. One more Lhuna vote might have doomed it. (
-1) Day 4: Votes for
TGWBS.
Don't know anything. (
0)
Gurthang - Day 1: Votes for
LMP.
I felt that there was something between Eomer and LMP and sided with Eomer. (
0) Day 2: No Vote. Day 3: Votes for
Lhuna. Creats tie and double-lynch. (
+1) Day 4: No Vote.
SamwiseGamgee - Day 1: Votes for
Garin. Breaks
Garin-Guy tie.
I doubt a wolf would be worried enough to break a tie between two innocents. (
+1) Day 2: Votes for
Lhuna. Brings
Lhuna into a tie for the lead.
Unlikely for a wolf to bring up a fellow, even if he suspected her of Loverness. (
+1) Day 3: Votes for
Lhuna. Widens gap between wolf and Lover-wolf.
In light of the planned double-lynch, I really don't like this. (
-1) Day 4: Votes for
Glirdan.
Can't tell anything. (
0)
Extra
+'s and
-'s:
+1 for being first to vote for a wolf. A wolf would be less likely to push another wolf forward for lynch:
Thinlómien
Formendacil
+1 for voting for a wolf twice:
Caranlondien
Formendacil
LMP
Samwise
-1 for voting for an innocent thrice:
Lalaith
So, totals(top + extra):
Caranlondien -> -1 + +1 = 0
tar-ancalime -> +2 + 0 = +2
Cailín -> +1 + 0 = +1
Naria -> 0 + 0 = 0
Valier -> 0 + 0 = 0
Lalaith -> -1 + -1 = -2
Thinlómien -> 0 + +1 = +1
Formendacil -> 0 + +2 = +2
littlemanpoet -> 0 + +1 = +1
Gurthang -> +1 + 0 = +1
SamwiseGamgee -> +1 + +1 = +2
So, lastly, the inevitable list for this little voting test (by that I mean Part 4):
Looks innocent:
tar-ancalime
Formendacil
SamwiseGamgee
Looks suspicious:
Lalaith
Caranlondien
Naria
Valier
I'll return later and finally decide what I conclude from all that.