At the time I decided to flip, the vote count looked like this:
Originally Posted by AvatarVecna
The Outsider (4): Elenna, Apogee1, Grek, AvatarVecna
Apogee1 (3): Elenna, Valmark, The Outsider
Grek (2): Apogee1, Valmark
Valmark (2): The Outsider, AvatarVecna
Elenna (1): Grek
Look at that bolded bit from my perspective. I know that the statistically best votes to avoid LYLO are either Book Wombat/Valmark, Elenna/Outsider, Outsider/Grek or Grek/Elenna. I know that I'm Town, so that eliminates the last two options from my consideration: I want to make today's vote be either Book Wombat/Valmark or Elenna/Outsider if there's any way to make that happen. Unfortunately, as things stand, neither is a serious possibility and sticking with my initial vote means that the ultimate vote is going to be Outsider/Apogee1. A mixed strategy like Valmark/Outsider is less good than a pure strategy, but still statistically better than Outsider/Apogee1. Seriously, check the math on it:
Apogee/Outsider avoids a Thing if Elenna was the N1 convert (1/3) and Outsider was not the N2 convert (4/5) for a total of 11/15 (73%) that the Town gets at least one thing.
Conversely Outsider/Valmark avoids a Thing only if (2/3) TO wasn't picked N1 (2/3), neither was picked N2 (2/4) and Valmark wasn't picked N3 (3/4), for a total of 36/48 (75%) that the town gets a Thing.
Both of these are inferior to Outsider/Elenna (which requires (1/3) followed by (2/4) for a total of 10/12 aka 83% chance of getting a Thing), but as previously noted Outsider/Elenna wasn't something that I could make happen.
Outsider/Valmark was, so I went for it.
Unfortunately, that has apparently resulted in us ending up at Outsider/Grek. Which is just as good from the perspective of non-Grek Town as Outsider/Elenna (good!), but known to me from my perspective to include at least one wasted vote (uh oh). So while I like that everyone is going for one of the logical voting options now... please reconsider?
- - - Updated - - -
Wombat/Valmark isn't guaranteed to catch a Thing. Rather, the logic behind that vote pairing is that you two the only ones who seem like plausible N2 AND N3 conversion targets, making for a mirror of the N1/N2 situation with Elenna/Outsider/me. The raw stats are better going by AV's numbers, but again as you've noted those are AV's numbers and are premised on the Things definitely not double-converting untested people. If we allow for a small chance that they did double convert untested people, either through scum error or as a weird WIFOM ploy, you're basically back down to even odds with Elenna/The Outsider, Elenna/Grek or Grek/The Outsider as a vote. Hence why those four were my possible considerations in the post above.
Originally Posted by Valmark
This is overall a very weird post. While the actual vote is a mirror of my own for their position, it doesn't look like TO arrived at it by the same logic that I did. Instead justification is that I've been more active than usual and do good analysis. That doesn't seem like a good reason to me. The rest of the post is arguing that we shouldn't vote for them because everyone things TO is scum in this game, even when they're not. But that's exactly why the Things would have converted him. He's been previously tested and has a built in excuse for looking suspicious and for making weird calls. I am overall pretty confident in Outsider as a vote, which is why I'm not in Full Panic Mode over the current vote going after me - we'll probably get a Thing in TO and then jump onto the obvious other Thing from there. (Not posting just yet who I think the obvious other Thing is, you'll see once I'm cleared.)
Originally Posted by The Outsider
Side note: Everyone keeps noting that I seem like I post more than I do. Which, yeah, fair cop. I'm a fairly terse writer and struggle to spell out my thoughts explicitly. But my analysis has been consistently useful, as people have noted, and that makes it so that the stuff I do post has an outsized impact on the conversation. I don't feel like that should be a scum tell so much as a 'Grek is just like that' tell.