Moiraine

How is it "really interesting" ? Gerrymandering is awful. It disenfranchises voters, particularly minorities. Do you really want the Republicans to win every time, even when they lose the popular vote? Does that seem very Democratic to you? The electoral college has gotten it wrong before (2000) but that's extremely rare. Using congressional districts, it would be wrong a lot more often because of gerrymandering. Now if they made MUCH stricter laws about gerrymandering, (e.g. - a district must have 6 sides max unless it borders a body of water) it could be an option. But that would never happen.

What on earth does this mean? Let me guess. She doesn't have a penis and military guys only respect people with penises? (in case anyone gets confused, I'm not saying I think military guys only respect people with penises).

I really liked Corso growing up. I watched GameDay every Fall Saturday in the '90s. Also, I had a crush on Kirk Herbstreit for awhile.

To be fair, I don't think they should base their decisions about whether a team is good or not. It should be mostly random. The problem is it doesn't seem random. It seems... unrandom, in the opposite direction.

I did mention NV I said if Clinton loses NH then NV is an option

Haven't we heard this before? WikiLeaks has never specifically said "this date is the day". WikiLeaks has named "remember, remember, the 5th of November"... Yes it has. Assange gave a specific day in October. Then it ended up just being some kind of book promo. There were stories of people staying up all night to hear the news in the early morning. The other rumor is that it's video of Bill Clinton having sex with minors. The fact that I've now heard two completely different rumors makes this less believable but I guess we can wait with baited breath until tomorrow.

Here's mine: Given what I have on my map and recent polling: If Trump wins NH then Clinton will have to take NC or FL or NV If Clinton wins NH then Trump will have to take CO or VA or MI It doesn't matter if Trump loses Utah. If both candidates stay under 270, Trump wins the election. The only thing he needs to do is keep Clinton from 270, and she's not winning Utah regardless. To make your own map go here: http://www.270towin.com/ Click reset if you need to, click on the states to change their colors, click "Share Map," then under "Customize your map by changing one or more states; return here to share it" click the </> Image icon.

Nah they'll just gerrymander harder although that won't help with the senate.

the day they release all 20,000 pics of wieners wiener being mailed to hillary? It's very important in this instance that you get your apostrophes right.

Why doesn't our game with Iowa just count for ours?

I've obsessed over this stuff from 2004 on. 270towin was the big site back in 04 or 08. Anyway, Trump has been leading in Ohio and Iowa for a long time so it's a good thing Clinton isn't counting on those.

The % on 539 has flat-lined for about 12 hours. I don't think they have a list of the exact polls they used to calculate it though. If any of those polls contain days before Oct. 30 then the % will go down further. On the other hand, once we get polls that don't include Oct. 28-29, I expect Clinton's % to flatten out or maybe even go back up, because after those dates the news turned out to be less of a big deal. It's still potentially a big deal, but sounds a lot less ominous for Clinton now that there are more details. One thing she really needs to do is strengthen her lead in New Hampshire and Virginia and try not to lose Michigan. Those seem to be the iffiest states (that she has a lead in) right now. Nevada, North Carolina, and Florida are essentially tied, but Clinton doesn't have to have them if she holds the others. She's still in a better position than Trump, because he MUST have them, but not by much. On RealClearPolitics: http://www.realclearpolitics.com/epolls/2016/president/us/general_election_trump_vs_clinton_vs_johnson_vs_stein-5952.html If you look at the graph you can see that Clinton's % didn't go down from the 29th on (it's actually up 0.1% now), but Trump's has been up as much as 2.4% from where he was on the 28th. That seems to show that Clinton didn't lose any support, but she convinced more former likely Johnson/Stein voters to instead vote for Trump. I find it fascinating how emotional people are. I understand coming to a decision to vote for Trump after thinking the whole thing through for months, if you hate Clinton and are strongly pro life. But I don't understand sudden bumps in the polls over small things. Even with the Trump audio leak, it didn't make sense, because a lot of people already thought he was a misogynist. I'm sure the "small things" comment will be argued against, but I contend that it is small because the people who don't like Clinton already think she's lying about all the e-mail stuff. What type of person is suddenly convinced by this that they'll vote against Clinton? Why weren't they already convinced?

You'd think I would've heard that one by now. Where'd you find this treasure?

Speaking of ballots in general - does it piss anyone off how they (both parties are guilty I'm sure) try to trick you with the names of the laws/measures, and also try to trick you with the wording? For instance, and I'll exaggerate a little, they'll name a proposed law something like "We love Veterans!" and what the law really does is take money away from Veterans and uses it to fund the Yacht Club. And the description for it will sometimes have double or even triple negatives to deliberately confuse the voter. Here's an example I found immediately in a google search. If you read the amendment it sounds like a good thing for people who want to use solar energy, and I don't understand why from the article but apparently it's the opposite of good. That's what Al Gore and Jimmy Buffett (and environmentalists and solar panel manufacturers) say. https://ballotpedia.org/Florida_Solar_Energy_Subsidies_and_Personal_Solar_Use,_Amendment_1_(2016) http://www.nytimes.com/2016/10/28/science/florida-solar-power-referendum.html

Indeed

This isn't general election stuff, but oh well. This is one of the dumbest things I've ever heard in a debate and one of the best responses. This is a debate for Illinois senator.

BAM! The blob after dct = {} (that stands for dictionary, dirty people) is the part I needed help with and then the rest I figured out from that. import nltk import pickle import pandas raw_data = { 'first_name': ['Bob', 'Teacher', 'Moiraine', 'Alastair', 'Knap', 'Matilda'], 'last_name': ['Zoogs', 'CD', 'Damodred', 'Damodred', 'Plc', 'Plc'], 'gender': ['Male','Male','Female','Male','Male','Female'] } df = pandas.DataFrame(raw_data, columns = ['first_name', 'last_name', 'gender']) last_name_dict = dict(iter(df.groupby("last_name"))) def gender_features(word): return {'first_letter': word[0], 'last_letter': word[-1] } dct = {} for last in df.last_name.unique(): dct[last] = [] for first,gender in zip(last_name_dict[last].first_name,last_name_dict[last].gender): dct[last].append((gender_features(first),gender)) # TRAIN in a loop traindct = {} class_dct = {} for last in df.last_name.unique(): traindct['train_set_%s' % last] = dct[last][0:] class_dct['Classif_%s'% last] = nltk.NaiveBayesClassifier.train(traindct['train_set_%s' % last]) # CLASSIFY Test: Classif_Zoogs.classify(gender_features('Rob')) >>> Male Classif_Damodred.classify(gender_features('Elaine')) >>> Female

It's at 65.9 now. I'm guessing the rate of decrease in her lead will slow down a lot once we stop getting new polls that include Oct. 30 in them. Even yesterday some included days before the 30th, so new ones would have her lower. I guessed before the "scandal" it'd be 65-35 on the day of the election but now I'm guessing 60-40.

Moiraine: Nebraska 24 - Ohio State 20

That'd be pretty great.

I'd vote Stein or Johnson if I didn't live in Omaha.

I miss having Thursday games.

Me too. Didn't want PSU to win that game.