Jay Cost shares his predictions for the election...a GOP gain of 61 seats in the House and 8 seats in the Senate. My gut feeling two months ago was 60 House and 8 Senate seats, so I feel pretty good about that one now that Cost has weighed in with almost identical numbers.

Sean Trende has this excellent piece which asks the most important question concerning the election; who will come out to vote? The most difficult task facing pollsters as they try to accurately gauge voter sentiments and intentions is the attempt to ascertain whether or not the people they poll will really turn out and vote. Trende believes most pollsters are accurately determining how people will vote based on their self-identified party status. For this election, in just about any poll you look at, 90-95% of Republicans and Democrats say they will vote for the candidate of their party, while independents are breaking for GOP candidates about 55-45% or so. Trende believes pollsters are, deliberately or not, giving Democrats a 3-4% advantage because they are predicting a voter pool that looks more like 2006 and 2008, which were good years for Democratic turnout. He believes 2004 is a better model for this election. If that is the case, look at each race (my opinion) and add 3 points to the Republican's score. So, let's experiment with this technique by looking at the most hotly contested Senate races (using RCP averages)...

CT - Blumenthal (D) 52% McMahon (R) 45%
CA - Boxer (D) 45% Fiorina (R) 43%
CO - Buck (R) 46% Bennett (D) 45%
IL - Kirk (R) 41% Giannoulias (D) 39%
KY - Paul (R) 46% Conway (D) 42%
NV - Reid (D) 45% Angle (R) 45%
PA - Toomey (R) 46% Sestak (D) 44%
WA - Murray (D) 49% Rossi (R) 47%
WV - Manchin (D) 46% Raese (R) 44%
WI - Johnson (R) 51% Feingold (D) 45%

I'm not including Missouri and New Hampshire, where the GOP candidate in each contest has an average lead of 10 points. I think national Democratic strategists are writing both of them off, as they are both for open Republican seats.

If you add 3% to the GOP candidate's total, then the Republicans are leading in nine out of the ten races, rather than five out of the ten. Only the real poll on election day will tell, of course, but I will be looking for this trend early that night.

The Pew Research Center has a new poll showing a more intense ground campaign than 2006, and more intense Republican enthusiasm, among other things.

Juan Williams writes about why he was fired from NPR.

The Wall Street Journal blasts NPR for their actions.

The Washington Post also criticizes the decision.

The National Review calls for them to lose their taxpayer funding.

Paul Krugman pans the British government for their spending cuts, while Sally McNamara praises them.