Jim Manzi offers a peculiar prescription for Republicans in the final two weeks of the campaign at The Corner.  Declaring the presidential race all but finished, Jim wants the GOP to transfer all remaining resources to battling for the Senate:

Here’s why from the point of view of McCain/Palin supporters:

McCain has been an underdog throughout this race. According to InTrade, he’s spent 214 of the past 220 days as a less than even-money bet. His chances have collapsed within the past month.  he betting line on him as I write this is about 6-1 against. Such long-shots do come in sometimes, but it is very unlikely that McCain will be president. Unless somebody has information that is unavailable to the public markets, this estimate should guide allocation of resources.

Further, Democrats are literally 10-1 to 100-1 favorites to control each house of congress. It’s time to accept all this.

Why — just because some people who like to bet say so?  InTrade is a bookie service, and nothing more.  InTrade has a very brief history, in which it has demonstrated no predictive value whatsoever.  It’s as reliable as an on-line poll.  It measures only those who feel motivated to buy shares (place bets) on one candidate based on the individual belief of their likelihood of victory.

It’s not at all scientific, although if one has enough spare cash, I’m certain it can be at least as entertaining as betting on the Vikings.  Bookies don’t determine who wins the game, and InTrade doesn’t even measure real political support.  After the collapse of Fannie Mae, Freddie Mac, and mortgage-backed securities, I find Manzi’s reliance on the wisdom of investors somewhat amusing, and uncompelling.

Both the presidential and the Senate races are important, but if we want to talk odds, then the Senate’s a much bigger long shot than the Presidency.  McCain has continued to close in on Obama in the last few days and has two weeks to tie or surpass him.  In the Senate, though, the Republicans have to defend ten more seats than the Democrats.  They’re not going to win control of the Senate, and they have an outside shot of losing enough seats to give Harry Reid a filibuster-proof majority, although that’s a long shot as well.

Under those circumstances, it’s much more important that we fight for the McCain ticket.  If Obama gets elected, the composition of the Democratic majority in the upper chamber won’t make any difference at all.  They will approve his judicial appointments and his policies will sail through Congress for at least two years, until the Republicans have a better balance in the midterm elections.  The only way to stop that from happening is to get McCain elected, not to hold the line on a Democratic majority to 59 seats.

Update: My friend King Banaian parses the probabilities, but he makes a fundamental error:

The InTrade data give us virtually a representation of the probability of McCain winning and Obama winning.

No, it doesn’t.  It gives us the odds as seen by those who are betting on the outcome.  There’s a significant difference between those two things.  Who are the bettors?  Do they represent, as a sample, the electorate as a whole?  Are they even all American?  I reject the notion that InTrade represents the probability of either candidate winning the election.  It just gives us an idea of the opinions of those willing to place bets, just as Vegas bookmaking odds do with college football games.