This post has been updated with more pundits' opinions.

I've been asking state and local political pundits for their predictions for tonight's elections.

So far, five state and local writers---Chris Grygiel and Joel Connelly of the PI.com, David Brewster and Knute Berger from Crosscut, David Goldstein from Horsesass, and Erik Smith from the Washington State Wire---have responded. So far, everyone agrees with me: Patty Murray will win tonight.

• Grygiel didn't want to get into details (superstitious?) but said, "I believe Murray will win a very tight race. I'll leave the point spread to others."

• Connelly, in contrast, was (much) more specific, predicting Murray would win 50.08 to 49.92, but that we won't know the outcome until Thanksgiving; that the Democrats will continue to control the US Senate 52-48, that the Republicans will win the US House by a margin of 226-209; that Democratic incumbent Rick Larsen will make it through against challenger Rick Larsen in the 2nd Congressional District, "barely"; that Republican Jaime Herrera will beat Democrat Denny Heck in the 3rd and "get to decorate House Republican press conferences" (?); that Initiative 1098, the high-earners' income tax, will lose; and that I-1082, the workers' comp privatization initiative, will lose.

• Brewster, too, thinks Murray will win, by a two-point spread; that the Democrats will retain control of the state House and Senate in Olympia; that the US House will go to Republicans; that both Herrera and Larsen will win; that 1098 will lose by a whopping 10 points; and that there will be two "surprise winners": Referendum 52, retaining the bottled-water tax to help fund energy retrofits at schools; and Initiative 1100, the Costco-backed liquor-privatization measure.

• The Washington State Wire's  Smith Murray says will win by 3 points. "I'm not sure the Rossi campaign caught fire with voters; it all seemed to be run from D.C. But a big wind from the east might prove me wrong."

On the initiatives, Smith predicts that 1053 (the Tim Eyman initiative that would require a two-thirds majority of the legislature to raise taxes) will pass; that 1098 will fail; that 1100 will pass; that 1105, the other liquor-privatization initiative, will lose; that 1107, which would repeal the candy and soda taxes, will pass; and that 1082 will fail. His explanation is a little over my head, so I'm printing it in full:

On 1082, what really happened isn't so much a matter of voter confusion. Yeah, maybe there's some of that. The issues are pretty complicated. But I think the real story is that Labor and Industries played things pretty shrewdly and denied its opponents a key campaign argument. What happened is terrible from an open-government perspective, but it was pretty clever.

L&I generally makes an announcement in August about the "indicated rate" for next year's workers' comp insurance. That's an actuarial estimate of the rate hike that will be required in order to keep the insurance fund solvent. And all indications are that this figure would have been absolutely enormous -- perhaps about 20 percent or so, maybe even more. Had it made the announcement, I suspect the campaign would have caught fire -- employers would have gone absolutely bananas.

Normally L&I follows this announcement in September with a "proposed rate," which can be a bit lower. L&I has a little more leeway than an ordinary insurance company, because the state insurance funds are backed by the credit of the taxpayers. That would have leavened things a bit, but it would have still been a dark headline. More anger would have ensued.

The effect of these two announcements would have been huge. You would have seen yes-on-1082 messages on every business readerboard in the state.

Well, what happened is that L&I decided to postpone the announcement of these two figures until after the election. They're going to be released Nov. 10. Nothing required the department to announce in August and September. The official explanation was that this year there were technical difficulties in collating data, etc. And it said that because 1082 would have required a change in the way taxes are calculated, it didn't make sense for the department to release numbers that might change, because it might "confuse" employers. So we still don't have a good idea where the insurance funds stood on June 30, 2010, months after the end of the fiscal year, because a full report has not been released. We do know that a couple of indicators are looking worse -- time loss figures per claim are up by two weeks over the last year, hinting at higher costs, and the accident fund ate $360 million into reserves over the last year. We also know that some sort of an indicated rate was calculated in August, because of a public records act request done by a couple of Republican legislators. We've seen the document, but the actual figure was blacked out, on the grounds that the number was in draft form.

So even though this is a front and center issue for business associations in Olympia, there was no reason for your mom-and-pop small businesspeople to explode.

• Goldy at Horsesass says his "gut feeling" is that Murray will win by five points or more; that Herrera will win in the 3rd; that no incumbent US Reps will lose in Washington State; that the Democrats will lose the US House but keep control of the Senate; and that every initiative except 1107 and 1053 will lose, "though 1100 will be awfully close."

• Finally (for now, anyway),  Crosscut contributor Berger's predicts that Murray will beat Rossi by five points; that Koster will beat Larsen in the 2nd; that Herrera will beat Heck in the 3rd; that Herrera will beat Heck in the 3rd; that the Democrats will lose control of both the US House and the Senate; both 1082 and 1098 will fail; and that 1100 will pass.