Here are my predictions for some of the closer midterm Senate races.  In a previous article, I explained why Republicans could see an election wave as high as 12 points in parts of the country.  Even so, I am trying to rein in my enthusiasm, read the polls, and give realistic predictions.

You can see the Pew Research state party affiliation numbers here.  These numbers include leaners and are just for comparison with the party splits in individual polls.

For reference, most 2020 Biden polling error numbers are taken from this New York Times article or the RealClearPolitics 2020 polling averages.  Here, Biden +4 means that polls overestimated Biden's vote by 4 points.  Biden -4 means the opposite.

Just for the record, I don't like most university polls because they tend to lean left.  So does the N.Y. Times.  I dislike newspaper-sponsored polls because they are often hot, fast, cheap, and inaccurate.

Arizona: Kelly-D vs. Masters-R

Arizona party affiliation is 39-D, 40-R, 21-I.  The 2020 polling error was Biden +2%. 

The big news is that Libertarian Marc Victor has dropped out and endorsed Masters.

Obviously, this race is very close.  The real question is, where do the undecided voters go, along with 2–3% backing Marc Victor who have not yet cast a ballot?  That could put Masters into the lead within the margin of error.

Prediction: Masters by 1.

Georgia: Warnock-D vs. Walker-R

Georgia party affiliation is 41-D, 41-R, 18-I.  The 2020 polling error was Biden +2%. 

Rasmussen and Trafalgar both have the Libertarian candidate in the 4% range.  However, I doubt that that will hold up.  The N.Y. Times gives the Libertarian 1% with 4% undecided.  The undecided voters and undercounted conservative support of 1–2% may also factor into this race.

Prediction: Walker by a field goal, 3 points.

New Hampshire: Hassan-D vs. Bolduc-R

Party affiliation in the Shire is 44-D, 35-R, 20-I.  The 2020 polling error was Biden +3%. 

According to the St. Anselm poll, the right track/wrong track split is 18%/71%, with 12% unsure.  Hassan's favorability is also underwater by -8%.  These are not good signs for Hassan.

I have two key observations.  First, the Emerson poll has roughly double the number of undecided voters you would expect.  No idea what is up with that.  Two, every poll has 3% undecided, similar to the polling error in 2020.  So does Kauffman pull enough of the undecided vote to swing the election to Hassan, or does the undecided vote go to Bolduc?  My guess is Bolduc gets them and maybe some of Kauffman's voters, too.

Prediction: The General marches to victory by 2.

Nevada: Masto-D vs. Laxalt-R

Nevada party affiliation, including leaners, is 46-D, 37-R, 18-I.

According to the Susquehanna poll, President Biden's job approval among independents is minus 31 points.  That is not a good sign for Masto.

The problem with the Nevada Independent poll is the 10% undecided.  What you see here is social desirability bias, which reflects a person's fear of being judged by others.  I doubt that all of these folks are undecided.  They simply do not want to tell a newspaper they are voting for Laxalt.

Prediction: Laxalt by 5.

Pennsylvania: Fetterman-D vs. Oz-R

The 2020 polling error was Biden +4%.  Pennsylvania party affiliation is 46-D, 39-R, 15-I.

In the Susquehanna poll, 57% of voters believe inflation and the economy are the most critical issues.  And Biden's job approval is minus 15 points.

In the Emerson poll, 51% believe that the economy is the top issue.  Joe Biden's approval is minus 17 points.  Also, 50% of voters say the Fetterman/Oz debate lowered their opinion of Fetterman.

Every poll listed above has 4% undecided.  In a close election, this may be the deciding factor.  I think at least half will go to Oz

Prediction: The Wizard by 4.

Ohio: Ryan-D vs. Vance-R

The 2020 polling error was Biden +7%, meaning the polls were off by 7% in Ohio.  Ohio party affiliation is 40-D, 42-R, 18-I.

According to the Progress poll, Joe Biden's favorability is minus 19 points.

Cygnal has the right track/wrong track split at 23%/72%.  In this poll, Biden's favorability is underwater by 22 points.  These are not good signs for Ryan.

It is important to note that Ohio was ground zero for the hidden conservative vote in 2020.  So chances are that most of the undecided voters will break for Vance.

Prediction: Vance wins by 10.

Washington: Murray-D vs. Smiley-R

The Washington state party affiliation split is 44-D, 33-R, 23-I.

According to the Emerson poll, Joe Biden's favorability is only -1%.  Murray's favorability is +5%.  Smiley's favorability is +3, with 19% either undecided or never heard of her.  Only 28% see the economy as the top campaign issue.  Crime is fourth at 11%.  Unfortunately, the Emerson poll was taken at the end of September and may be missing the Smiley surge.

The Moore Group poll is from the end of October and is not listed on the RCP website.  They are a private polling firm, but their results should be as valid as anyone's.  They have Murray's favorability at minus 3 and Smiley's favorability at +3.  Here is the link. 

There are enough undecided voters in each of these polls to either put the outcome within the margin of error or give Smiley the victory.  However, Washington is an awfully blue state.

Prediction: Smiley wins by a fistful of votes.

Image via Pexels.