Sabato's First 2016 House Ratings

[SWE]

http://www.centerforpolitics.org/crystalball/2016-house/

Nothing for Senate or Governor yet.

[Frodo]

Why isn't Gwen Graham in the toss-up category? 

[Maxwell]

I'd say McSally is a strong fit for the district should she win, she'll be fine. He could move Blum, Hardy, and maybe Katko to Leans D in exchange moving Brad Ashford to Leans R. This seems like a fair enough ranking though.

[ElectionsGuy]

Surprised he would do this so early.

Why isn't Gwen Graham in the toss-up category? 

Supposedly she's the right Democrat. Though of course 2014 can prove no matter what Democrat you are, if you're in a red district you're in trouble. Northern Florida in particular though has been quite reluctant on always choosing Republicans.

[Vega]

Glad to see PA-08 as a toss-up, though I can't help but feel it will be won by a Republican.

[Devils30]

Graham will benefit from better turnout in Tallahassee and from African-Americans. Also a solid candidate to build inroads with the district's white voters. I would have put ME-2 and FL-26 in tossup.

[Negusa Nagast 🚀]

Surprised CA-25 is in Safe R. The PVI is only R+4.

[publicunofficial]

Calling CA-25 non-competitive aside, pretty good list.

[JRP1994]

RIP John Barrah, his granddaddy's Smith and Wesson, and his perennial "Leans Dem" district.

[Vosem]

http://www.centerforpolitics.org/crystalball/2016-house/

Nothing for Senate or Governor yet.

First couple of quibbles: Ashford, Blum, and Hardy probably all start out as at least slight underdogs. By contrast, McSally should be fine with incumbent advantage in a Romney district, Dold barely lost in 2012 with an Illinoisan Democrat atop the ticket and probably starts off Leans R, and PA-8 isn't Tossup category either. By contrast, Gwen Graham, Will Hurd, and Scott Peters should probably all start out as Tossups.

But it's a reasonable place to start, all in all

[Lief 🗽]

RIP John Barrah, his granddaddy's Smith and Wesson, and his perennial "Leans Dem" district.

I wonder if he'll run again. I assume he could be competitive in a less anti-Democratic year with higher turnout.

[Devils30]

AZ-02 is a 50-48 Romney district in an area with demographic trends that suggest it will be blue fairly soon. Wouldn't shock me if it moves left in 2016 even if the country moves right.

[Attorney General, LGC Speaker, and Former PPT Dwarven Dragon]

Calling GA-12, CA-25, and MN-02 (Obama district in 2012, rep. is fairly popular, but possibly vulnerable in a presidential year (2016) with the right candidate) completely uncompetitive aside, this looks pretty good at such an early stage.

[Nichlemn]

Nothing strikes me as conspicuously out of place, but I just think that rating only 56 districts as something other than "Safe" two years out before an election is too few. If we are to take "Safe" as representing a probability below (say) 5% of the seat switching,  I think pretty much any remotely swingy district should qualify, because there's so much time for surprise retirements/scandals/huge waves to emerge. (IIRC, quite a few of the GOP's 2010 House gains were rated "Safe D" early in the cycle). Most of the time, most of these districts would get moved to "Safe" by early 2016, in the likely event that extraordinary occurences would not materalise. CT-05 and OH-06 are examples of such seats.

Unfortunately, pretty no-one bothers to "score" the early ratings. You could just pull ratings out of your ass and no-one will care so long as your final ratings do okay.

[SPC]

Nothing strikes me as conspicuously out of place, but I just think that rating only 56 districts as something other than "Safe" two years out before an election is too few. If we are to take "Safe" as representing a probability below (say) 5% of the seat switching,  I think pretty much any remotely swingy district should qualify, because there's so much time for surprise retirements/scandals/huge waves to emerge. (IIRC, quite a few of the GOP's 2010 House gains were rated "Safe D" early in the cycle). Most of the time, most of these districts would get moved to "Safe" by early 2016, in the likely event that extraordinary occurences would not materalise. CT-05 and OH-06 are examples of such seats.

Unfortunately, pretty no-one bothers to "score" the early ratings. You could just pull ratings out of your ass and no-one will care so long as your final ratings do okay.

http://www.centerforpolitics.org/crystalball/articles/2014-house-ratings-democratic-potential-republican-predictability/

Republicans won 4 of 7 tossups and all but one each in the Lean/Likely categories; Democrats lost 9 of 13 in the Lean category and 4 of 14 in the Likely category, plus ME-2.

[Nichlemn]

Nothing strikes me as conspicuously out of place, but I just think that rating only 56 districts as something other than "Safe" two years out before an election is too few. If we are to take "Safe" as representing a probability below (say) 5% of the seat switching,  I think pretty much any remotely swingy district should qualify, because there's so much time for surprise retirements/scandals/huge waves to emerge. (IIRC, quite a few of the GOP's 2010 House gains were rated "Safe D" early in the cycle). Most of the time, most of these districts would get moved to "Safe" by early 2016, in the likely event that extraordinary occurences would not materalise. CT-05 and OH-06 are examples of such seats.

Unfortunately, pretty no-one bothers to "score" the early ratings. You could just pull ratings out of your ass and no-one will care so long as your final ratings do okay.

http://www.centerforpolitics.org/crystalball/articles/2014-house-ratings-democratic-potential-republican-predictability/

Republicans won 4 of 7 tossups and all but one each in the Lean/Likely categories; Democrats lost 9 of 13 in the Lean category and 4 of 14 in the Likely category, plus ME-2.

ME-02 is a example of a seat that should definitely been "Likely D" early on, as Michaud retiring and/or running for another office was not out of the question, and at D+3 it was competitive on paper.

[SPC]

Nothing strikes me as conspicuously out of place, but I just think that rating only 56 districts as something other than "Safe" two years out before an election is too few. If we are to take "Safe" as representing a probability below (say) 5% of the seat switching,  I think pretty much any remotely swingy district should qualify, because there's so much time for surprise retirements/scandals/huge waves to emerge. (IIRC, quite a few of the GOP's 2010 House gains were rated "Safe D" early in the cycle). Most of the time, most of these districts would get moved to "Safe" by early 2016, in the likely event that extraordinary occurences would not materalise. CT-05 and OH-06 are examples of such seats.

Unfortunately, pretty no-one bothers to "score" the early ratings. You could just pull ratings out of your ass and no-one will care so long as your final ratings do okay.

http://www.centerforpolitics.org/crystalball/articles/2014-house-ratings-democratic-potential-republican-predictability/

Republicans won 4 of 7 tossups and all but one each in the Lean/Likely categories; Democrats lost 9 of 13 in the Lean category and 4 of 14 in the Likely category, plus ME-2.

ME-02 is a example of a seat that should definitely been "Likely D" early on, as Michaud retiring and/or running for another office was not out of the question, and at D+3 it was competitive on paper.

Should that be an indictment of his early ratings that seats switched in larger proportions than their ratings would indicate or praiseworth that he only failed to label one seat that would flip as competitive?

[publicunofficial]

http://www.centerforpolitics.org/crystalball/2016-house/

Nothing for Senate or Governor yet.

First couple of quibbles: Ashford, Blum, and Hardy probably all start out as at least slight underdogs. By contrast, McSally should be fine with incumbent advantage in a Romney district, Dold barely lost in 2012 with an Illinoisan Democrat atop the ticket and probably starts off Leans R, and PA-8 isn't Tossup category either. By contrast, Gwen Graham, Will Hurd, and Scott Peters should probably all start out as Tossups.

But it's a reasonable place to start, all in all

Rothenberg is very pro-incumbent in his early rankings. Rarely does he ever brand a member of the House as the underdog right out of the gate, even obvious driftwood like Chip Cravaack.

[Nichlemn]

Nothing strikes me as conspicuously out of place, but I just think that rating only 56 districts as something other than "Safe" two years out before an election is too few. If we are to take "Safe" as representing a probability below (say) 5% of the seat switching,  I think pretty much any remotely swingy district should qualify, because there's so much time for surprise retirements/scandals/huge waves to emerge. (IIRC, quite a few of the GOP's 2010 House gains were rated "Safe D" early in the cycle). Most of the time, most of these districts would get moved to "Safe" by early 2016, in the likely event that extraordinary occurences would not materalise. CT-05 and OH-06 are examples of such seats.

Unfortunately, pretty no-one bothers to "score" the early ratings. You could just pull ratings out of your ass and no-one will care so long as your final ratings do okay.

http://www.centerforpolitics.org/crystalball/articles/2014-house-ratings-democratic-potential-republican-predictability/

Republicans won 4 of 7 tossups and all but one each in the Lean/Likely categories; Democrats lost 9 of 13 in the Lean category and 4 of 14 in the Likely category, plus ME-2.

ME-02 is a example of a seat that should definitely been "Likely D" early on, as Michaud retiring and/or running for another office was not out of the question, and at D+3 it was competitive on paper.

Should that be an indictment of his early ratings that seats switched in larger proportions than their ratings would indicate or praiseworth that he only failed to label one seat that would flip as competitive?

Not really either. There's such a wide variance in possible outcomes that "calling" the races from early on is close to impossible. It's that knowing what we knew in early 2013, ME-02 shouldn't have been Safe D, and that remains true even if it hadn't ended up flipping.

2010 was the real failure of these models. I can't find Sabato's early predictions, but Cook only had 50 Democratic House seats as less than "Safe" early on. Of course, Republicans netted 63 seats while not winning all of those seats, so ~20 "Safe D" seats ended up voting Republican.

[publicunofficial]

Nothing strikes me as conspicuously out of place, but I just think that rating only 56 districts as something other than "Safe" two years out before an election is too few. If we are to take "Safe" as representing a probability below (say) 5% of the seat switching,  I think pretty much any remotely swingy district should qualify, because there's so much time for surprise retirements/scandals/huge waves to emerge. (IIRC, quite a few of the GOP's 2010 House gains were rated "Safe D" early in the cycle). Most of the time, most of these districts would get moved to "Safe" by early 2016, in the likely event that extraordinary occurences would not materalise. CT-05 and OH-06 are examples of such seats.

Unfortunately, pretty no-one bothers to "score" the early ratings. You could just pull ratings out of your ass and no-one will care so long as your final ratings do okay.

http://www.centerforpolitics.org/crystalball/articles/2014-house-ratings-democratic-potential-republican-predictability/

Republicans won 4 of 7 tossups and all but one each in the Lean/Likely categories; Democrats lost 9 of 13 in the Lean category and 4 of 14 in the Likely category, plus ME-2.

ME-02 is a example of a seat that should definitely been "Likely D" early on, as Michaud retiring and/or running for another office was not out of the question, and at D+3 it was competitive on paper.

Should that be an indictment of his early ratings that seats switched in larger proportions than their ratings would indicate or praiseworth that he only failed to label one seat that would flip as competitive?

Not really either. There's such a wide variance in possible outcomes that "calling" the races from early on is close to impossible. It's that knowing what we knew in early 2013, ME-02 shouldn't have been Safe D, and that remains true even if it hadn't ended up flipping.

2010 was the real failure of these models. I can't find Sabato's early predictions, but Cook only had 50 Democratic House seats as less than "Safe" early on. Of course, Republicans netted 63 seats while not winning all of those seats, so ~20 "Safe D" seats ended up voting Republican.

Sanatorium/Cook/Rothenberg don't try to predict the outcome of the election, merely try to measure the current standings between candidates.

If I say a seat like, say, MI-05 is Safe D, and two weeks before Election Day Dan Kildee is revealed to be North Korean spy; then I can't really be criticized for saying Kildee's seat was safe. If I then kept the rating as Safe D after the fact, I could then be criticized for not being able to read the race's dynamics.

I also don't like the "This seat would be competitive if the incumbent retires! Likely D/R" train of thought , races should be judged by current candidates not potential ones.

[IceSpear]

Why isn't Gwen Graham in the toss-up category? 

Because she just defeated a Republican incumbent in a low turnout Republican wave.

LOL at having NV-04 and IA-01 as "toss up".

[IceSpear]

These ratings as a whole seem to give way too much credit to Republicans who rode the wave. Poliquin, Young, and Zeldin as lean R? Hardy and Blum as toss up? Jolly as likely R? He barely won the special election and didn't face a Democratic opponent this time because the FL Dems screwed themselves over. MacArthur as likely R? Slaughter and Delaney not being safe?

If you assume 2016 will be a Republican wave, then these ratings seem more plausible.

[Attorney General, LGC Speaker, and Former PPT Dwarven Dragon]

These ratings as a whole seem to give way too much credit to Republicans who rode the wave. Poliquin, Young, and Zeldin as lean R? Hardy and Blum as toss up? Jolly as likely R? He barely won the special election and didn't face a Democratic opponent this time because the FL Dems screwed themselves over. MacArthur as likely R? Slaughter and Delaney not being safe?

If you assume 2016 will be a Republican wave, then these ratings seem more plausible.

There were absolutely NO national dollars in MD-6 or NY-25, yet the republicans almost won in both districts. Without knowing the climate of 2016, they can't be rated as safe just yet.

[IceSpear]

These ratings as a whole seem to give way too much credit to Republicans who rode the wave. Poliquin, Young, and Zeldin as lean R? Hardy and Blum as toss up? Jolly as likely R? He barely won the special election and didn't face a Democratic opponent this time because the FL Dems screwed themselves over. MacArthur as likely R? Slaughter and Delaney not being safe?

If you assume 2016 will be a Republican wave, then these ratings seem more plausible.

There were absolutely NO national dollars in MD-6 or NY-25, yet the republicans almost won in both districts. Without knowing the climate of 2016, they can't be rated as safe just yet.

Yes, there are some surprisingly close margins in giant waves. But if we're going to account for that possibility, then the "likely R" column should be much, much larger. Not knowing the political climate, it's pretty asinine to have say, CA-25 and AR-02 as "safe R" but only have CA-16, NY-25, etc. as "likely D". Considering our volatile political climate, it seems pretty silly to be so overly dependent on the results of the most recent election, as Stu Rothenberg would know with his "there's NO WAY Republicans will retake the House!" article in early 2009.

[Nichlemn]

Nothing strikes me as conspicuously out of place, but I just think that rating only 56 districts as something other than "Safe" two years out before an election is too few. If we are to take "Safe" as representing a probability below (say) 5% of the seat switching,  I think pretty much any remotely swingy district should qualify, because there's so much time for surprise retirements/scandals/huge waves to emerge. (IIRC, quite a few of the GOP's 2010 House gains were rated "Safe D" early in the cycle). Most of the time, most of these districts would get moved to "Safe" by early 2016, in the likely event that extraordinary occurences would not materalise. CT-05 and OH-06 are examples of such seats.

Unfortunately, pretty no-one bothers to "score" the early ratings. You could just pull ratings out of your ass and no-one will care so long as your final ratings do okay.

http://www.centerforpolitics.org/crystalball/articles/2014-house-ratings-democratic-potential-republican-predictability/

Republicans won 4 of 7 tossups and all but one each in the Lean/Likely categories; Democrats lost 9 of 13 in the Lean category and 4 of 14 in the Likely category, plus ME-2.

ME-02 is a example of a seat that should definitely been "Likely D" early on, as Michaud retiring and/or running for another office was not out of the question, and at D+3 it was competitive on paper.

Should that be an indictment of his early ratings that seats switched in larger proportions than their ratings would indicate or praiseworth that he only failed to label one seat that would flip as competitive?

Not really either. There's such a wide variance in possible outcomes that "calling" the races from early on is close to impossible. It's that knowing what we knew in early 2013, ME-02 shouldn't have been Safe D, and that remains true even if it hadn't ended up flipping.

2010 was the real failure of these models. I can't find Sabato's early predictions, but Cook only had 50 Democratic House seats as less than "Safe" early on. Of course, Republicans netted 63 seats while not winning all of those seats, so ~20 "Safe D" seats ended up voting Republican.

Sanatorium/Cook/Rothenberg don't try to predict the outcome of the election, merely try to measure the current standings between candidates.

If I say a seat like, say, MI-05 is Safe D, and two weeks before Election Day Dan Kildee is revealed to be North Korean spy; then I can't really be criticized for saying Kildee's seat was safe. If I then kept the rating as Safe D after the fact, I could then be criticized for not being able to read the race's dynamics.

I also don't like the "This seat would be competitive if the incumbent retires! Likely D/R" train of thought , races should be judged by current candidates not potential ones.

But two years out from an election, there is often very little idea of who the challenger will be, and reasonable uncertainty as to whether the incumbent will run again. There is no "race as it stands now". It has to involve some kind of prediction about how the future will turn out.

In your scenario it wouldn't be unreasonable to have ranked the race as Safe D, because the probability of a Congressman having such an extraordinarily scandal uncovered just before the election is very low, which is consistent with "Safe" meaning a probability such as 95%+. But if we're talking 2 years out in a seat that is quite swingy at the Presidential level, there are many paths that could lead to the party losing the seat. Even if they're each quite unlikely by themselves, together they could add up to something quite substantial.

Like if I could get a sneak peak to 2016 and found out that MI-05 had gone Republican, I'd be amazed. Something must have gone terribly wrong for Kildee and/or national Democrats. But if I saw that CT-05 had gone Republican, it would be be a bit of surprise but nothing extraordinary. Maybe Republicans won the Presidential election decisively and got some coattails, or maybe Esty had a tax fraud problem or something, or maybe Republicans had a really good candidate, or most likely some combination of the prior.