Daily Kos Elections Governors Forecast: Democrats could win in 9 of the 10 most populous states

Map of competitiveness of 2018 gubernatorial elections (dark grey states are Tossups; light grey states aren't up this year)

The typical way to think about how to “keep score” in gubernatorial elections is kind of a dumb one: essentially, it’s one point for each state, so if your party ends up electing the governor in 26 or more states, you’re the “winner.” As I’ve pointed out before, though, you don’t get any sort of bonus for controlling the majority of state houses. The party with 26 governors doesn’t get to tell the governors in the other 24 states what to do. (In fact, the majority party doesn’t even get to elect the chair of the National Governors’ Association, which isn’t much of a prize anyway, policy-wise; the chair simply rotates between the parties every cycle.) And if you’ve elected 26 governors in 26 disproportionately-small states, then your party is still representing only a minority of the country.

Our colleagues at FiveThirtyEight released their governors forecast today, and they did one thing that was really innovative: the topline result, instead of how many states the parties are on track to win, was how much of the population each party is on track to represent. Their answer is: the Democrats are on track to represent 194 million, or 59 percent of the nation, while the Republicans are on track to represent 135 million, or 41 percent. (Beyond that, their predictions are pretty much the same as where we are as of today: Democrats are on track to pick up eight governorships, which would take them up to 24.)

If you’re wondering how 24 states make up that much of the nation’s population, well, here’s a chart of our own creation that illustrates that point. Nine of the ten most-populous states are up this year, and while six of them are currently held by Republicans, the polling averages in three of them are pointing to Democratic pickups (with two more right on the cusp, either tied or with a low-single-digits deficit). Here are the races organized, rather than according to how competitive they are, according to which are the ten most populous states:

STATE	D CAND.	D AVG.	R CAND.	R AVG.	DIFF.	FLIP?
CALIFORNIA	Newsom	53	Cox	38	+15	STAYS D
TEXAS	Valdez	36	Abbott (inc.)	56	-20	STAYS R
FLORIDA	Gillum	46	De Santis	45	+1	D FLIP
NEW YORK	Cuomo (inc.)	47	Molinaro	25	+22	STAYS D
PENNSYLVANIA	Wolf (inc.)	50	Wagner	37	+13	STAYS D
ILLINOIS	Pritzker	48	Rauner (inc.)	29	+19	D FLIP
OHIO	Cordray	40	De Wine	40	0	???
GEORGIA	Abrams	45	Kemp	47	-2	STAYS R
NO. CAROLINA	NOT UP		THIS YEAR			STAYS D
MICHIGAN	Whitmer	47	Schuette	38	+9	D FLIP

Keep in mind that only five of these races appear on our usual “totem pole,” because, in addition to North Carolina not being up this year, California, New York, and Pennsylvania are all designated “Safe Democratic” while Texas is “Safe Republican” (we only include the ones that we consider competitive, in our qualitative ratings). Ironically, the Democratic candidates in California and Pennsylvania are winning by less in our polling averages than J.B. Pritzker in Illinois (which we still consider “Likely Democratic,” though, given incumbent Bruce Rauner’s near-hopeless position, that may change soon).

Now let’s switch over to the normal table, where we list all the competitive races in order of their margins, in today’s polling averages:

STATE	D CAND.	D AVG.	R CAND.	R AVG.	DIFF.	FLIP?
ILLINOIS	Pritzker	48	Rauner (inc.)	29	+19	D FLIP
RHODE ISLAND	Raimondo (inc.)	43	Fung	33	+10
MICHIGAN	Whitmer	47	Schuette	38	+9	D FLIP
COLORADO	Polis	46	Stapleton	38	+8
MAINE	Mills	47	Moody	39	+8	D FLIP
CONNECTICUT	Lamont	45	Stefanowski	38	+7
MINNESOTA	Walz	46	Johnson	41	+5
NEW MEXICO	Lujan Grisham	47	Pearce	43	+4	D FLIP
OREGON	Brown (inc.)	43	Buehler	39	+4
IOWA	Hubbell	39	Reynolds (inc.)	36	+3	D FLIP
WISCONSIN	Evers	47	Walker (inc.)	45	+2	D FLIP
SOUTH DAKOTA	Sutton	45	Noem	43	+2	D FLIP
FLORIDA	Gillum	46	De Santis	45	+1	D FLIP
KANSAS	Kelly	38	Kobach	38	0
OHIO	Cordray	40	De Wine	40	0
GEORGIA	Abrams	45	Kemp	47	-2
NEVADA	Sisolak	42	Laxalt	45	-3
OKLAHOMA	Edmondson	41	Stitt	46	-5
ARIZONA	Garcia	40	Ducey (inc.)	50	-10
NEW HAMPSHIRE	Kelly	37	Sununu (inc.)	51	-14
MARYLAND	Jealous	35	Hogan (inc.)	54	-19
ALASKA	Begich	24	Dunleavy	44	-20	I TO R
SOUTH CAROLINA	Smith	33	McMaster (inc.)	56	-23

If you look back at last week’s version, at that point Democratic candidates were leading in the polling averages for six races (down from eight the week before). But that’s back up to eight this week! One of the races that accounts for that is Wisconsin, where the Marquette Law School poll released last week with a one-point lead for Scott Walker bumped the race down to a tie. That was the first poll since June to give Walker a lead. That tie only lasted a day or so, though, as another poll from Marist for NBC (who’ve generally seen this race in much rosier terms than Marquette has) showed up, giving Democrat Tony Evers at 50-42 lead, pushing Evers back above water.

It might be tempting to say “well, that Marquette poll shows the Kavanaugh bump, and then it went away.” For one thing, though, that theory doesn’t jibe with the field dates, if you look at the fine print: the Marist poll was in the field mostly before the Marquette poll, and only was released slightly later than Marquette. But more than anything, it should be another reminder that shifts in the polling averages generally aren’t about anything that happened in the news; most people are partisans whose voting intent is already baked in. What varies from week to week, more than anything else, is which pollsters are releasing polls that week, and what kind of model those pollsters use. (As I alluded to, across multiple polls, Marist has been giving noticeably friendlier results for Democrats in its polls of midwestern states than other pollsters are finding, Wisconsin included.)

Keep in mind, though, that one pollster’s model of the expected electorate might give it a bit of a house effect in one state, while not having that same effect in a different state with a different demographic mix. In fact, Marist’s poll of Nevada’s governor’s race, performed at the same time as their Wisconsin poll, is one of the main reasons that Democratic candidate Steve Sisolak has fallen out of the lead in the Silver State in the last few weeks; they found him trailing Republican Adam Laxalt 44-40.

So if Nevada was a state where the Democrat led in the polling averages, but has fallen out since then, who’s the new state that’s taken its place (in order to get the Democratic range of pickups back up to eight)? In a bit of a surprise, it’s Maine! The Pine Tree State is by far the biggest riser of the week, shooting up from a tie to an eight-point lead for Democratic Attorney General Janet Mills; she’s currently leading Republican Shawn Moody in the average by a 52-44 margin.

Much of the problem in Maine was simply how underpolled it was: that persistent tie was based on the only poll that had ever been taken of the Mills-Moody matchup, an August poll from Suffolk. (The result didn’t seem that odd to us, though, given that Maine’s blue-state status has taken something of a beating in the last few years, with the re-election of terrible Republican Governor Paul LePage and the loss of the House seat in Maine’s 2nd district in 2014, and then the loss of ME-02’s electoral vote in 2016. That decline seems consistent with Maine’s demographics and the shifting party coalitions: it’s one the whitest and oldest states in the nation.)

This week, though, two more polls showed up, each with an eight-point lead for Mills. I’m not sure how far you want to trust either of those polls, which are both from partisan pollsters; one is from Democratic firm Change Research, who’ve put up some dubious-looking results elsewhere this cycle and who didn’t seem to account for third-party candidates (of which there are several prominent ones, this being Maine). The other is from Slingshot Strategies, on behalf of one of those third-party candidates, Terry Hayes, apparently intended to show that Hayes is in the thick of this. (The poll had Hayes at 10, though, so Hayes isn’t really in the thick of this.)

As sketchy as those polls might be, though, collectively, they seem to show Mills taking command of the race, to the extent that it’s now the Democrats’ third-best pickup opportunity, margin-wise. Interestingly, while Maine is well-known for a long run of female Senators, Mills would be noteworthy for being Maine’s first female governor.