November 7th, 2017 saw a pretty remarkable night in Virginia electoral politics. Most of the attention went to the Democrats sweeping statewide offices, and they did. The second biggest story, although I think it is the biggest story, was the 15 Democrats that flipped Republican House of Delegates seats.
That is a truly big deal. It moves Democrats from an extremely weak position, 1 seat away from Republicans having a super majority in the chamber, to a position of near equity in the chamber. There are some fantastic stories there as well. Transgender Delegate Elect Danica Roem brought down Republican Bob Marshall the author of the infamous Bathroom Bill. Delegate Elect Elizabeth Guzman became the first Latina to win a seat in Virginia’s state legislature. 13 other amazing Democrats add to that sampling of good news for Democrats.
There are even 3 races still in recounts. Recounts are hard to win from behind, but it remains a big deal what these candidates have done as well.
So what happened in the Virginia election.
Going into election night Republicans controlled 66 of 100 seats in the Virginia House of Delegates. That seemed like a safe number for them. It is no longer. In 2015 Democrats did not even contest 44 races, leaving Republicans the task of only needing to win 7 contested races to secure a majority in that cycle.
As the old saw goes in sports “you miss 100% of the shots you don’t take.” In 2015 we missed 44 shots because we did not take them, leaving us making only 34 of 56 shots we did take. If you are a basketball fan you know hitting 61% of the shots you take is pretty darn good, but you still need to take enough shots to beat the opponent.
This year Democrats challenged 88 seats out of the hundred, leaving only 12 districts uncontested.This time we only hit 55% of our shots (which is still an amazing number in basketball) but we actually took enough shots to win this thing, and we nearly did. So that is great news.
That said, not taking those 12 other shots represent part of a big missed opportunity here. We also missed executing our fundamentals in other shots that had an impact on us missing a chance to clearly secure the majority in the chamber.
So what were the missed opportunities?
Not fielding a candidate has negative impacts on vote totals. The numbers are pretty irrefutable when you look at them. Before I dig into some numbers I want to make a quick note on methodology here. To assess the impacts of running versus not running I have looked at the Gubernatorial campaign results in all districts. This is one race where Republican and Democratic votes occurred in every single House of Delegate seat, so it provides some stability across districts for comparison across election cycles.
Looking at the vote totals in the Governor’s race by House of Delegates District has the added bonus of shining some light on the old up the ballot vs down-ballot impacts.
So, after reconciling the data (Virginia reports governors results by congressional district, not House of Delegate District AND they have split precincts, so reconciling this was a tricky and tedious business) and sort of living in it for a few weeks I have observed a few things:
- Running matters. Running in down-ballot races matters to up-ballot vote totals, so not fielding candidates in 12 House of Delegates races resulted in lower vote totals for Democrats at the top of the ticket.
- Properly funding a broader range of races will yield better results up, down and across the ballot. Resources matter and, aside from the 12 races we did not challenge, there were some races we ignored as a party that with some attention would have yielded better results across the board.
I will mix in narrative with some data tables and bullet points as we dig into the data.
Running Matters — Democratic Votes in Democratic Wins
It was a big year in Virginia. Turnout was through the roof for an odd-year election. The Democratic vote for Governor increased 393,556 votes from 2013 to 2017. That is an average increase of 3,935 votes per House of Delegates District. The totals in the 49 races we know we won we drew 266,386 additional votes more than 2013 at an average of 5,436 vote increase per district.
Democratic Votes for Governor by House of Delegates District
Race Type |
Count |
2013 Gov Vote |
2017 Gov Vote |
Total Increase |
Vote/District 2013 |
Vote/District 2017 |
Increased
Vote/District
|
Democratic Wins |
49 |
600,436 |
866,822 |
266,386 |
12,254 |
17,690 |
5,436 |
Republican Wins |
51 |
413,648 |
540,818 |
127,170 |
8,111 |
10,604 |
2,494 |
Totals |
100 |
1,014,084 |
1,407,640 |
393,556 |
10,141 |
14,076 |
3,936 |
In these races, we raised $10,000,000 more in races that we won than in the races we lost. More importantly, that represented $16 per vote in races we won versus just $7 per vote in races we lost. For the sake of comparison, it is worth translating that into additional votes won rather than all votes. In that case, we raised $22 more per vote. The average winning race for Democrats raised $277,822 versus the $73,356 for the races we lost.
Money Raised for Democrats per district
Race Type |
Count |
Dem Money |
% of Total Raised |
Dem Money/District |
$ per Vote |
$ per Added Vote |
% Change in Total Vote |
Democratic Wins |
49 |
$13,613,263 |
78% |
$277,822 |
$16 |
$51 |
44% |
Republican Wins |
51 |
$3,741,162 |
22% |
$73,356 |
$7 |
$29 |
31% |
Totals |
100 |
$17,354,425 |
|
$173,544 |
$12 |
$44 |
39% |
To dig deeper, let’s consider the obvious. Democrats held 34 seats coming into this election cycle. comparing governors race vote totals in those seats from 2013 to 2017 we see a huge increase in Democratic vote total at around 180,057 votes. That is an increase in the Democratic vote of 5,296 votes per district. An important thing to note about these 34 districts is that Democrats have run for these seats in each of the last 3 cycles and we held them all this year, 33 of those were Democratic seats in each election.
Let’s also look at the 15 seats we have flipped. The increase in the overall Democratic vote in these districts was around 86,329 or 5,755 votes per district.
DEMOCRATIC WINS BY TYPE: DEMOCRATIC VOTES FOR GOVERNOR BY HOUSE OF DELEGATES DISTRICT
Race Type |
Count |
2013 Gov Vote |
2017 Gov Vote |
Total Increase |
Vote/District 2013 |
Vote/District 2017 |
Increased Vote/District |
Democratic Seat |
34 |
449,342 |
629,399 |
180,057 |
13,216 |
18,512 |
5,296 |
Democratic Flips |
15 |
151,094 |
237,423 |
86,329 |
10,073 |
15,828 |
5,755 |
Total 2017 Dem Wins |
49 |
600,436 |
866,822 |
266,386 |
12,254 |
17,690 |
5,436 |
In terms of resources raised per race, there was $2.5 million more raised for the 15 seats we flipped than the seats we already held. We raised on average $546,572 per seat we flipped. There was a wide range in that group. We will be breaking down the Democratic flips further down this post. The 15 seats we flipped do represent 47% of all the money raised by Democrats for all of the House races we competed for. That is nearly half the money raised allocated to 15% of the potential races within the state.
Democratic Wins by Type: MONEY RAISED FOR DEMOCRATS PER DISTRICT
Race Type |
Count |
Dem Money |
% of Total Raised |
Dem Money/District |
$ per Vote |
$ per Added Vote |
% Change in Total Vote |
Democratic Seat |
34 |
$5,414,679 |
31% |
$159,255 |
$9 |
$30 |
40% |
Democratic Flips |
15 |
$8,198,584 |
47% |
$546,572 |
$35 |
$95 |
57% |
Total 2017 Dem Wins |
49 |
$13,613,263 |
78% |
$277,822 |
$16 |
$51 |
44% |
The already held seats represent 31% of the money raised by Democrats did yield 46% of the total increased vote for the Democratic gubernatorial campaign. That said, the Northam campaign probably spent heavily in these districts too, so you likely found this group sort of turbo charged.
Detail on the Democratic Flips
There were several varieties of contests here. This is where we begin to see some signs of why running in down-ballot races is critical.
- There were 9 of those 15 races that we flipped that had a Democratic challenger in each of the last 3 cycles. The vote increase for these districts came in at 5,477 votes per district.
- There were 3 of those 15 HoD races that we flipped that had a Democratic challenger in only the 2013 and 2017 campaigns. These races yielded a 5,677 vote increase per district.
- 1 of those races had a Democratic candidate in both 2015 and 2017. This race yielded 8,305 more votes for the Democratic gubernatorial candidate this year rather than in 2013. If you are running a state legislative campaign I recommend you contact this Democratic Delegate, Dawn Adams. She did this on $5 per vote.
- Finally, 2 of the 15 flipped districts were only contested by Democrats this cycle out of the last three. These races yielded a 5,850 increase in votes for Governor over the 2013 election cycle.
DEMOCRATIC Flips: DEMOCRATIC VOTES FOR GOVERNOR BY HOUSE OF DELEGATES DISTRICT
Race Type |
Count |
2013 Gov Vote |
2017 Gov Vote |
Total Increase |
Vote/District 2013 |
Vote/District 2017 |
Increased Vote/District |
Ran all 3 years |
9 |
85,437 |
134,729 |
49,292 |
9,493 |
14,970 |
5,477 |
Ran in 13 and 17 only |
3 |
30,712 |
47,744 |
17,032 |
10,237 |
15,915 |
5,677 |
Ran in 15 and 17 only |
1 |
13,919 |
22,224 |
8,305 |
13,919 |
22,224 |
8,305 |
Only ran in 17 |
2 |
21,026 |
32,725 |
11,699 |
10,513 |
16,363 |
5,850 |
Total for Flipped Seats |
15 |
151,094 |
237,423 |
86,329 |
10,073 |
15,828 |
5,755 |
I mean, this part of the analysis is a no-brainer, where we won we did really well. One take away, generally speaking, the races we touched fewer times in past years showed the greatest improvement in votes for Governor from 2013 to 2017 than ones we have had more focus on in past cycles.
This is likely a result of the district we compete for every cycle, win or lose, being somewhat better organized so there is slightly less room for growth, while campaigns that we don’t contest every year are somewhat less organized and a viable Democratic candidate for these seats boost our numbers at a higher rate. Having Delegate candidates clearly added vote totals to the Governor’s Democratic vote total. While the differences may seem narrow, we have seen elections come down to razor thin margins before and they will again.
Let’s take a peek at the resource breakdown for these 15 flipped seats:
Democratic Flips: MONEY RAISED FOR DEMOCRATS PER DISTRICT
Race Type |
Count |
Dem Money |
% of Total Raised |
Dem Money/District |
$ per Vote |
$ per Added Vote |
% Change in Total Vote |
Ran all 3 years |
9 |
$5,363,205 |
31% |
$595,912 |
$ 40 |
$109 |
58% |
Ran in 13 and 17 only |
3 |
$1,969,664 |
11% |
$656,555 |
$ 41 |
$116 |
55% |
Ran in 15 and 17 only |
1 |
$103,949 |
1% |
$103,949 |
$5 |
$13 |
60% |
Only ran in 17 |
2 |
$761,766 |
4% |
$380,883 |
$23 |
$ 65 |
56% |
Total for Flipped Seats |
15 |
$8,198,584 |
47% |
$546,572 |
$35 |
$95 |
57% |
In terms of resources, we did raise huge amounts of money per vote. 31% of the $17 million raised for all delegates races by Democrats were in the 9 seats we contest regularly. That is a massive proportion out of 100 potential races. It is a large amount out of the 88 races we did contend as well. On the flip end we got more value per dollar out of the 2 races where we only ran in this cycle. We aren’t talking about a large data pool here, but we must work with the info in front of us.
In terms of the up-ballot impact, these races raised 47% of all of the money Democrats raised this cycle. They only added 22% of the Democratic increase in votes, however. That is a weak yield in the end.
Running Matters — Democratic Votes in Republican Wins
Republicans current win count is 51, pending 3 recounts. In those races, we drew an additional 127,170 votes or so for the Democratic Governor’s races for 2,494 additional Democratic votes per district. Now that seems like these districts did not matter so much compared to the ones we won, but, when you dig down, you find that running matters even more in these races. Let’s look at the Democratic performance in a variety of types of races to see what I mean:
- In the 6 races where Republicans won where Democrats contested in each of the last 3 cycles saw an average Democratic vote increase of 4,046 votes. The total added votes for Governor in these 6 races is 24,277.
- In the 4 races where Dems did not compete in 2013 but did in 2015 and this year we saw an average increase of 3,741 Democratic votes per district. The total added votes for Governor in these 4 races is 14,965.
- In 9 races where Dems competed in 2013 and 2017 but not 2015 the Democratic vote increase dropped to 2,311 per district. The total added votes for Governor in these 9 races is 20,802.
- In the 20 races where we only competed this year, but not the prior 2, we saw an average increase of 2,728 Democratic votes for Governor. The total added votes for Governor in these 20 races is 54,544 (14% of total Democratic Governor vote increase).
Contested Republican Wins: DEMOCRATIC VOTES FOR GOVERNOR BY HOUSE OF DELEGATES DISTRICT
Race Type |
Count |
2013 Gov Vote |
2017 Gov Vote |
Total Increase |
Vote/District 2013 |
Vote/District 2017 |
Increased Vote/District |
Dem Ran All 3 Years |
6 |
57,757 |
82,034 |
24,277 |
9,626 |
13,672 |
4,046 |
Ran in 15 and 17 |
4 |
36,680 |
51,645 |
14,965 |
9,170 |
12,911 |
3,741 |
Dem Ran 13 and 17 |
9 |
68,790 |
89,592 |
20,802 |
7,643 |
9,955 |
2,311 |
17 only |
20 |
162,682 |
217,236 |
54,554 |
8,134 |
10,862 |
2,728 |
Totals in Contested Republican Wins |
39 |
325,909 |
440,506 |
114,597 |
8,357 |
11,295 |
2,938 |
So the above 4 race types do make it clear we competed in districts we think we can win, and those did show the most improvement. Turning to resources per race this is what we find:
Contested Republican Wins: Democratic Monies Raised per House District
Race Type |
Count |
Dem Money |
% of Total Raised |
Dem Money/District |
$ per Vote |
$ per Added Vote |
% Change in Total Vote |
Dem Ran All 3 Years |
6 |
$1,685,286 |
10% |
$280,881 |
$21 |
$69 |
42% |
Ran in 15 and 17 |
4 |
$383,819 |
2% |
$95,955 |
$7 |
$26 |
41% |
Dem Ran 13 and 17 |
9 |
$536,034 |
3% |
$59,559 |
$6 |
$26 |
30% |
17 only |
20 |
$1,136,023 |
7% |
$56,801 |
$5 |
$21 |
34% |
Totals in Contested Republican Wins |
39 |
$3,741,162 |
22% |
$73,356 |
$7 |
$29 |
37% |
None of these race types raised even the $35 average amount raised per vote for the Flipped seats. Basically this data shows a large number of underfunded races. You saw 22% of all money raised go to these races and they returned 30% of the total Democratic vote increase for the gubernatorial campaign. That is good value, particularly considering the Northam campaign did not put as many resources as he had in the 34 Democratic seats coming into the election. Imagine had we properly funded even some of these races what the yield could have been.
Combined with the higher vote increases per district found in the seats we won it might be a knock on the concept of a positive impact on up-ballot races. That is until you compare this group to the Republican wins where we did not field any candidate:
- In the 6 races where we only fielded a candidate in 2013 and not this year our Governor vote total only increased by 1,188 votes.
- In the 2 races where we competed in 2015 but not this year the Democratic Governor only picked up 1,100 votes per district more than the 2013 total.
- In the 4 races we have not contested at all over the 4 cycles the Democratic governor only picked up 812 more votes than the Dem Gov pulled in in 2013.
Ignored Districts: DEMOCRATIC VOTES FOR GOVERNOR BY HOUSE OF DELEGATES DISTRICT
Race Type |
Count |
2013 Gov Vote |
2017 Gov Vote |
Total Increase |
Vote/District 2013 |
Vote/District 2017 |
Increased Vote/District |
Dem Ran only 13 |
6 |
43,735 |
50,860 |
7,125 |
7,289 |
8,477 |
1,188 |
Ran only 15 |
2 |
15,184 |
17,384 |
2,200 |
7,592 |
8,692 |
1,100 |
No Dem Any Year |
4 |
28,820 |
32,068 |
3,248 |
7,205 |
8,017 |
812 |
No Democratic Candidate Totals |
12 |
87,739 |
100,312 |
12,573 |
7,312 |
8,359 |
1,048 |
On average, in each of these categories we still saw improvement. Even though we did not invest here. That said we only saw around 1/3rd of the increaded vote per district that we saw in similar races we had only contested in 2017. That is 1,800 votes per district added fewer. That is potentially 21,600 potential votes that, even if we ran an underfunded candidate, we left on the table. We did not need them this time, but maybe in another cycle that could be the difference (Think Wisconsin 2016).
I will skip the money table for this group since they raised none however they raised 0% of the money and returned 3% of the increased vote totals.
RESOURCE ALLOCATiON Tables
Resource allocation has been a big topic in the special elections this year. Many of us felt we way way way overspent on the GA6th for example. Let’s see the resource allocation totals in Virginia in a few summary tables.
Democratic Financial Resource Allocation by Win Type
Win Type |
Count |
Total 2017 Dem Votes for Governor |
% of Vote Total |
Total Dem Money Raised |
% of Dem Money Raised |
% Point Yield |
Democratic Seats |
34 |
629,399 |
45% |
$5,414,679 |
31% |
14% |
Democratic Flips |
15 |
237,423 |
17% |
$8,198,584 |
47% |
-30% |
Contested Republican Wins |
39 |
440,506 |
31% |
$3,741,162 |
22% |
10% |
Uncontested Republican Wins |
12 |
100,312 |
7% |
$- |
0% |
7% |
Totals |
100 |
1,407,640 |
100% |
$17,354,425 |
100% |
0% |
So, what you see in this table are the votes and money raised by win type. The 34 seats that were already held by Democrats brought in the most votes at 629,399 votes, which represents 45% of the total vote Northam drew within the state. $5,414,679 were raised in those races, 31% of all the money Democrats raised for these seats. The final column represents the difference in rate from the amount of vote 45% minus the rate of money raised 31%. The vote yield at a rate of total percent raised versus the number of votes drawn is 14 percentage points.
I do think the most interesting thing here is what I mentioned above, nearly half of the money raised by Democrats was in races that got 17% of the votes we needed for the governor's race. Not included in any of these tables is the vote margins per district. We won the Democratic flips by roughly 4,200 votes per district while we lost the contested Republican races by 3,900 votes on average. Tweaking how resources were allocated might have yielded us many more wins when you consider these facts.
Of course, I have spent most of the time looking at Democratic results, we run against Republicans. So what happened in terms of resource allocation for our opponents? Let’s take a look:
Republican Resource Allocate by Win Type
Win Type |
Count |
Republican Votes |
% of Vote Total |
Total Rep Money Raised |
% of Rep Money Raised |
Republican Yield |
Democratic Seats |
34 |
222,671 |
19% |
$1,494,401 |
8% |
11% |
Democratic Flips |
15 |
173,305 |
15% |
$6,724,264 |
38% |
-23% |
Contested Republican Wins |
39 |
593,839 |
51% |
$7,695,985 |
43% |
7% |
Uncontested Republican Wins |
12 |
185,106 |
16% |
$1,955,254 |
11% |
5% |
Totals |
100 |
1,174,921 |
100% |
$17,869,904 |
100% |
0% |
Both Republicans and Democrats spent much more on those 15 seats Democrats flipped than returned in overall governor’s vote totals for their candidate. I think the most telling figure however is that 43% of Republican money was raised to defend the 39 seats Democrats contested. More than half of that amount (23% of total raised by Republicans) went to defend 20 seats that Democrats did not contest in 2015 or 2013. Tying down 23% of total Republican respurces on races we ordinarily do not contest provide reason enough to push to challenge for every single seat.
Virginia Republicans have a long history of fundraising dollars coming from other campaign warchests. Tying that money down in races they normally do not have to defend probably had a huge impact in enabling to flip the seats we did.
SOME CONCLUSIONS
There is a lot of data to consider up there. And looking race by race compared to these aggregate numbers could help us find some real missed opportunities. For example 6 of our seats we run every year, but lose, only lost by an average of 70 votes. 4 other races lost by an average of 758 votes as well. More money shifted there could have easily made the difference.
All in all, Virginians did well this cycle, just running additional candidates tied up Republican resources creating openings for us to score some of our flips. More importantly, the 20 races we ran in this cycle for the first time in the last three, returned a great value per increased vote over 2013 Democratic votes for the governor’s race at $21 per added vote versus $51 per added vote for the races we won.
Running in “red districts” state legislative districts is a great value-add for up-ballot races.
What are your thoughts?