Analysis of the 2016 Presidential Race via R Shiny

Clicking on the OFFICE select list shows that the OpenElections data for Michigan also includes the votes for the State House. Substituting the State House for the U.S. House in the prior steps results in the following steps:

1. Set STATE to MI
2. Set YEAR to 2016
3. Set ELECTION to 20161108__mi__general
4. Set COUNTY to (all)
5. Set OFFICE to President
6. Click the "ADD RACE" button
7. Set OFFICE to "State House"
8. Click the "ADD RACE" button
9. Set Units to "Percent ratio"
10. Click the "Area Plot2" tab
11. Uncheck the "Show all labels" checkbox
12. Set "Y From,To,Step,Tick" to "95,115" (without the quotes)
13. Uncheck the X-axis checkbox
14. Set "Color (regression)" to def

As can be seen, the regression line is a little higher than for the U.S. House, slanting from about 105 percent in the strong Republican precincts to about 103 percent in the strong Democratic precincts. This would seem expected since the State House race is likely further "down-ballot" than the U.S. House election.

In order to look more closely at those precincts for which the ratio is below 100 percent, do the following steps:

1. Check "Show all labels"
2. Set "Label type" to County
3. Set "Y From,To,Step,Tick" to "95,100" (without the quotes)
4. Set "Limit (regression)" to "95,115" (without the quotes)

This should display the following plot:

The above plot shows the counties of those Michigan precincts in which there were surprisingly more votes in the State House race than in the Presidential race. As before, many of the most solid Democrat precincts (on the far right) were in Wayne County. Also noticable is that there appears to be far fewer precincts than there were when comparing to the U.S. House though the range seems similar.

The prior steps can be changed to compare the Presidential and U.S. House votes in just Wayne County as follows:

1. Set STATE to MI
2. Set YEAR to 2016
3. Set ELECTION to 20161108__mi__general
4. Set COUNTY to Wayne
5. Set OFFICE to President
6. Click the "ADD RACE" button
7. Set OFFICE to "U.S. House"
8. Click the "ADD RACE" button
9. Set Units to "Percent ratio"
10. Click the "Area Plot2" tab
11. Uncheck the "Show all labels" checkbox
12. Set "Y From,To,Step,Tick" to "95,115" (without the quotes)
13. Uncheck the X-axis checkbox
14. Set "Color (regression)" to def

As can be seen, the regression line is again a little higher than for the U.S. House in all of Michigan. More noticable, however, is that it slants quite a bit more than the prior one-plus percentage, slanting about 4 percent from about 106 percent in the strong Republican precincts to about 102 percent in the strong Democratic precincts. Also interesting is what appears to be a near-solid horizontal line at 100 percent on the y-axis and between about 90 and 98 percent on the x-axis.

In order to look more closely at those precincts for which the ratio is below 100 percent, do the following steps:

1. Check "Show all labels"
2. Set "Label type" to Area
3. Set "X From,To,Step,Tick" to "70,100" (without the quotes)
4. Set "Y From,To,Step,Tick" to "98,100" (without the quotes)
5. Set "Limit (regression)" to "95,115" (without the quotes)

As can be seen, nearly all of the precincts appear to be in Detroit. In fact, you can get a list of the precincts via the following steps:

1. Click the Areas2 tab
2. Set Units to Count
3. Set "Areas2 (col)" to 14
4. Check the Desc checkbox below the "Areas2 (col)" input

Wayne County, MI: Shift in Margin Votes from 2016 President to 2016 U.S. House (Count)

     COUNTY                        AREA Clinton_D Trump_R MARGIN1  TOTAL1 Dingell_D Gorman_R MARGIN2  TOTAL2   DEM_SH   REP_SH   MAR_SH  TOT_SH
1     WAYNE            DETROIT CITY 274       295       8     287     303         0        8      -8     308     -295        0     -295       5
2     WAYNE       DETROIT CITY 960 AVCB       214       5     209     221         0        5      -5     224     -214        0     -214       3
3     WAYNE            DETROIT CITY 270       214       5     209     226         0        8      -8     229     -214        3     -217       3
4     WAYNE       DETROIT CITY 968 AVCB       226       3     223     229         0        3      -3     231     -226        0     -226       2
5     WAYNE       DETROIT CITY 962 AVCB       217       3     214     220         0        2      -2     222     -217       -1     -216       2
6     WAYNE       DETROIT CITY 961 AVCB       312       9     303     324         0        7      -7     326     -312       -2     -310       2
7     WAYNE       DETROIT CITY 959 AVCB       366       5     361     373         0        4      -4     375     -366       -1     -365       2
8     WAYNE            DETROIT CITY 490        19       0      19      21         0        0       0      23      -19        0      -19       2
9     WAYNE            DETROIT CITY 434       364       6     358     379         0        6      -6     381     -364        0     -364       2
10    WAYNE            DETROIT CITY 423       344      11     333     369         0       21     -21     371     -344       10     -354       2
11    WAYNE            DETROIT CITY 401       322       5     317     329         0        6      -6     331     -322        1     -323       2
12    WAYNE      DETROIT CITY 1015 AVCB       364      15     349     385         0       17     -17     387     -364        2     -366       2
13    WAYNE      DETROIT CITY 1007 AVCB       404       3     401     409         0        0       0     411     -404       -3     -401       2
14    WAYNE       DETROIT CITY 975 AVCB       306      10     296     320         0       11     -11     321     -306        1     -307       1
15    WAYNE       DETROIT CITY 971 AVCB       332       3     329     335         0        3      -3     336     -332        0     -332       1
16    WAYNE       DETROIT CITY 953 AVCB       266       3     263     272         0        3      -3     273     -266        0     -266       1
17    WAYNE       DETROIT CITY 916 AVCB       309      22     287     338         0        0       0     339     -309      -22     -287       1
18    WAYNE            DETROIT CITY 446       244       8     236     259         0        5      -5     260     -244       -3     -241       1
19    WAYNE            DETROIT CITY 290       247       2     245     251         0        8      -8     252     -247        6     -253       1
20    WAYNE            DETROIT CITY 218       215       0     215     217         0        6      -6     218     -215        6     -221       1
21    WAYNE            DETROIT CITY 161       517      13     504     559         0       14     -14     560     -517        1     -518       1
22    WAYNE            DETROIT CITY 142       279       5     274     288         0       10     -10     289     -279        5     -284       1
23    WAYNE             DETROIT CITY 13       271       5     266     279         0        0       0     280     -271       -5     -266       1
24    WAYNE      DETROIT CITY 1069 AVCB        47       3      44      50         0        0       0      51      -47       -3      -44       1
25    WAYNE      DETROIT CITY 1066 AVCB        47       1      46      48         0        1      -1      49      -47        0      -47       1
26    WAYNE      DETROIT CITY 1063 AVCB        79       0      79      79         0        1      -1      80      -79        1      -80       1
27    WAYNE      DETROIT CITY 1041 AVCB       440       5     435     451         0        5      -5     452     -440        0     -440       1
28    WAYNE      DETROIT CITY 1035 AVCB       367      25     342     393         0       20     -20     394     -367       -5     -362       1
29    WAYNE      DETROIT CITY 1023 AVCB       353      13     340     369         0        5      -5     370     -353       -8     -345       1
30    WAYNE        HIGHLAND PARK CITY 7       107       1     106     112         0        2      -2     112     -107        1     -108       0
31    WAYNE       HIGHLAND PARK CITY 25       150       6     144     159         0        4      -4     159     -150       -2     -148       0
32    WAYNE       HIGHLAND PARK CITY 15       420      16     404     440         0       11     -11     440     -420       -5     -415       0
33    WAYNE       DETROIT CITY 987 AVCB       236       6     230     247         0        3      -3     247     -236       -3     -233       0
34    WAYNE       DETROIT CITY 986 AVCB       224      32     192     262         0        0       0     262     -224      -32     -192       0
35    WAYNE             DETROIT CITY 94       493      20     473     518         0        0       0     518     -493      -20     -473       0
36    WAYNE       DETROIT CITY 918 AVCB       316       6     310     323         0        0       0     323     -316       -6     -310       0
37    WAYNE             DETROIT CITY 87        59       0      59      59         0        0       0      59      -59        0      -59       0
38    WAYNE             DETROIT CITY 58       245       5     240     256         0        0       0     256     -245       -5     -240       0
39    WAYNE            DETROIT CITY 469       214       2     212     223         0        6      -6     223     -214        4     -218       0
40    WAYNE            DETROIT CITY 455       118       4     114     122         0        2      -2     122     -118       -2     -116       0
41    WAYNE            DETROIT CITY 453       235       2     233     238         0        2      -2     238     -235        0     -235       0
42    WAYNE            DETROIT CITY 443        38       0      38      38         0        1      -1      38      -38        1      -39       0
43    WAYNE            DETROIT CITY 437       288       4     284     298         0        6      -6     298     -288        2     -290       0
44    WAYNE            DETROIT CITY 417       222       1     221     227         0        0       0     227     -222       -1     -221       0
45    WAYNE            DETROIT CITY 361       585       6     579     602         0        8      -8     602     -585        2     -587       0
46    WAYNE             DETROIT CITY 36       125       3     122     129         0        4      -4     129     -125        1     -126       0
47    WAYNE            DETROIT CITY 327       358      10     348     370         0        6      -6     370     -358       -4     -354       0
48    WAYNE             DETROIT CITY 28        56       5      51      64         0        5      -5      64      -56        0      -56       0
49    WAYNE            DETROIT CITY 239       365       8     357     383         0        0       0     383     -365       -8     -357       0
50    WAYNE            DETROIT CITY 224       190       2     188     192         0        0       0     192     -190       -2     -188       0
51    WAYNE            DETROIT CITY 195       253       4     249     261         0        0       0     261     -253       -4     -249       0
52    WAYNE            DETROIT CITY 194       339       7     332     346         0        0       0     346     -339       -7     -332       0
53    WAYNE            DETROIT CITY 180       137       2     135     140         0        0       0     140     -137       -2     -135       0
54    WAYNE            DETROIT CITY 122       213       8     205     222         0        7      -7     222     -213       -1     -212       0
55    WAYNE      DETROIT CITY 1071 AVCB        97       3      94     101         0        3      -3     101      -97        0      -97       0
56    WAYNE      DETROIT CITY 1070 AVCB        45       1      44      46         0        0       0      46      -45       -1      -44       0
57    WAYNE      DETROIT CITY 1067 AVCB        44       2      42      46         0        0       0      46      -44       -2      -42       0
58    WAYNE      DETROIT CITY 1058 AVCB        27       0      27      27         0        0       0      27      -27        0      -27       0
59    WAYNE      DETROIT CITY 1054 AVCB        33       4      29      38         0        3      -3      38      -33       -1      -32       0
60    WAYNE            DETROIT CITY 105       140       3     137     144         0        0       0     144     -140       -3     -137       0
61    WAYNE      DETROIT CITY 1019 AVCB       332       4     328     337         0        6      -6     337     -332        2     -334       0
62    WAYNE      DETROIT CITY 1018 AVCB       327       8     319     337         0        6      -6     337     -327       -2     -325       0
63    WAYNE        HIGHLAND PARK CITY 4       349       5     344     356         0        6      -6     355     -349        1     -350      -1
64    WAYNE        HIGHLAND PARK CITY 1       207       2     205     210         0        3      -3     209     -207        1     -208      -1
65    WAYNE       DETROIT CITY 994 AVCB       262       2     260     265         0        0       0     264     -262       -2     -260      -1

The fact that the regression lines all slant downward from left to right indicates that, on average, there were fewer votes for President than for the U.S. House or the State House in Michigan in the 2016 general election. I would very much seem that this should be investigated and not merely explained as "It happens" as in the Washington Post article. It would be useful to know how many of the undervotes for President are occurring in totally blank ballots (if those are, in fact, counted) or on ballots with only one or two races marked. More important, it would seem prudent to, if possible, examine them all manually. It's possible that some of the undervotes are very light marks that were not picked up by the optical scanners. In fact, I believe, that most optical scanners have a sensitivity setting. This would be especially concerning if the sensitivity could be set differently for different races. This would allow someone to set the Democrat Presidential candidate to be less sensitive than other candidates, resulting in more "missed" votes for that candidate. Of course, this could also be accomplished with malicious code that randomly ignored some small percentage of the votes for certain candidates. It would seem critical that undervotes be examined and not simply ascribed to the intent or a mistake by the voter.