Plenty to see in the latest #s. Clark County remains a really tough nut to crack for the GOP but there is plenty of good news once we dig into the numbers. On Day 4 of early voting Democrats cast 15,310 ballots while Republicans cast 9911 ballots. This was a 2% day-over-day (dod) increase for the Democrats and a -3% dod decline for the Republicans. Basically, Democrats held serve on Day 4 marginally reversing the compression between the two parties proportional representation of the overall early vote total. Despite the one-day flattening of the trend, Republicans remain far motivated and active in early voting nearly doubling their 2008 output while Democrats are failing to even meet their 2008 levels beyond the first day pop.
Election over election comparison at the same point in 2008:
- Democrats overall lead today is 23,490
- Democrats lead in 2008 was 32,380
- Higher mathematics says that is an 8890 reduction
Comparison versus same party performance:
- Versus 2008 Democrats have 978 fewer ballots cast overall
- Versus 2008 Republicans have 7912 more ballots cast overall
- Democrats are averaging 244.5 fewer ballots cast every day
- Republicans are averaging 1978 more ballots cast every day
- At those rates, Democrats early vote advantage would shrink from 86.5k in 2008 to 55.5k by election day
The Clark County Model
We’ve been using our crude model created after just one days #s which admittedly left a lot to be desired in predictive value. But now after four days we have enough data to make some tweaks to model assumptions. As the #s above show, 2012 has been a repeat of 2008 which is what we used as the basis for our original assumptions. 2008 saw greater day to day volatility while 2012 has seen comparatively stable daily results. Possibly THE question to answer to form a predictive model at this point is whether Republicans continue their historical trend of daily improvement versus Democrats or whether the comparative percentages flatten out. Although 2008 saw Republicans steadily gain on Democrats, admittedly not without a set-back or two, the last three days in Clark County have been remarkably consistent. Democrats % of the overall vote has stayed within a 1.1% range (50.1 to 49.0) while Republicans % has stayed within a 1.4% range (33.4 to 32.0). Interestingly, if you refer back to the 2008 chart, on Day 4 the Democrats achieved the same reversal of the GOP rising tide before the trend went back to compressing the proportional make up between the two parties. As such for our model we will maintain an expected growth in the GOP’s percent of the vote relative to the Democrat’s performance.
- In the 8-days following the Day 1 pop in 2008, Democrats averaged a turnout relative to the Day 1 total of 88.8% before steadily ramping up the turnout into election day. In 2012 Democrats’ post-Day 1 average is only 78.6% of the Day 1 turnout at 14,446 ballots cast. Other than a one day spike in 2008, Democrats turnout was fairly stable for the remaining days until the ramp-up into election day.
- We are going to adjust our input for Democrat turnout to their current 3-day average up to the ramp-up point (Oct 29) and grow the turnout at 2008’s rate.
- The 2008 rate is calculated using the actual turnout relative to the Day 1 total.
- The rate of change for Republican growth relative to Democrat turnout in 2012 has been 4.2%. Our earlier model set the growth rate at 2.4%. We are going to insert the new 4.2% growth rate into the model building off the Day 4 result which was Republicans at 65% of the Democrats turnout.
Under these new assumptions using 2012 actual data and overlaying them on the 2008 results, our model expects the Democrats advantage in Clark County heading into election day to be only 40,820 — a -51% decline from 2008 (83,633) or an advantage of -42,813 fewer ballots cast.