The Long View
Opportunities hidden in the numbers
By Manuel L. Quezon III
Philippine Daily Inquirer
First Posted 02:06:00 02/08/2010
YESTERDAY this paper reported there are 50, 723, 734 voters registered for the May 10 elections. These voters will be organized into 329, 389 voting precincts organized further into 75,471 clustered precincts in 37,226 voting centers.
The Comelec expects a relatively high voter turnout – 80 percent or 40.5 million voters. If we use the income classes advertisers and pollsters use, this translates to: 4,057,898 voters from Classes ABC (10 percent of voters); 30, 434, 240 from Class D (75 percent); 6, 086, 848 from Class E (15 percent).
What could this mean in actual votes?
Without fussing with margins of error, the January surveys put Aquino anywhere from 36 percent (Standard), 37 percent (Pulse), 42 percent (SWS); Villar – 35 percent (SWS, Pulse), 36 percent (Standard); Estrada – 2 percent (Pulse), 13 percent (Standard, SWS); Teodoro – 4 percent (SWS), 5 percent (Standard, Pulse); Villanueva – 2 (all); Gordon – 1 (Pulse, Standard), 2 (SWS); Madrigal – 0.4 (SWS), 0.5 (Pulse), 1 (Standard); De los Reyes – 0.2 (SWS) 0.3 (Pulse), 05 (Standard); Perlas – 0.05 (Pulse) 0.1 (SWS), 0.3 (Standard).
Pulse has the freshest numbers. If honest elections had been held on Jan. 22-26, the results could have been: Aquino – 15,014,225 votes; Villar – 14,202,645; Estrada – 4,869,478; Teodoro – 2,028,949; Villanueva – 811,579; Gordon – 405,789; Madrigal – 202,894; De los Reyes – 121,736; Perlas – 20,289; and nuisance candidate Acosta of the KBL – 81,157.
Put another way, had the election taken place on the survey dates, over a million votes would have separated frontrunner Aquino from his leading contender Villar.
SWS (Jan. 21-24) has Aquino at 42 percent and Villar at 35 percent, a seven-point difference that translates into a lead of 2,840,529 votes. If the difference is actually 5 percent, the lower end of the margin of error, that’s an Aquino lead of 2,028,949; if it’s 9 percent, the maximum end of the margin of error, that’s an Aquino lead of 3,652,108.
The most recent Pulse Asia poll (Jan. 22-26) puts Aquino ahead of Villar by 2 percent. (With a different base and methods, the Manila Standard Today survey has the same number.) Much has been made of this statistical dead heat, though it actually means – factoring in the margin of error – that the two leading candidates are anywhere from being tied to 2-4 percent apart. So 2 percent of the expected turnout is a 811,579-vote lead for Aquino or the two could be exactly tied; at four percent, Aquino’s lead could be as high as 1,623,159.
In terms of socioeconomic classes, Pulse has Aquino leading Villar (37-22) among ABC; that’s 15 percent of that class or a 608,684-lead in votes. Aquino also leads Villar in Class D (40-34), so that’s a 6-percent lead, meaning, 1,826,054 votes more than Villar; while Villar leads Aquino in Class E (39-31) which means a 9-point lead for Villar or 486,947 votes over Aquino.
This all presumes that if 80 percent do vote, all their votes will be counted, not only properly but also expeditiously.
The problem, of course – in the immortal words of President Macapagal-Arroyo, who moved heaven, earth and Garci to ensure she had a lead of 1 million votes in 2004 – is, ” yung dagdag, yung dagdag.” Plus, as her subordinates might put it, “yung bawas, yung bawas.” There are many ways to do this without even padding or shaving votes once cast.
In the mock polls (a simulation of the voting, counting of votes and transmission of results) over the weekend (involving 50 voters per precinct in a total of nine precincts in Quezon City, Taguig, Baguio, Cebu and Davao), the Comelec proclaimed there was no problem in the counting and transmission.
Overlooked was, considering the tiny numbers participating, the relatively high number of votes that weren’t accepted for counting. In Quezon City, only 46 out of 50 ballots were counted. (Four were rejected, apparently for “improper shading.“) In Taguig, the machine also refused to accept three ballots. Four out of 50 is 8 percent; so let us assume this is a reasonable number of spoiled/invalid ballots to expect from people not following directions. That’s 3,246,318 votes out of the immediate counting and canvassing.
The Comelec earlier put much higher the percentage of voters who might have problems because of the machines – 30 percent, which means it’s preparing for manual counting for 6,086,848 votes.
In either case, if human error alone might put 3.2 million-6 million votes in a grey area (thus requiring further scrutiny and manual counting, with each ballot bogged down in examinations and arguments), then neither of the two leading contenders in a close race could be proclaimed.
As the formal campaign begins tomorrow, Aquino’s lead, while still formidable, gives (a false, I think) impression of being smaller than it actually is when translated into percentages. Villar’s catching up in terms of percentages still has a long way to go when seen in terms of actual votes. On the other hand, neither side can rest easily because neither has a comfortable enough percentage to make them immune to the administration’s ace in the hole.
And what’s the administration’s ace in the hole? It can deny either of the leading contenders victory – unless. Which may be why Gary Olivar has thanked Villar for not hitting GMA; why Press Secretary Jun Icban said that thereâ€™s no GMA “kiss of death” despite talk she’s in league with Villar. The enemy of my enemy, the Palace broadly hints, will be my friend.
Here is Social Weather Stations’ Jan. 21-24 report in full; here is Pulse Asia’s Jan. 22-26 report. Mon Casiple has an analysis, while Marocharim Experiment has a generation-specific reflection. For background reading, see PCIJ’s report, Jeckyll-and-Hyde Campaign;Â John Nery’s The 2010 race is set and my entry,We, the People: How Candidates view The People as Electors, my columns Brains without bodies (2) and Back to the Future, and Pulse Asia’s February 2009 Survey on the May 2010 Elections: The Undecided in Alphanumeric.