Brazil Election - 5 October 2014

[buritobr]

I will write the numbers of my predictions only tommorow evening, after the last polls. But I think that Aécio Neves will go to the second round. He performed much better than Dilma Roussef and Marina Silva.

Concerning the state elections, I think that Padilha and Lindberg will have a much better result than they polling now. Dilma has 35% in Rio de Janeiro. Many Dilma voters will vote for Pezão, Garotinho and Crivella, but I think that many Dilma voters who have not decided yet the vote for governor will decide to vote for the candidate of the same party. Considering that Dilma has 25% in São Paulo and Padilha has only 11%, I think that also in São Paulo there is room for Padilha to grow in the eve of the election.
Many people decide for whom they will vote for president many months before the election, and decide for whom they will vote for governor only few days before. In São Paulo in 1998, 2002, 2006 the polls were predicting that the PT candidate for governor would have a much worse result than Lula, and than, the PT candidate for governor used to finish only a little bit worse than Lula.

[buritobr]

I learned in this forum that there is a big coincidence. Bulgaria will also hold elections tomorrow. Bulgaria is the home of the Rousseff family. So, in this year, the bulgarians are not watching Brazilian elections, as they were in 2010.

[buritobr]

My prediction is not based on the national Ibope poll, but on the 27 state Ibope polls, weighted according to the population of each state. I considered that all the "undecided" voters will vote for Aécio Neves, because he is on a growing trend. The prediction is

Dilma Rousseff 43.6%
Aécio Neves 28.3%
Marina Silva 23.9%
Others 4.2%

In the "others", probably Luciana Genro will win and Eduardo Jorge, Pastor Everaldo and Levy Fidelix will be tied.

I showed the table of the calculation in my blog
http://blogdomarcelobrito.blogspot.com.br/2014/10/projecao-do-resultado-do-primeiro-turno.html

[buritobr]

Details of the Datafolha poll October 4th

All: Dilma 40%, Aécio 24%, Marina 22%

Age
16-24: Dilma 36%, Aécio 24%, Marina 25%
25-34: Dilma 38%, Aécio 22%, Marina 26%
35-44: Dilma 43%, Aécio 22%, Marina 21%
45-59: Dilma 42%, Aécio 25%, Marina 20%
60-: Dilma 42%, Aécio 27%, Marina 19%

Grade of school
Elementary: Dilma 50%, Aécio 19%, Marina 18%
High School: Dilma 38%, Aécio 24%, Marina 24%
College: Dilma 26%, Aécio 34%, Marina 26%

Household Income
Less than 2 minimum wages: Dilma 49%, Aécio 17%, Marina 20%
Between 2 and 5 minimum wages: Dilma 37%, Aécio 26%, Marina 24%
Between 5 and 10 minimum wages: Dilma 28%, Aécio 33%, Marina 25%
More than 10 minimum wages: Dilma 24%, Aécio 41%, Marina 25%
One minimum wage in Brazil ~ US$300 monthly

Region
Southeast: Dilma 32%, Aécio 27%, Marina 25%
South: Dilma 38%, Aécio 32%, Marina 15%
Northeast: Dilma 55%, Aécio 13%, Marina 21%
Center-West: Dilma 33%, Aécio 32%, Marina 23%
North: Dilma 47%, Aécio 22%, Marina 20%

[buritobr]

Wait, what ?

Last time I checked (which was a month or 2 ago) Silva was easily making the run-off and was even tied with Rousseff.

And now she's even struggling to come in 2nd ?

What the hell happened ?

Very bad campaign. She tried to show to the anti-Dilma voters that she would have more chances to win against Dilma in the runoff, but she never convinced these voters why would she be a better presidente than Aécio Neves.

[buritobr]

The defeat of PT in the Congress is the result of the complete colapse of the PT in the middle class.
Since 2006, the PT presidential candidate have much more votes from the poor than from the middle class. However, almost every middle class person who votes for Lula/Dilma votes also to a PT candidate for the Congress. Poor people usually vote for Lula/Dilma but not necessarily to PT for the Congress.
In 2014, Dilma had large defeat even in cities with large middle class population in which Lula and Dilma still did well in 2006 and 2010, like Brasília, Rio de Janeiro and Campinas. PT lost a large number of middle class voters after the scandal of 2005, but still performed not so bad in the middle class in 2006 and 2010. In 2014, PT almost lost all middle class voters

[buritobr]

This Brazilian election looks like recent European elections. Right does well, far left does well, center-left suffers a large defeat.

In Rio de Janeiro, PSOL became the biggest left-wing power, bigger than PT. Lindberg (PT) had a little bit more votes than Tarcísio Motta (PSOL) only because of his hometown Nova Iguaçu. PSOL elected more state deputies than PT.

[buritobr]

I think most of the Marina Silva votes will go to Aécio Neves in the runoff. One have to consider that although she was attacked by Aécio campaign, she was much more attacked by Dilma campaign. It is possible that Marina endorse Aécio in the runoff.

However, I don't believe that 100% of the votes for Marina Silva will go to Aécio. I observed the city-by-city results and discovered that unlike 2010, Marina Silva had more votes in low income cities than in high income cities.
When I observed the metropolitan area of Rio de Janeiro, I saw that Marina Silva had more votes in São Gonçalo, Duque de Caxias, Nova Iguaçu, Belford Roxo and São João de Meriti than she had in Niterói. When I observed the metropolitan area of São Paulo, I saw that Marina Silva had more votes in Diadema and Guarulhos than she had in São Paulo. When I observed the metropolitan area of Campinas, I observed that she had more votes in Sumaré and Hortolândia than she had in Valinhos and Vinhedo. I don't think that all the votes for Marina in these cities where the population is poor will go to Aécio.

[buritobr]

Considering group of candidates

Left (Dilma Rousseff + Luciana Genro + Eduardo Jorge + Zé Maria + Mauro Iasi + Rui Pimenta)
45,661,331 votes (43.1%)

Right (Aécio Neves + Marina Silva + Pastor Everaldo + Levy Fidelix + Eymael)
58,362,471 votes (56.1%)

Exactly the opposite of Dilma victory in 2010.


However, despite being "right", as I said before, I don't believe that 100% of Marina votes will go to Aécio.
The map below shows how the districts of the city of Rio de Janeiro voted. In the city of Rio de Janeiro, the three major candidates almost tied (~30% for each one). Aécio won all the high income districts. Dilma won some low income districts and Marina won some of them. In the poor districts Marina won, Dilma won by a huge margin in the runoff against Serra in 2010. I don't think that all these poor people who voted for Marina will vote for Aécio now.
http://g1.globo.com/politica/eleicoes/2014/rj/apuracao-zona-eleitoral-presidencia/1-turno.html
In the first round of 2010, the map was different. Dilma had the pluraity in the city, Dilma won all the poor districts and Marina and Serra shared the rich districts. The vote for Marina in 2014 became poorer.

http://g1.globo.com/politica/eleicoes/2014/sp/apuracao-zona-eleitoral-presidencia/1-turno.html
In São Paulo, Marina won no district. But she had more votes in the red districts than in the blue districts. In Grajaú, a poor district located in the South of the city, Dilma had 45,7%, Marina had 24,2%, Aécio had 23,8%, Luciana had 3,3%, Levy Fidelix had 1,2% and Everaldo had 0,8%. I don't think it is possible for Aécio to have 50% (Aécio + Marina + Levy + Everaldo) in Grajaú in the runoff. In 2010, Dilma had 71% against José Serra there. In the election for mayor in 2012, Haddad (PT) had 80% against José Serra there.

[buritobr]

Two questions:

1. Why is PSDB so freaking strong in SP? (and PT so weak?)
2. While Aecio made a remarkable comeback, he just lost his home state of Minas, where he was suppoused to be a God. Would that make any effect or it doesn't matter? 

1. In São Paulo, the percentage of people living below the poverty line is much lower than in the rest of the country. Unlike Rio de Janeiro and Brasília, the large majority of the middle class from São Paulo work in the private sector. That's why they are more willing to support "small government". In Rio de Janeiro and Brasília, most of the middle class work in the public sector, and that's why, many of them used to vote for PT. But this year, Dilma didn't win Brasília and Rio de Janeiro too.
Despite having huge rejection in São Paulo, PT used to have safe 30% there. In this year, it didn't happen. São Paulo is the home of PT and PSDB.

2. Minas Gerais is a sample that represents Brazil as a whole. The north of the state is poor like the neighboring Bahia, the south of the state is rich like the neighboring São Paulo. Aécio won in the south of the state, but lost in the north

[buritobr]

In 1994, the ratio GDP per capita from São Paulo / GDP per capita from Brazil was 1.60. In 2011, this ratio was 1.51. Actually, São Paulo has lower GDP growth rates than the Brazilian average since mid 1970s.

Even before being pro-PSDB, São Paulo was anti-left. In 1989, Collor (from Alagoas) defeated Lula (from São Paulo) by 53-47 in the whole country and by 58-42 in São Paulo. Lula was supported by the PSDB candidate Mário Covas.

[buritobr]

Marcelo Crivella was endorsed by Anthony Garotinho today. Since Garotinho has a strong disapproval rating, this move was not seen as good way to start. Nevertheless, Crivella has shown some low rejection ratings and seems to be a difficult challenge for Pezão in Rio.A Datafolha Poll indicated Pezão beating Crivella by only two points (51% vs 49%).

Pezão could win easily against Garotinho or Lindberg. Crivella is the most difficult chalenger. I think I will vote for Pezão, but if he endorses Aécio officially, I would not vote for anyone.

[buritobr]

Historical vote of the "nanicos"

José Maria Eymael (PSDC) far right
1998: 171,831 (0.25%) 9th
2006: 63,294 (0.07%) 6th
2010: 89,350 (0.09%) 5th
2014: 61,250 (0.06%) 9th

José Maria de Almeida (PSTU) far left
1998: 202,659 (0.30%) 7th
2002: 402,236 (0.47%) 5th
2010: 84,609 (0.08%) 6th
2014: 91,209 (0.09%) 7th

Rui Costa Pimenta (PCO) far left
2002: 38,619 (0.05%) 6th the last one
2010: 12,206 (0.01%) 9th the last one
2014: 12,324 (0.01%) 11th the last one

Levy Fidelix (PRTB) far right
2010: 57,960 (0.06%) 7th
2014: 446,878 (0.43%) 7th

Fidelix increased eight times his vote after saying that people cannot reproduce through the anus

[buritobr]

Wich is the most vargasist party ? PTB, PDT, PTN or PTdoB ?

PDT

Although PTB was refounded by Vargas' daughter in 1979, the politicians who joined PTB after that had nothing to do with Getúlio Vargas labor policies.

The PDT, founded by Leonel Brizola in 1979 too, is the party who supports Vargas' ideas. Brizola was a political ally from Vargas in Rio Grande do Sul.

[buritobr]

I voted for Eduardo Jorge in the first round. Eduardo Jorge is endorsing Aécio Neves now...

But I will vote for Dilma Rousseff.



It is not surprise this endorsement. The Green Party (PV) has already been partner in PSDB administrations.

[buritobr]

In 1989, 1994 and 1998, nobody knew Reinaldo Azevedo. Lula lost three times.

In 2002, 2006 and 2010, Reinaldo Azevedo was one of the biggest anti-PT voice. Lula won in 2002 and 2006, and Dilma won in 2010.

Until now, he was not very sucessful.


He is a Brazilian version of American conservative pundits like Ann Coulter, Rush Limbaugh, Billy O Reily and Sam Hannity.

[buritobr]

I wrote a blog post about pro-PT and pro-PSDB journalists

http://blogdomarcelobrito.blogspot.com.br/2014/10/as-maquinas-de-spining-do-pt-e-do-psdb.html

[buritobr]

After the cover of three weekly magazines (Saturday) and after the endorsement of Marina Silva (Sunday), Aécio Neves did not wide the gap.

This Datafolha poll was done in two days: October 14th and October 15th. The first day of interviews took place before the debate. The second day of interviews took place after the debate. So, we can not evaluate the impact of the debate yet.

Even the columnists of the conservative newspaper O Globo considered that Dilma Rousseff performed better in the debate.

[buritobr]

The Economist endorsed Aécio Neves. No surprises.

In the USA, The Economist has already endorsed republican and democratic candidates. In the UK, The Economist has already endorsed conservative and labour candidates. In Italy, The Economist has already been against Berlusconi. Outside these three countries, this newspaper had always endorsed the candidates of the Right.

[buritobr]

State polls Ibope October 16

Rio Grande do Sul
Aécio 51% Dilma 49%

Paraíba
Dilma 60% Aécio 40%

Pará
Dilma 56% Aécio 44%

Rio de Janeiro
Dilma 53% Aécio 47%

Ceará
Dilma 78% Aécio 22%


Results of the first round

Rio Grande do Sul
Dilma 43,2%  Aécio 41,4%  Marina 11,5%  Luciana 2,3%

Paraíba
Dilma 55,6%  Aécio 23,4%  Marina 18,8%  Luciana 0,8%

Pará
Dilma 53,2%  Aécio 27,6%  Marina 16,3%  Everaldo 1,1%  Luciana 1,1%

Rio de Janeiro
Dilma 35,6%  Marina 31,1%  Aécio 26,9%  Luciana 2,7%

Ceará
Dilma 68,3%  Aécio 15,0%  Marina 14,1%  Luciana 1,2%


In some states, Dilma is not having much more than Dilma + Luciana votes. A large majority of Marina votes are going to Aécio. In Rio de Janeiro (the third most populous state), Dilma still has the minority of Marina votes, but has a significant share of these votes.

[buritobr]

The Economist is not as conservative as Veja and Época. The british magazine has a more positive view on Lula than the Brazilian magazines have.

[buritobr]

If we ask a monkey who will be the winner, the probability of the monkey say the right answer is 50%

[buritobr]

Sensus does not work for PSDB. Sensus works for Aécio Neves.

In 2010, this institute had pro-Dilma biased polls.

[buritobr]

Also, the PSDB is usually underpolled nationally.

That may be true for the 1st round, but not for the 2nd.

Sure. Datafolha predicted 55-45 for Dilma in the 2nd round in 2010. It was 56-44.

[buritobr]

Exit polls are not necessary anymore. Before the electronic vote, the time needed to process the results were five days. Nowadays, it is necessary only four hours.