In these times of polarization and noise, when information and junk messages are mixed, debates arise where there should not be any. One case is the precision of the CIS. The matter returned to the news this week, when the Government confirmed José Félix Tezanos as head of the organization. The spokesperson for the Executive and Minister of Education, Pilar Alegría, justified it to the press: “You will understand that that sociologist who is right in the surveys is confirmed.”
Some experts said (rightly) that talking about failures and successes is not the best way to judge a measuring instrument under uncertainty. But it is more serious than that: the statement is false. We are not facing an opinionable issue, I dare say, but rather a factual fact: the CIS estimates with Tezanos have been imprecise and biased.
1. The CIS did not “get” the result of 23-J
Although it was more precise than in the past, its estimate of seats had as its central scenario a different one than the one that later occurred: the CIS gave PSOE and Sumar around 171 or 172 seats, which would have allowed them to govern without the support of Junts and ERC, but in reality they stayed at 152 and it was necessary to negotiate with those forces for months and months. Our survey average deviated in the opposite direction, but less: It gave about 177 seats to the sum of PP and Vox, which remained at 170 and without options to govern. Which survey came closest? The 40dB polling. that this newspaper published, which placed PP and Vox at 173 seats and the sum of PSOE and Sumar at 148.
2. The CIS estimates would still be in question even if it had been correct on 23-J
The reason is simple: the center has accumulated errors in multiple elections since Tezanos arrived at the center in 2018, so that one success was not enough to compensate for that history. To measure it, I have prepared the following table, which compares the deviation of half a dozen pollsters in a dozen electoral events since 2018.
The data is eloquent. The CIS with Tezanos has been wrong, on average, more than all the other pollsters. On four occasions she deviated more than any other – three times between 2018 and 2021, which was her worst period – and she always erred more than our polling average.
Calculation in three steps: (1) for each election and pollster, I calculate the average error per party (for those with 2% of the vote or more); (2) for each election, I calculate the average error of the pollsters; and (3) for each pollster and election, I now calculate its relative error, subtracting the average error in that election from the pollster’s error. Thus, a positive error means that the pollster deviated more than the average of pollsters, and a negative error means that she made less mistakes.
3. Most importantly: the worst problem with CIS estimates is not the magnitude of their errors, but their direction
Since Tezanos’ arrival, the CIS has overestimated the left’s vote in 36 of 37 elections. Also on 23-J. In July the left surprised with a better result than all the private surveys anticipated, but even in those circumstances the CIS once again overestimated it. He was surprised on the contrary: his estimate said that the PSOE and Sumar would prevail by 5 points over the PP and Vox, but in reality they were 1.4 points below.
![](https://datawrapper.dwcdn.net/o2kny/full.png)
Yesterday, an editorial in EL PAÍS criticized the ratification of José Félix Tezanos, arguing that his speech contaminates the image of impartiality of the center, they said, “regardless of the success or error of his forecasts.”
Going further: I think those errors are an indication of bias. Because, it is true that there are technical reasons to explain why the CIS polls deviated to the left, once and several times, but when that deviation becomes systematic, and is repeated after four years and 30 elections, the question is inevitable. : why is such bias not corrected?
In other news
⚽ Rodrygo returns to the average. I was interested in these data from David Álvarez about the Real Madrid forward. The graph shows the evolution of Rodrygo’s success this season, comparing his accumulated goals (dark blue) with his expected goals, according to the number and characteristics of his shots (light blue).
![](https://datawrapper.dwcdn.net/C1rMg/full.png)
The graph says that Rodrygo’s chance production has been more or less constant. The player appeared in shooting positions with a regular frequency, worth four expected goals in his first 10 games and another four in the next 10. What changed was not that, but his accuracy. Rodrygo managed to “owe” more than three goals, back in game 11, when he had only one goal with shots worth four. But in the last five games his success has been extreme—seven goals—to compensate for that debt and put the statistics in his favor.
There are many ways to explain this back and forth. Part of it will be simple chance: an exceptionally bad streak must be followed by a normal, therefore better streak. But it’s also reasonable to think about the mind of a player. In how failing or scoring conditions your game, and in the advantage of achieving calm, that automatic mode that José Adán, responsible for a neurotraining unit, refers to: “a state in which everything flows. The brain does what it has to do at all times, without obstacles, without rumination of thoughts, worries or expectations.”
🔥 2023 is the hottest year in millennia. Why is this and how do we know? Manuel Planelles, Clemente Álvarez and Laura Navarro tell it in this visual analysis of the evolution of the climate, where they follow the human footprint and the traces left by temperatures in the trees and glacial ice.
![](https://datawrapper.dwcdn.net/g63pF/full.png)
✏️ Has AI already taken away some jobs? A recent study looked at advertisements on a job website freelance for editors and writers. The month after ChatGPT appeared, a year ago, a drop in offers was observed: fewer writers were sought and they were paid less. He told it John Burn-Murdoch in Financial Times.
You help me? Forward this newsletter to whoever you want, and if you are not subscribed, sign up here. It is an exclusive newsletter for EL PAÍS subscribers, but anyone can receive it for a trial month. You can also follow me on Twitter, at @kikollanor write to me with clues and comments, to [email protected].
Subscribe to continue reading
Read without limits
_
#Tezanos #doesnt #Fails #worst #direction