Spain's July 23 Election: Why Senate and Congress Votes Differ

Senate and Congress on July 23: 600,000 null votes and mismatched tallies

A voter takes the ballot for the Congress of Deputies, places it in their envelope, then encounters the Senate ballot: orange paper, a list of names, and three checkboxes. They mark an X for the PP and, for whatever reason, another for a different party. Counting these at midnight, after fourteen hours of voting, produces figures that alarm people. The Senate and the Congress rarely align. This is normal, not an exception.

On July 23, this discrepancy became a trending argument. A senior police official reportedly warned Javier Cárdenas of an anomalous difference between votes for the Congress and the Senate, a gap claimed to be unprecedented in magnitude and summarized as: "people are not crazy enough to vote for the PP in the Senate and the PSOE in the Congress." The tweet spread rapidly. The technical explanation was far less publicized.

Why do the Senate and Congress yield different results?

Because they are two entirely different systems, despite being voted on the same day at the same polling station. In the Congress, voters choose a closed list by province with proportional allocation. In the Senate, a fixed number per constituency is elected —four senators per province— and each party can propose only three names. The allocation is majoritarian, heavily rewarding the winner in each province beyond what their vote share would suggest in other calculations.

The second effect is split-ticket voting. On the orange ballot, individuals are marked, not party logos, up to three boxes. This encourages mixing. A voter might support the PSOE in the Congress but tick a specific PP name in the Senate without hesitation. Some argue that constituency allocation, especially in low-allocation districts, explains the entire difference alone. No conspiracy is needed: just look at how the two chambers are counted.

The police warning turned into a tweet: what can be proven?

Regarding the alleged warning from that senior police official, all we know is what the tweet says: that an anomaly exists and had never been seen before. No report, no document, no name. Electoral fraud claims without evidence are a recurring hypothesis whenever counts surprise, and this time the surprising element was the Senate's own allocation system.

It is crucial to separate two things. That results differ between both chambers is mathematically expected, as explained above. That this difference proves rigging is a completely different matter, unsupported by any evidence. Raising suspicion is not proof, especially when the explanation is standard in any electoral law manual.

What are the 600,000 null votes in these elections?

These are ballots invalidated because voters spoiled them: marking more names than allowed, ticking two parties, or incorrectly filling out the envelope. The figure cited, nearly 600,000 null votes, stands out due to its size, with a large portion concentrated where confusion peaks: the Senate.

Here lies the detail few mention. In 2015, a poll worker counted Senate ballots with an X for the PP and another for an independentist party, along with many voided votes because voters had no idea what to do. This is not fraud; it is ignorance. Null votes forgive everyone and punish no one.

Minutes, envelopes, and Indra: what can be verified and what cannot

This is where genuine distrust begins. The count does not happen in a basement: it occurs at each polling station, publicly, with representatives from all parties present. The minutes are filled out by hand and delivered immediately, using carbon copy paper: each party keeps a copy. With thousands of monitored tables, coordinated tampering would require absolute silence from everyone.

The other front is postal voting. It is claimed that these are counted last but reported first, causing initial data fluctuations. As one forum user explained, the actual mechanism is simpler: the postal envelope goes into the box and is counted with all others, so a table's result reflects the whole table, not the last batch. Then there is the fate of the papers: according to online discussions, votes are destroyed except for null ones or those subject to dispute. From there to concluding that Indra determines the outcome is a leap no one has made with evidence.

The Senate almost no one votes for, yet appoints Constitutional Court judges

The chamber generating the anomaly is also the one people care least about, creating a paradox. Some skip voting for the Senate believing it serves no purpose. It does: it elects members of the Constitutional Court and the CGPJ, among other less visible functions.

Added to this is a logistical detail with consequences. A forum user noted that several parties sent home the white Congress ballots but not the orange Senate ones. Result: those wanting to vote but lacking the ballot wrote candidate names in pen. If that doesn't generate null votes and splits, nothing will.

---

With the Senate system in mind, the interesting question is not why votes don't match, but why anyone expected them to. The gap between the chambers doesn't appear because some moved envelopes: it appears because voting differs, counting differs, and fundamentally, very few people know what they are voting for when they pick up the orange ballot.