Most good proposals still lose

We went through 563 evaluation results the European Commission has published for Horizon Europe and Erasmus+ topics since 2021, and pulled out one number that almost nobody looks at. Not the success rate — the share of proposals that cleared the quality threshold and were then actually funded.

The median, as at 6 August 2026, is 30%.

So in a typical call, roughly seven out of every ten proposals judged good enough to be funded were turned down anyway, because the budget ran out. Across those 563 results that adds up to 10,693 proposals that passed on quality and received nothing.

Why this changes what a rejection means

A proposal that scores above threshold and loses has not failed on quality. It has failed on ranking, and ranking is a different problem with different fixes. Rewriting the methodology section because the proposal "failed" treats a budget outcome as a quality verdict; the evaluation summary already said the quality was there.

Why this changes the arithmetic before you start

A raw success rate mixes together two things that behave nothing alike: proposals that were not good enough, and proposals that were good enough and still lost. In the results we checked, more than half of submitted proposals typically clear the threshold — so the real contest is not against the bar, it is against the other proposals that also cleared it.

And the spread is enormous. Across these 563 results, the funded share of above-threshold proposals runs from 3% to 100%. Two calls that look identical on their topic pages can be a lottery or a near-certainty. That gap is the subject of a separate analysis.

What a project team should do

Before committing six weeks to a proposal, the useful question is not "can we write something strong". It is "in this call, is strong enough". If you are choosing between two calls, look up what share of above-threshold proposals each one funded last time, and treat that as the real hurdle rate rather than the published success rate.