2021 Dublin Bay South Bye-Election

17-08-2026 by Eoghan Kelly

Back-testing against a real bye-election

Dublin Bay South’s 2021 bye-election — called after Eoghan Murphy’s resignation and won comfortably by Labour’s Ivana Bacik. This is a further test our transfer model. The trasnfer system is designed for the present day, so the further back in time we go the less accurate it is likely to become.

Our Projection

Running Bacik and Geoghegan’s first preferences through our baseline transfer model has Bacik winning with 54.1% to Geoghegan’s 34.9% once everyone else is eliminated.

Bacik led from the first count and extended that lead steadily, picking up the bulk of transfers from Sinn Féin, the Greens, and the smaller left-wing candidates as they dropped out.

Results

First preferencesCount 1Count 2Count 3Count 4Count 5Count 6Count 7Count 8Count 9Count 10Quota (50.0%)Bacik (Lab)30.2%Geoghegan (FG)26.2%Boylan (SF)15.8%Byrne (Green)8.0%Conroy (FF)4.6%Flynn (Ind Centre)3.3%Durcan (SD)3.2%Purcell (PBP)2.8%Tóibín (Aontú)2.8%Dooley (Ind Left)1.0%Barrett (Ind Right/NP)0.7%Cahill (Ind Right)0.6%Gilbourne (Ind Right/Renúa)0.6%Keigher (Ind Other)0.1%O'Keeffe (Ind Other)0.1%

Real Life

Bacik (Labour) defeated Fine Gael’s Geoghegan by 49.8% to 34.4% after all votes were counted. Sinn Féin’s Boylan and the Green’s Byrne were the last two eliminated before the final count, same as in our projection.

Bacik led from the first count here too, and built her winning margin gradually as the field narrowed, rather than in one late surge.

Results

First preferencesCount 1Count 2Count 3Count 4Count 5Count 6Count 7Count 8Count 9Count 10Quota (50.0%)Bacik (Lab)30.2%Geoghegan (FG)26.2%Boylan (SF)15.8%Byrne (Green)8.0%Conroy (FF)4.6%Flynn (Ind)3.3%Durcan (SD)3.2%Purcell (PBP)2.8%Tóibín (Aontú)2.8%Dooley (Ind)1.0%Barrett (NP)0.7%Cahill (Ind)0.6%Gilbourne (Renua)0.6%Keigher (Ind)0.1%O'Keeffe (Ind)0.1%

Differences

The model got the shape of the race right — Bacik led throughout and won comfortably, with Boylan (Sinn Féin) and Byrne (Green) as the last two eliminated before the final count in both scenarios. The final margin is where it drifts: we had Bacik winning by 19.2 points, real life had it at 15.4.

With an updated set of baseline transfer rates, the model now also gets the order of the last few eliminations right — PBP’s Purcell goes before Aontú’s Tóibín in both our projection and reality, whereas an earlier version of the model had that pair flipped. With margins this tight between minor candidates, small differences in transfer assumptions are still enough to flip who’s eliminated in which order, even when it doesn’t change the final outcome.

We also ran one more count than the real result did. As with our Dublin Central back-test, the real count bundles minor and clearly-hopeless candidates into single eliminations where we process every candidate one at a time — it’s an unnecessary extra step for us since we’re not trying to speed up a live count, just get the transfer model right.

Reading the charts

Use the play button on each chart to step through the rounds, or use the round selector to jump directly to a specific count. The bar shading shows which candidates are still in contention; eliminated candidates drop to zero before their preferences flow to the remaining candidates.

These are illustrative scenarios, not predictions — first-preference shares and transfer patterns will both depend on who actually stands and how the campaign plays out.

💜 We’ve brewed up this site with a lot of love (and caffeine). 💜

Building this website took a lot of late nights and a genuine passion for making election projections that are accessible.

If you’ve enjoyed using the site or found it helpful, consider supporting us with a coffee.

It’s a small gesture, but it means the world to us and helps keep projects like this going.