100% transfer of surplus in final count of Sligo Leitrim

[llgon]

This is related to the thread on the distribution of surpluses but I've put it in a new thread to avoid confusion with the discussion on the Kerry count.

I think it's strange that on the final count of the Sligo/Leitrim/West Cavan/South Donegal constituency there was no non-transferable vote out of 759.

The circumstances were as follows:
Count 12: Following the elimination of O'Rourke (FF) his 6083 votes were distributed with MacSharry (FF) getting 1707 leading to his election with a surplus of 759. As this surplus was not enough to elect a candidate or save another candidate from elimination the returning officer decided to put these votes aside for possible later use and eliminate the next lowest candidate.

Count 15: With three remaining candidates for the final two seats the only votes left to transfer were MacSharry's surplus of 759 (which all came from O'Rourke) and these were distributed as follows: Scanlon(FF) 637, Reynolds (FG) 101, McLoughlin(FG) 21. This resulted in Scanlon overtaking Reynolds and being elected along with McLoughlin. The margin in the end between Scanlon and Reynolds was 405 so no recount.

My understanding is that all the 1707 votes transferred in Count 12 should have been counted and proportioned into the 759 surplus. With no non-transfers in the 759 it is possible that there could have been one that was non-transferable out of the 1707 but not two. Seems unlikely to me.

 

[ajapale]

With no non-transfers in the 759 it is possible that there could have been one that was non-transferable out of the 1707 but not two. Seems unlikely to me.

https://docs.google.com/spreadsheets/d/116NZgZHL3NcaVRaFd4l-WPeU8OjM3mwka9A96bYS8gQ/htmlview

The link above should point to the excellent today fm Google docs count.

It does seem remarkable that on the 14th count mc starry had such a low proportion of non transferable votes that it registered at less than 1 / 759!

At this late stage you would expect the proportion of non transferable votes to increase as they will go stop at a candidate allready elected or eliminated.

Perhaps when it's down to a two horse race they just count the preference for the two remaining candidates and divvy up the surplus on that basis??

 

[Brendan Burgess]

Quota 12,468

This is what I thought would have happened.

On the 12th count, 1,707 papers were transferred to McSharry.
For count 15, these should have been put into sub-parcels - say:

Scanlon: 1,024 (60%)
Reynolds: 170(10%)
McLoughlin: 85(5%)
Non-transferrable: 428(25%)

So the surplus of 759 should have gone:

Scanlon: 455 (60%)
Reynolds: 76(10%)
McLoughlin: 38(5%)
Non-transferrable: 189(25%)

It seems wrong to me, but as it wouldn't have made any difference, they probably didn't bother worrying about it.

Under my system, Scanlon would have got 455 instead of 637, or 182 fewer
Reynolds would have got 25 fewer (101 -76)
So Scanon got relatively 157 more.

Take that away, and he still beat Reynolds by over 200.

 

[llgon]

Thanks for the replies. I totally agree that it would have had no effect on the outcome but am surprised if either there was only one/no transferable vote or the correct procedures were not followed because the effect on the outcome would not be decisive. However the latter would appear to be the case to me.