2020 Election Analysis – Part II – How the cookie crumbled…

As a follow up to my initial analysis, here’s some additional interesting statistics from the 2020 election.

In my last post I was focused primarily on means as the statistic highlighted. This showed in particular that:

  1. The PLP vote declined on average by 2.4% and the OBA vote by 30.5%.
  2. If I remove the constituencies where the OBA didn’t have a candidate this election, their average voter decline was actually 26%.

However, if one wants to get an idea of the total voter decline between 2017 and 2020 for each party, as well as the increase number of non-voters (and for this one must exclude the three uncontested seats), one finds this:

  1. The PLP vote declined by -7%.
  2. The OBA vote declined by -45%.
  3. The non-voter vote increased by +27%.

You will note I have not modeled the FDM into this. This is solely due to the 2020 election providing the baseline for them, so I couldn’t really model them compared to 2017. The best comparison would be with the category ‘Independent’ from 2017, in which case the vote increased here by +5,159%, which to me isn’t too useful a metric.

What to me is more interesting for the FDM vote here is seeing what happened to the declining PLP and OBA vote in relation to the non-vote and the FDM vote here.

The OBA saw a total loss of 4,990 voters, while the PLP saw a total loss of 1,980 voters. The total voter decline then was 6,970, of which 72% was OBA and 28% was PLP.

On the other hand, the increase in voting categories was that of FDM+Independents (1,451 or 21%) and non-voters (5,340 or 79%).

Now, it isn’t possible for me to say for sure how the breakdown between PLP/OBA voter decline went between FDM/Independents and non-voters. If I assume that the voter decline of the PLP/OBA broke according to the 21% and 79% aspect and in proportion to their respective declines (28% and 72%), one can start formulating some hypotheses.

However, as the FDM didn’t run candidates in every seat, I feel the analysis here should be restricted solely to the seats with FDM candidates. This, of course, leads to further complications in that in two of those seats there was no OBA candidate. I don’t think the OBA understands how inconsiderate they were in not fielding 36 candidates, both for the voters and statisticians… lol

Reducing the pool further to just FDM challenged seats to see how the vote split between PLP/OBA declines and FDM/Independents/non-voter increases, gives us the following bits of data:

  1. In these seats the PLP vote declined by -22% overall (but -1.6% on average).
  2. The OBA vote declined by -217% overall (but -15.5% on average).
  3. The non-voter category increased by +63% (but +4.5% on average).

To get a better idea of how the vote broke down one should remove the two seats without OBA candidates. This then changes the above numbers to:

  1. The PLP vote declined by -25% overall (but -2% on average).
  2. The OBA vote declined by -189% overall (but -15.75% on average).
  3. The non-voter category increased by +76% (but +6.3% on average).

So how did this vote decline split between PLP and OBA in these seats? (just ones with a 3-way fight):

  1. The total loss of PLP/OBA votes was 2,958.
  2. The PLP loss here was 1,042, or 35%.
  3. The OBA loss here was 1,916, or 65%.
  4. The increase in FDM/Independents/Non-voters was 2,911.
  5. The FDM/Independent was 1,092, or 38%.
  6. The Non-voter was 1,819, or 62%.

If I assume the PLP/OBA vote decline went 62% to non-voters and 38% to FDM/Independents one may hypothesize that:

  1. 62% of the PLP lost voters went to non-voters = 646 voters.
  2. 38% of the PLP lost voters went to FDM/Independent = 396 voters.
  3. 62% of the OBA lost voters went to non-voters = 1,188 voters.
  4. 38% of the OBA lost voters went to FDM/Independent = 610 voters.

What does this mean for the composition of the FDM vote?

  1. About 5.7% of 2017 PLP voters found the FDM attractive.
  2. About 14.3% of 2017 OBA voters found the FDM attractive.
  3. It isn’t possible to gauge how many of 2017 non-voters found the FDM attractive here.

By all means, please double-check my calculations. I do not at all claim to be an expert at statistics and it has been many years since I dabbled at it.

My basic conclusions though are:

  1. The FDM can peel off some of the PLP voting base, however the core growth area for the FDM would be from disaffected OBA voters.
  2. That the FDM couldn’t attract more disaffected OBA voters is interesting. It is possible that the candidate selection and main personalities behind them are not palatable for the OBA voting base. After all, Mr. Bean as a former PLP leader was the subject of demonisation by the OBA at the time, so perhaps he is both the FDM’s biggest asset and liability in this context.
  3. Had the OBA and FDM ran in every constituency it is unlikely that the FDM would have won any seats.
  4. It is unlikely the OBA had any pathway to securing a majority at all.
  5. Had the OBA been able to get their voter base out they would likely have at least retained their 11 seats and probably have even increased their numbers to a maximum of 15 seats (the final would have been a PLP majority of 21 to an OBA Opposition of 15).
  6. That the OBA lost 5 seats instead is indicative of a spectacular failure of their organisation – to what degree this was due to their leadership, their communications campaign, their electoral machinery (voter lists and GOTV systems) I cannot say. My personal feeling is all of the above in almost equal doses and this should mean the OBA need to do a thorough restructuring.
  7. The FDM has a significant challenge ahead of them to increase their attractiveness to OBA voters (their most fertile growth area), and will find it difficult (but not impossible) to supplant the OBA in time for the next election. Things are certainly dire for the OBA, however they have a significant institutional advantage over the FDM.
  8. There are options for additional fragmentation of the Opposition parties, including room for further 3rd parties. This would, however, be a transitional period in such an event, with survival of the fittest coming into play. The key growth area would be in those OBA voters who abstained from voting altogether. You can be sure that both the OBA and the FDM will be studying this group to see how to either regain them or attract them. The PLP will likely also be looking to see if any of these OBA boycotters can be persuaded to vote PLP going forward. Key takeaway? All those seats that flipped to the PLP will be the focus of intensive canvassing cultivation between now and the next election.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s