Here’s the third state in my senate series – Western Australia. I have decided to do WA next to give myself an extra day of time to determine the impact Clive Palmer will have on QLD and NSW.
Most likely elected Senate (WA):
1 GRN, 2 ALP, 3 LNP
I am constantly updating my calculation of primary votes to allow for the latest polls, and slightly tweaked my method for calculating minor party support to allow for more national consistency. Variation is also slightly higher to represent more uncertainty than was previously allowed for.
Primary votes:
LIB: 40.8% (+/-3.3%)
ALP: 27.4% (+/-2.1%)
GRN: 11.1% (+/-1.7%)
NAT: 4.5% (+/- 1.1%)
Minor parties: 16.1% comprising selected parties:
PUP: 3.6% (+/-1.8%)
LDP: 1.8% (+/-0.9%)
SXP: 1.0% (+/-0.5%)
Shooters and Fishers, DEM, Christians: 0.6% (+/-0.3%) each
No Carbon Tax, Sport: 0.12% (+/-0.06%) each
Likelihood of election:
LIB: 3 elected
ALP: 1 elected, and a 64% chance of 2nd senator
GRN: 83% likelihood
PUP: 32%
SXP: 10%
LDP: 6%
SFP: 2%
No Carbon Tax: 2%
DEM: 1%
SPORT: 0.2%
NAT: 0.2% (A poor result for a vote in the range of 3.4% to 5.6%)
Christians: 0.1%
Everyone else: 0
Ideological splits:
3 Left, 3 Right: 58% likely
2 Left, 4 Right: 42% likely
Summary:
- One ALP, Three LNP, almost certainly
- Greens and Labor will both get elected about half the time
- If not enough “Left” votes, then it becomes a right wing lottery.
- David Wirrpanda (NAT) cannot even get elected with a high vote
- PUP is most likely non-major “Right” party
- An unlikely but possible outcome is the election of the Sport Party, No Carbon Tax, or possibly a different Right micro party.
Unlike the previous two states, WA represents the best chance for the coalition to “make up” a senate seat by securing a 4-2 Right outcome. However, if this were to occur it would be with the unpredictable PUP holding perhaps 2 senators. The Coalition would much prefer a conventional conservative, like FF or DLP, but may have to work outside the box in negotiating safe passage of bills with Clive Palmer.
Over the coming days, I will publish equivalent summaries for other states. I won’t talk about the methodology – this is largely covered in prior posts. On election day, I will republish my full final results, updated for latest polling data.
Summaries so far:
WA: 1 GRN, 2 ALP, 3 LNP
SA: 1 GRN, 2 ALP, 1 XEN, 2 LNP
VIC: 1 GRN, 2 ALP, 2 LNP, 1 FF
Running total: 3 GRN, 6 ALP, 1 XEN, 7 LNP, 1 FF
Can I put my Queensland Senate thoughts in an off-topic post?
ReplyDeleteThe major party vote falls from the House of Reps to the senate, this means you must subtract a few percent off the Opinion Poll figures for calculation.
So using the Numbers here: http://blogs.crikey.com.au/pollbludger/2013/09/03/nielsen-53-47-to-coalition-in-queensland/
I subtracted a number of percent as per 2010:
QLD HOR result http://www.abc.net.au/elections/federal/2010/guide/sop.htm#QLD
QLD Senate result
http://www.abc.net.au/elections/federal/2010/guide/sqld-results.htm
My adjusted party votes:
LNP 38.72%
ALP 24.94%
PUP 7.86%
GRN 7.86%
KAP 3.91
FFP 3.36%
All other parties the same as 2010. New parties I gave micro votes between 0.3 or 0.01 depending on the attractiveness of their name.
Result:
LNP 2
ALP 2
PUP 1
Fishing and Lifestyle 1 or LNP 1
Nice analysis - very soundly based on data. The variation here becomes interesting, but your analysis has the sum of the major 3 parties on 71.5%, which is lower than what I've modelled and lower than historical elections. This is not to say that it's wrong, we'll all find out about this time tomorrow! If I get time tonight, I'll put your numbers through my model with reasonable variations.
DeleteWhoops, I'd been looking at my WA numbers! Qld coming soon.
DeleteWhat about the Wikileaks Party? Good brand recognition but poor campaign (nearly all the volunteers left because of their lead candidate Gerry) - range of 0.2% and 1.0% would be my best guess.
ReplyDeleteAgree. My "standard" variation for micros is +/-50% of their vote. I had the WIKI party midpoint at 0.59%, implying a variation of 0.3-0.9%. But I didn't get them elected in any scenario...
Delete