LogFAQs > #949774291

LurkerFAQs, Active DB, DB1, DB2, DB3, DB4, DB5, DB6, Database 7 ( 07.18.2020-02.18.2021 ), DB8, DB9, DB10, DB11, DB12, Clear
Topic List
Page List: 1
TopicContest Stats and Discussion - Part 1365
LusterSoldier
01/21/21 10:45:25 PM
#326:


Poll updates and trend charts for the poll from 3 days ago:

https://docs.google.com/spreadsheets/d/1JSmVbKoUwNzqlgGCLq2hGfo0xWhVz813h45-w5g4NSI/pub?gid=305399188

This poll is a good example of how a pandemic has affected poll results, and even the trends as well.

In terms of historical trends for this poll, "None" has the most pronounced trends with a strong early vote and maintaining that strong early vote throughout the entire night vote and a brief portion of the early morning vote. It immediately collapses after the first 7-8 hours of the poll. The options "Five to ten", "More than 10", and "More than 100" all start off the poll very terribly, but improve themselves after the Power Hour ends with their best time period being the morning vote, then they all drop off hard again with their SNV being similar to the Power Hour.

The ongoing pandemic has clearly affected the trends very significantly here. This is most clearly visible with the trends for "None", which show a less extreme drop off in its trends from the Power Hour. It still performs best during the board vote and Power Hour, but it performed way better during the morning vote and day vote compared to the 2019 poll. In the most recent version of this poll, "None" lost a total of just 1.77% from 6:00 AM to the end of the poll. For reference, "None" lost a total of 4.90% after 6:00 AM in the 2019 poll. In 2018, it was 3.29% and the 2016 poll was 4.55%. So the trends were far weaker as a result of the pandemic, but this can be blamed on the options that were hurt the most by a pandemic (more on that later). What little trends "None" actually had in this poll were the caused by the options "One" and "Two to four" chipping away at None's strong early vote.

"One" is the only major option whose trends were mostly unaffected by the pandemic, as they are very similar to past versions of this poll. "One" still has a bad board vote and Power Hour, and continues performing terribly during the night vote and morning vote. It begins to perform better during the day vote, while the SNV is still its best time period just like in the previous versions of this poll.

"Two to four" also has very similar trends to the past versions of this poll, with the board vote and Power Hour being terrible and the night vote being only slightly better. The late morning vote and day vote is the best time period here, while the SNV is only slightly below average.

The options most heavily affected by the pandemic are "Five to ten", "More than 10", and "More than 100". Those 3 options were the main reason why "None" experiences a very noticeable drop off after its very strong first 7-8 hours of the poll. In the 2019 poll, the last 4 options (including the "outside" option) had 5 consecutive hours where they combined for 10% of the vote and 6 hours overall with the single best hour maxing out at 13.13%. But in the 2021 poll, the last 4 options never combined for more than 5% of the vote in any single hour (4.82% was the high point). These options simply didn't show up at all in the 2021 poll, so they really didn't have any trends to speak of.

As for the results of the poll, the pandemic has affected the results by causing "None" to perform almost 3% better compared to the 2019 poll. "One" has also done better during the pandemic, with an improvement of about 1.5% from 2019. "Two to four" has performed only slightly worse (5.71% to 5.45%). "Five to ten" has dropped from 1.54% to 0.81%. The most severe drop off came from "More than 10", which fell from 3.38% to 0.70%. "More than 100" fell from 1.13% to 0.53%. "I'm outside right now" has held steady with an identical 1.01% in both 2019 and 2021.
---
Luster Soldier --- ~Shield Bearer~ | ~Data Analyst~
Popular at school, but not as cool as azuarc, Guru Champ!
... Copied to Clipboard!
Topic List
Page List: 1