100 Warm Tunas of 2020 was crazy! We saw our predictions rapidly change throughout the voting period, with Spacey Jane’s Booster Seat predicted for first when we first published, to be trumped by Glass Animal’s Heat Waves on the final day of data collection.
This year we introduced a new feature on our website to allow users to select the social mediums to be included in the prediction, allowing users to fine-tune the prediction that you see.
This year we continued to collect data across the following five sources:
- Instagram Posts
- Instagram DMs
- Instagram Stories
- Twitter Posts
- Reddit Comments
Summary
- We collected 4,026 entries (18.3% decrease since 2019 🔻)
- We tallied 36,156 votes across these entries (19.9% decrease since 2019 🔻)
- triple j counted 2,790,224 votes.
- Therefore, we collected a sample of 1.3%.
- We successfully predicted #1
- We predicted 8 out of the top 10 songs.
- We predicted 12 out of the top 20 songs.
- We predicted 75 out of the top 100 songs played in the countdown.
- Throughout December and January, 100warmtunas.com was loaded over 180,000 times by over 54,000 users.
Despite collecting less data than previous years, we still maintained a relatively stable sample size of 1.3%. This was due to the actual submitted vote count as reported by triple j also decreasing from 3,211,596 to 2,790,224.
Technical Analysis
This year, we were successfully able to predict #1 (Heat Waves - Glass Animals) to win. Lets take a look at the top 10 of the official countdown and match it up with the predicted places in 100 Warm Tunas:
| Artist | Title | ABC Rank | Tunas Rank | Difference |
|---|---|---|---|---|
| Glass Animals | Heat Waves | 1 | 1 | 0 |
| Spacey Jane | Booster Seat | 2 | 3 | 1 |
| Flume | The Difference [Ft. Toro y Moi] | 3 | 5 | 2 |
| Ball Park Music | Cherub | 4 | 2 | 2 |
| Tame Impala | Lost In Yesterday | 5 | 7 | 2 |
| Cardi B | WAP [Ft. Megan Thee Stallion] | 6 | 14 | 8 |
| G Flip | Hyperfine | 7 | 4 | 3 |
| The Jungle Giants | Sending Me Ur Loving | 8 | 8 | 0 |
| Hilltop Hoods | I’m Good? | 9 | 6 | 3 |
| Billie Eilish | Therefore I Am | 10 | 19 | 9 |
Lets pull apart this table a bit more and grab some statistics about how we did with our prediction overall:
| Predicted | Out Of Top N | Percentage |
|---|---|---|
| 8 | 10 | 80.0% |
| 12 | 20 | 60.0% |
| 20 | 30 | 66.7% |
| 27 | 40 | 67.5% |
| 33 | 50 | 66.0% |
| 40 | 60 | 66.7% |
| 52 | 70 | 74.3% |
| 64 | 80 | 80.0% |
| 70 | 90 | 77.8% |
| 75 | 100 | 75.0% |
| 155 | 200 | 77.5% |
From the above data, we can see that:
- The average error for the top ten ranks was 3 positions (about the same as 2019).
- Warm Tunas predicted 8 out of the top 10 songs.
- Warm Tunas predicted 12 out of the top 20 songs.
- Warm Tunas predicted 75 out of the 100 songs played in the countdown.
To compare Warm Tuna’s predictions vs actual rankings, a scatter plot has been provided below. We can see as we get closer to rank 1, the 100 Warm Tunas prediction gets better and converges upon the actual rankings played out on the day:
Accuracy of Predicted Ranks
Making exact predictions is difficult, especially when you only have a small fraction of the data. The data will generally give you a good indication, but will never allow you to predict, with certainty, that a particular outcome will occur. This is the reason we implemented confidence intervals on our predictions for the top 10.
The following tables depict the range of error exhibited across different cross-sections of the countdown.
Accuracy of the Top 10:
| Positions Off | Num Occurences | % | Positions Off (Cumulative) | Ocurrences | Cumulative % |
|---|---|---|---|---|---|
| 0-4 | 8 | 80% | 0-4 | 8 | 80% |
| 5-9 | 2 | 20% | 0-9 | 10 | 100% |
Accuracy of the Top 20:
Across the top 20, 75% of our predictions were accurate within 9 places of their actual position in the countdown.
| Positions Off | Num Occurences | % | Positions Off (Cumulative) | Ocurrences | Cumulative % |
|---|---|---|---|---|---|
| 0-9 | 15 | 75% | 0-9 | 15 | 75% |
| 10-19 | 1 | 5% | 0-19 | 16 | 80% |
| 20-29 | 2 | 10% | 0-29 | 18 | 90% |
| 40-49 | 1 | 5% | 0-49 | 19 | 95% |
| 50-59 | 1 | 5% | 0-59 | 20 | 100% |
Accuracy of the Top 50:
The trend continues across the top 50, with 40% of our predictions being accurate within 9 places of their actual position in the countdown and 78% being accurate within 29 places. This shows that 100 Warm Tunas is still reasonably good at predicting a general idea of what songs will play during the countdown.
| Positions Off | Num Occurences | % | Positions Off (Cumulative) | Ocurrences | Cumulative % |
|---|---|---|---|---|---|
| 0-9 | 20 | 40% | 0-9 | 20 | 40% |
| 10-19 | 5 | 10% | 0-19 | 25 | 50% |
| 20-29 | 14 | 28% | 0-29 | 39 | 78% |
| 30-39 | 3 | 6% | 0-39 | 42 | 84% |
| 40-49 | 3 | 6% | 0-49 | 45 | 90% |
| 50-59 | 3 | 6% | 0-59 | 48 | 96% |
| > 80 | 2 | 4% | > 0 | 50 | 100% |
Accuracy of the Top 100:
For completeness, the accuracy bands of the top 100 have also been provided
| Positions Off | Num Occurences | % | Positions Off (Cumulative) | Ocurrences | Cumulative % |
|---|---|---|---|---|---|
| 0-9 | 27 | 27% | 0-9 | 27 | 27% |
| 10-19 | 10 | 10% | 0-19 | 37 | 37% |
| 20-29 | 23 | 23% | 0-29 | 60 | 60% |
| 30-39 | 13 | 13% | 0-39 | 73 | 73% |
| 40-49 | 8 | 8% | 0-49 | 81 | 81% |
| 50-59 | 7 | 7% | 0-59 | 88 | 88% |
| 60-69 | 2 | 2% | 0-69 | 90 | 90% |
| 70-79 | 1 | 1% | 0-79 | 91 | 91% |
| > 80 | 5 | 5% | > 0 | 96 | 96% |
| Did not rank | 4 | 4% | - | - | - |
Vote Source Breakdown
100 Warm Tunas collects votes from a handful of different sources. Here is the break-down of how many votes were counted across the different sources:
| Source | Num Entries | Num Votes | Votes Per Entry |
|---|---|---|---|
| Instagram Feed | 2235 | 20531 | 9.1861 |
| Instagram DM | 481 | 4474 | 9.3015 |
| Instagram Story | 859 | 7238 | 8.4261 |
| 480 | 4150 | 8.6458 | |
| 137 | 1204 | 8.7883 | |
| Total | 4192 | 37597 | 8.9688 |
| Total (after de-duplication) | 4026 | 36156 | 8.9806 |
Vote Submissions
It comes as no surprise that the most popular day for people to share their votes was on the day of voting opened with a massive 503 entries collected, followed by the day voting closed with 359 entries collected. We also observed a slight spike on the eve and morning of Hottest 100 Day:
Our quietest day is observed to be Christmas day with only 9 submissions collected.
We can break down the votes collected for each song over time:
Wrap Up
This year, whilst picking the outright winner was a key highlight, we are most excited about being able to successfully predict 8 of the 10 of the songs which landed in the top 10. We believe this is the most important metric rather than exact positions. With the declining amount of data available, it’s getting harder and harder to pick the outright winner with certainty.
We’d like to thank everyone who shared their votes, visited our site or interacted with us on Instagram. Without our audience, 100 Warm Tunas wouldn’t be possible.