Creating a Prediction Model to Predict this year’s Grammy Winners

One of the biggest nights in music and one of the greatest and prestigious award to musicians, The Grammys signify a recognition of an artists’ best efforts getting rewarded. When it comes to some of the categories, there are is no shortage of talent to fill the nomination slots.

From nomination day to the Big Night, there is much push and pull about those who are on the Grammy plate: who is poised to win, who deserves to win, and who actually wins.

This year is no exception, there is an enormous pool of talent to pull from, and 2024 is one of those years that I am sure will be recognized and recognized as a year that brought a slew of innovative and enjoyable music, reigniting genres and artists, creating a night of real competition across every genre and category.

This is one of the first years in quite a while that the Best New Artist category is as contested as it is, with many of the artists in question making incredible breakthroughs with there music that break boundaries and make way for new modes of creativity and inspiration.

When it comes to conducting data analysis and creating exploratory projects, it almost always seems like history can repeat itself. Having a well-informed outlook at how a certain artist came to get to a nomination, as well as a good knowledge base on how the Grammy’s usually reward artists, should be able to almost create a pattern and a good base for predicting a crop of winners for that year.

So I thought, why not use my love for data analysis and pop culture and try to figure out for myself who would win this year’s Grammy’s? Could I come up with a robust collection of relevant data to build and train a prediction model that could accurately predict who could become this year’s lucky big 4 winners (in my case we’re looking at big 5 – but more on this later).

But first let’s get a bit familiar with everything.

how does the grammy process work?

According to the official Grammy’s site, this is how the voting process goes down.

Once the eligibility is open, artists and companies will submit their works and campaign for Grammy nominations.

The Academy’s voting members, all involved in the creative and technical processes of recording, then participate in (1) the nominating process that determines the five finalists in each category; and (2) the final voting process which determines the GRAMMY winners.

THE RECORDING ACADEMY GRAMMY AWARDS VOTING PROCESS

what is the big 4

The big 4 are the 4 most prestigious and most important awards that are presented at the Grammy’s ceremony. These awards encompass the total theme in which the awards ceremony is about – and rewarding artists for pushing boundaries or creating performances in all aspects of music that deserve its recognition.

This could be a reward in the terms of writing or performance – as seen in the Song of the Year and Record of the Year categories, or releasing an artist or body of work that can bring out fresh new conversations, as with Best New Artist or Album of the Year.

The 411 On The Big Four GRAMMY Categories

Song of the Year is an award that focuses primarily on the songwriting and the writers involved with making that year’s hit song. Many times this award can get confused with Record of the Year, which awards the song and its performance, as well as the production that was involved in creating the song.

Album of the Year is an all encompassing award, that awards the performer, the performance or performances, as well as the production which includes the producing, mixing, engineering, writing and more. This award looks the album as a complete body of work, and nominates based on its entirety

The Best New Artist award is a bit different than you might think. The Best New Artist can be described as an artist within the last couple of years who have made a breakthrough into the music scene with their introduction (or moment of final recognition) to the world.

The Big 4 are voted on by the selected and by vetted 12,000+ voters (F/N: 13,000 now as repeated by the Grammy presenters EVERY award, 🙂) who vote on the work best fits the criteria’s definition of that past year.

what is my big 5

when it comes to me I love pop and the conversation around it and what it does for the cultural zeitgeist. I love following my pop girlies, and watching some grow into mega stars.

Pop music is often pigeonholed into a basic category – and that pop enjoyers don’t have a substansial affinity for “real” music. Whatever that means.

In the past couple of years, pop has begun to expand to new heights and find a new definition of the mainstream. I can only describe today’s pop scene as bringing the left field style of music to the forefront.

Experimental hooks, slightly offbeat hooks, off-kilter visuals to match has come into the scene with much fervor and much subsequent acclaim. Many artists such as Charli XCX, Beyonce and my personal client JADE have made waves in the pop scene in the past years (and even longer!) reviving this type of music into the main scene and it is so exciting.

Other more traditional pop acts have padded the year out to be as great as it was, as Ariana’s return to the music scene, paired with Billie’s emotional HIT ME HARD AND SOFT and Tate McRae’s brave new entries onto the scene have made the genre as legit as it once was.

And of course, there are artists who can serve us both, with artists like Chappell Roan and the return of Lady Gaga tossing their nachos to their respective cooking device and creating a michelin star meal.

So what other category makes sense for me to be interested in as much as the other big 4 – as best pop vocal album?

To me, Album of the Year encompasses a body of work that had something to say beyond its initial lyricisms and choruses. Albums that can start conversations, movements, or a new wave of style or thinking, is what I think encompasses that Genre.

But when it comes to best pop vocal album, this category makes the most sense to my interests as I like to see who simply just brought it in the year – created a solid body of work that’s REAL GOOD POP MUSIC or brought about a pop performance that has you thinking its a certified hit right off the bat.

the project in question

Within the last couple of years, the Grammy’s have been almost predictable, albeit still enjoyable, in the picks that have been made up of the Big 5. Almost if it had become a formula as to which Grammy darling would take the cake for the year.

the winner’s of the past 5 years.

Year	Category	Album/Song	Artist
2020	Album of the Year	When We All Fall Asleep, Where Do We Go?	Billie Eilish
2021	Album of the Year	Folklore	Taylor Swift
2022	Album of the Year	We Are	Jon Batiste
2023	Album of the Year	Harry’s House	Harry Styles
2024	Album of the Year	Midnights	Taylor Swift
2020	Best New Artist		Billie Eilish
2022	Best New Artist		Olivia Rodrigo
2023	Best New Artist		Samara Joy
2024	Best New Artist		Victoria Monet
2020	Best Pop Vocal Album	When We All Fall Asleep, Where Do We Go?	Billie Eilish
2021	Best Pop Vocal Album	Future Nostalgia	Dua Lipa
2023	Best Pop Vocal Album	Harry’s House	Harry Styles
2024	Best Pop Vocal Album	Midnights	Taylor Swift
2020	Record of the Year	Bad Guy	Billie Eilish
2021	Record of the Year	Everything I Wanted	Billie Eilish
2022	Record of the Year	Leave the Door Open	Silk Sonic
2023	Record of the Year	About Damn Time	Lizzo
2024	Record of the Year	Flowers	Miley Cyrus
2020	Song of the Year	Bad Guy	Billie Eilish
2021	Song of the Year	I Can’t Breathe	H.E.R.
2022	Song of the Year	Leave the Door Open	Silk Sonic
2024	Song of the Year	What Was I Made For?	Billie Eilish

I wanted to see if there was truly a pattern on these past wins, and what makes each entry the winning candidate for each category. I also wanted to see what internal and external influences could have an impact on the chances of a piece of work coming out out on top, and if I could calculate these influences and come out with a prediction for myself – ahead of the committee’s final decision.

Some of the factors I looked at was an artist’s history with the Grammy’s, the impact and success of a work (and in some cases with the help of a collaborator), and the impact and success of the artist themselves.

To get a historical look at the history of the Grammy’s and the artists that shaped it, I gathered relevant data for the creation of two prediction models, a simplified and detailed model, which was then combined and worked into a final prediction model that could calculate these past influences together and create a work or artist’s probability of winning.

the data

To create a robust model, I first started with collecting the nominee data from the past 25 years, collecting the data on the nominees for all five major categories from the year 2000 to 2025.

(shoutout to web scraping, beautifulsoup4, and totalmusicawards.com for letting me work smarter and not harder – I created a script to pull out a list for me which included the year, category, work and artist.)

From there, I wanted to calculate each artist and their relationship to the Grammy’s themselves. I wanted to create a system to calculate the quote unquote “Grammy Darlings”, or artists that have good rapport with the award show and system, and could have a higher chance of winning.

My equation included 3 different types of metrics that were calculated from the Big 5 categories and combined to create a sort of hierarchy score. This included:

• Historical Nominations (Nominations before the year 2020) – every nomination here was worth 1 point.
• Recent Nominations (Nominations from 2020 until this year)- every nomination here was worth 1.5 points.
• Total Wins (Wins in Big 5 category)- every nomination here was worth 2 points.

The top 10 “Grammy Darlings” according to the Weighted Grammy Calculations

Contributor	Historical Nominations	Recent Nominations	Total Wins	Weighted Score
Taylor Swift	12	15	6	46.5
Billie Eilish	0	17	7	39.5
Adele	11	4	9	35
Beyonce	8	10	1	25
Bruno Mars	12	1	5	23.5
Lady Gaga	10	5	1	19.5
U2	7	0	5	17
Kendrick Lamar	9	5	0	16.5
Norah Jones	6	0	5	16
Coldplay	6	3	2	14.5

From the 325 total nominees from the year 2000 (also keep in mind here, I also pulled out features in individual songs and considered the song as their own nomination as well – ex. Peaches by Justin Bieber featuring Daniel Caesar and Giveon – all three were given a point for nomination), I pulled out the nominees from the current year and the top 50 non-2025 nominees (I actually pulled 51, but you’ll see why in a second) and got a clean list of 60 artists (you see..) to pull further research on.

The weighted scores of all the 2025 Nominees

2025 Nominee	Historical Nominations	Recent Nominations	Total Wins	Weighted Score
Taylor Swift	12	15	6	46.5
Billie Eilish	0	17	7	39.5
Beyonce	8	10	1	25
Bruno Mars	12	1	5	23.5
Lady Gaga	10	5	1	19.5
Kendrick Lamar	9	5	0	16.5
Ariana Grande	3	5	1	12.5
Post Malone	2	6	0	11
Sabrina Carpenter	0	5	0	7.5
Chappell Roan	0	5	0	7.5
Jacob Collier	0	2	0	3
Shaboozey	0	2	0	3
Charli XCX	1	1	0	2.5
Andre 3000	0	1	0	1.5
Teddy Swims	0	1	0	1.5
Raye	0	1	0	1.5
Doechii	0	1	0	1.5
Benson Boone	0	1	0	1.5
the Beatles	0	1	0	1.5

This was all the data I needed for my initial and simplified prediction model; this model used an artist’s historical and current standing with the Grammy’s to calculate their likelihood of a continuous win.

With these 60 artists and all their nominations from the given time period, I was able to gather the real meat and potatoes, which would be used for the detailed prediction model.

From here, I split the data into two additional avenues of influence research: the impact of the artist and their cultural standing and and similar data for the artist’s nominated work*.*

Artist Impact	Artists’ Works Impact
Part of the top 50 Weighted?	Work Type
Are they a 2025 Nominee?	Category Nominated For
Spotify Popularity Score	Year of Nomination
Spotify Follower Count	Did this nomination win?
Last.fm Listeners	Spotify Popularity
Last.fm Scrobbles	Last.fm Listeners
Engagement Score	Last.fm Scrobbles
Retention Score (Last.fm)	Engagement Score
Retention Score (Spotify)	Retention Score (Last.fm)
Total Chart Score	Retention Score (Spotify)
Cultural Impact Score	Chart Score
Total Artist Score	Cultural Impact Score
Scaled Total Score	Total Works’ Score
Scaled Cultural Impact	Scaled Total Score
Final Artist Score	Scaled Cultural Impact
	Final Works’ Score

For the initial data, such as the artists’ popularity and listens, I used Last.fm and Spotify to scrape their listening data and create a sort of average. I wanted to look at these numbers holistically, so I created some equations again to come up with engagement and retention scores, to see how audiences are listening to an artist or a piece of work, as well as if they keep coming back to it.

Mixing those in with the popularity of the artist on Spotify, I was able to create the cultural impact score, the score that I used to basically influence most of my predictions. From there I could pretty much gauge the popularity of an artist or work, and how it impacted the culture.

One of the most important factors I wanted to look at to glean external influence in addition to an artist’s popularity was their history with the charts. If the grammy history was the meats of the data I wanted to collect, an artist and their work’s chart history were definitely the potatoes.

(again, shout out web scraping, beautifulsoup4, and this time musicchartsarchive.com)

From here, I calculated a whole bunch of formulas to create an average chart score on the artist and their works levels. Using their chart score, I looked at that in relation to the cultural impact score and created a final artist or works score.

Artist
Score	Formula
Charted Albums	How many albums an artist has seen charted
Charted Singles	How many singles an artist has seen charted
Average Album Peak Position	Mean of Album Peak Positions
Average Single Peak Position	Mean of Single Peak Positions
Albums Score	# of Charted Albums × 25
Singles Score	# of Charted Singles × 25
Album Peak Score	(100 – Avg Album Peak Position) × 25 / 100
Single Peak Score	(100 – Avg Single Peak Position) × 25 / 100
Final Chart Score	Sum of scores for albums, singles, and their peak positions
Final Artist Score	(Final Chart Score × 0.7) + (Cultural Impact Score × 0.3)

Work
Score	Formula
Peak Position Score	(100 – Peak Position) × 0.4
Weeks on Chart Score	Weeks on Chart × 0.3
Average Position Score	(100 – Avg Position) × 0.3
Final Work Score	(Chart Score × 0.7) + (Cultural Impact Score × 0.3)

When pulling these equations, it was easy to see that the numbers were fluctuating all over and across all equations, with artists being calculated with their own monthly listeners and popularity rates. So I went ahead here and did a min-max equation to sort of standardize the scores and create a pretty well rounded score for each artist and work.

Here I was able to get my finalized data on my artists and their nominations! I was able to standardize the scores to start the prediction model process.

Here’s an example of Sabrina Carpenter and her nominated work, Short n’ Sweet:

Artist	Total Chart Score	Cultural Impact Score	Total Artist Score	Scaled Total Score	Scaled Cultural Impact	Final Artist Score
Sabrina Carpenter	488.19	44	355	0.06	0.22	0.10

Work	Type	Category	Year	Engagement Score	Retention Score (Last.fm)	Works’ Chart Score	Cultural Impact Score	Total Works’ Score	Scaled Total Score	Scaled Cultural Impact	Final Works Score
Short n’ Sweet	Album	Album of the Year	2025	1.10	0.52	149.72	0.81	131.04	0.45	0.00	0.22

the prediction model

To get my most accurate prediction model, I used a combination of all this data I took my sweet time collecting.

All of my databases (artist scores, work scores, grammy weighted scores) were seperated, so I needed a bit of scripting magic to put all of the artist and work level scores together onto one level, with the artist detail being attached to every individual work.

Within this merging process, I had to remove the Best New Artist category and add it back later, as the lack of work in the rows would have caused issues in the merging process.

Once merged, there was some factors that I wanted to make sure were included in my model to kind of get a bird’s eye view of all the data that we were looking at. For each of the 2025 nominees, I pulled out all the past nominations that they have been granted and looked at a couple of things.

• I made sure to add a competitiveness factor for each category, looking at how many nominees were in each to get a more accurate look at how crowded the category is.
• If a work had a collaboration in it (ex. Die with a Smile with Lady Gaga and Bruno Mars) a collaboration bonus was added. A combined piece of work can usually pull from the popularity of both of it’s artist so I wanted to account for that advantage there.
• If a nominee is a first time nominee (this was a little messy in its definition, it only pulled if they got 1 nomination in 2025 and never had any before so artists like Chappell Roan didn’t get this benefit becasue she was nominated twice in the Big 5, while Doechii would since she got only 1), they would also receive a first-time nominee bonus. This accounts for the underdog and novelty factor that can lead to a potential win. This kind of helped me a little, but we’re getting ahead of ourselves!
• Within the chart scores I wanted to make sure everything was still included so I still brought back some data that was merged into calculated data and averaged it in a different way such as:

Score	Formula
Chart Stability Score	Weeks on Chart / (Peak Position + 1)
Chart Entry Ratio	Charted Albums / (Charted Singles + 1)
Peak Ratio	Avg Album Peak Position / (Avg Single Peak Position + 1)

• And finally, I took the chart scores, the engagement scores, and the cultural impact scores to generate a weighted engagement score on the artist and work level to kind of get an average of everything we’re looking at.

Then, I brought the Best New Artist nominations back to the chart, with their artist chart scores, engagement scores, and the cultural impact scores.

Now it was just time to train the model. For this I used Random Forest Classifier.

For the prediction model, I wanted 2 things to happen in this prediction process:

1. Train 5 models simultaneously for each category to only find probability within those nominated works
2. Output the likelihood of each nominee winning the award, with a probability percentage score.

From here, there were some pretty obvious consistencies with artist and work performance and their likelihood of winning I took the top two probable winners of each category and compared them for my final predictions.

so, who do you think will get the grammy?

And now, onto my model’s predictions.

best new artist: chappell vs raye

I was actually a lttle surpised with this one. Although I knew that Chappell would make it the top two, I was surprised to see her rating so low, and that Raye would be the one trailing in second place.

(Here, I predicted it to be Chappell vs Sabrina, but I love Raye dearly so there is no complaint over here on that!)

Both Chappell and Raye saw breakout success in vastly different, but also similar ways. Escapism became one of Summer 2023’s biggest hits thanks to TikTok and moderate success of other singles like Flip a Switch and Oscar Winning Tearsallowed Raye to find success in her 2022 album My 21st Century Blues.

Raye also took a SWEEEEP at last year’s Brit awards, cementing her as an artist to stay, and even without American American acclaim yet, she proved to be an artist that will do all right.

Raye: Singer-songwriter makes history winning six Brit Awards

RAYE MAKES HISTORY, WINNING SIX TROPHIES AT 2024 BRIT AWARDS

Chappell Roan saw quite a huge success, and at a much faster rate than her contemporaries. After her überviral Coachella performance hit the scene, Roan quickly saw many of her singles and eventually her previously released album The Rise and Fall of a Midwestern Princess begin to reach new heights.

A Timeline of Chappell Roan’s Rapid Rise From YouTube Covers to Global Stardom

It’s Chappell Roan’s world, we’re just living in it: The full story of her meteoric rise to fame

Both being first time nominees and fairly new to the world, I think this competition really came down to the cultural impact which is what lead to our winner.

my winner:

With Chappell, there’s no doubt that her rise to the top in the last year qualifies her for best new artist. Even with the steep competition, her swift rise and support from other pop stars no doubt puts her at the top of this category.

“Your favorite artist’s favorite artist”

–That’s what streets were saying, IDK

Although her probability score is low at a meager 14% win rate, I don’t think there’s a doubt in my mind that she’s gonna take it here.

song of the year: birds of a feather vs die with a smile

Both Billie Eilish and Lady Gaga are no stranger to the pop genre, with both being major players in the streaming game. Their chart histories are amongst the top highest, and both had a major standing in generating their prediction score.

Bruno Mars’ added impact added a bit of a bump here as well, as his streaming power also had an impact in “Die with a Smile’s” standing. The collaboration bonus in the model took note of this.

Eilish’s past Grammy performance edge gave her a bit of tilt here, with her 7 past wins in the Big 5 categories being a reflection of her success. Lady Gaga has the advantage when it comes to chart power, with her her artist score coming out with a higher score, due to having a longer period of highly performing singles and albums.

Overall my model gave Billie a 65% win probability for Birds of a Feather, with Die with a Smile trailing a little behind with a 48% win probability.

my winner:

So why does Billie come out on top here, as a top 10 Grammy darling in her few short years of her career, and her streaming giant status generating a high cultural impact score, the combined effect was undeniable here. Her past track with a previous Song of the Year win was a boosting factor.

record of the year: birds of a feather vs espresso

For this year’s Record of the Year crown, the showdown was something I found to be quite interesting. The top 2 contenders, couldnt have been more different in terms of standing as an artist, as seen in the Song of the Year battle with Lady Gaga and Billie Eilish.

This time, Billie sees herself again in the top two spot for Birds of a Feather along with superstar “newcomer” Sabrina Carpenter and her summer

Sabrina’s explosive and breakout year, gave her a quick and steep rise in her chart score and a cultural impact score that even rivaled longtime titan Billie.

For this, Sabrina came out almost victorious with a humble but still impressive 25% win probability score.

But it wasn’t enough for the Birds of a Feather artist.

my winner:

Again, Billie takes the hypothetical crown here due to her already existing almost Billie status in the Big 5 Categories. With 17 recent nominations and 7 wins within the sector, it is almost impossible for Billie to not leave there with something.

One thing to note here, is that when Billie took the first stab at darling status, she first had to dethrone already existing tough competitors in her category. So, maybe that can happen here with Sabrina…

onto the albums!

album of the year: hit me hard and soft vs short n’ sweet

Rematch time! For album of the year, my model again picked up Billie Eilish and Sabrina Carpenter as the the top two competitors for the given category.

Sabrina and her explosive year man. In addition to her albums #1 on the chart success, she also saw multiple singles within the album reach career heights as well, with the 3 singles (Espresso, Please Please Please, and Taste), all reaching top 10 debuts, with please please please taking the crown for reaching #1 status.

Her high chart score made her a real threat for this album of the year category, even edging out Grammy heavy hitters like Taylor Swift.

Sabrina here definitely showed explosive power, and her numbers as a newcomer to the Grammy scene was was surely a surprise to many. However, when it comes to Album of the Year, the committee usually rewards consistency and prolonged presence, which is why her probability score stayed low at a 28% win probability.

my winner:

Here again, my model proves why having past experience with the Grammy community can provide an edge with future success. Billie is known for slicing up the competition, so my model obviously feels like she can do it again, especially with the success that it had within the year.

When it comes to this prediction and the rematch effect, I actually think my model could be a little skewed and I can even see Sabrina taking this one. Although Billie is a titan and certified Grammy bully, I feel like her album as a whole wasn’t as in the zeitgeist as I feel short n’ sweet was? Idk besides Birds of a Feather and maybe Lunch, I didn’t really see much of a movement from this album.

I feel like Billie got more acclaim from her Charli duet Guess than she even did with Lunch? I don’t know guys, let me know.

best pop vocal album: hit me hard and soft vs eternal sunshine

When Billie won her first win for Best Pop Vocal Album, many considered her biggest competition to be Ariana’s offering to committee. However, Billie was able to come out on top in that same year that she started to gain her darling status.

And now the time has come for a rematch. Can Billie’s sustained presence stay steadfast, or can Ariana’s return to the pop scene shake her foundation so much that she relinquishes the crown back to her competitor?

With Eternal Sunshine and a dramatic return to the pop scene, after the completion of filming her back to back Wickedmovies, Ariana was quickly able to garnish a high **53% cultural impact score ****for her album.

But Ariana has had history when it comes to a snub for this category. Would it happen again here? She herself isn’t coming to the ceremony – should that be telling us something?

my winner:

Billie’s blend of chart success, past Grammy success, and sustained longevity, again was proved that Eternal Sunshines 37.67% win probability rate could not stand the test of time.

So now, we just have to see.

one, two, skip a few..

so, how did the prediction model go? (AFTER GRAMMY’S UPDATE)

well.

My model got ONE prediction right. Best New Artist. But I think we all saw the win for Roan coming from a mile away, right? I didn’t even need the prediction model for that one!

But, as for the rest.

What. Just. Actually. Happened.

There’s a lot of factors at play here that I think skewed my model completely left, but I can’t say I’m mad at the way my model decided on results and I can also proudly say that I’m glad with the way that the wins actually went?

I really expected Billie to get one of the big 5 here, maybe song of the year, the most, but it looks like snub city for her this year. So lets look at the actual wins and why my model could have been wrong.

who actually won?

best new artist: chappell roan!!!

Embed from Getty Images

My model’s first (and only lol) correct prediction! Roan’s explosive year following her meteoric and swift rise after last summer’s performance to Coachella, culminating to her historic performance and Lollapalooza, were immediate indicators to her predicted win.

With six singles charting in this past year, her chart score boasted pretty high on the scale in comparison to her peers. This years’ crop of contestants were a hard to beat crowd, and that’s why despite Chappell’s low 14% predicted win rate, it was still a sure thing for me.

And plus, some of her competitors found success in other categories as we will see later…

song AND record of the year: not like us by kendrick!!!

Embed from Getty Images

When it came to the song and the record of the year awards, my model could not have been more off. Not in terms of overall prediction, as it did give Lamar’s Not Like Us a 21.41% predicted win rate for both categories. My model instead prioritized chart score here, when a cultural impact based on more external factors would have done this category a bit more of a service.

The explosive battle between Lamar and fellow rapper Drake, had the internet by storm, and Not Like Us was the icing on the beef cake, with the song reaching the peak number 1 position, and for 36 weeks at that. Frontrunner Billie and her hit Birds of a Feather only peaked at 2, and was on the chart for 35 weeks. Hmmmm..

Actually though, when looking at it, Kendrick had the 10th highest chart score out of all the researched artists for this project, at a gargantuan score of 2291.09. So how could he not have automatically shot to the top of the rankings? (My model had him at 3rd for record of the year and 4th for song of the year)

I think the Grammy weighted score had something to do with it.

When it comes to the big 5, Kendrick has not won once. Rap usually rarely finds itself lauded in these categories, with the last one of note being Childish Gambino’s This is America.

This discounted Lamar heavily, and positioned Billie upwards, as well as gave way for Lady Gaga, another Grammy darling, to have a chance in this category.

My model does not account for nuance in these things, does it..

best pop vocal album: short n’ sweet!!!

Embed from Getty Images

Unlike the rest of the TRUE big 4 awards, Best Pop Vocal Album is actually quite known for rewarding newcomers with sizable impact in their rookie years. Some favorites include Dua Lipa’s Future Nostalgia, and Harry Styles’ Harry’s House.

If we were going on legacy and prestige here, it would make sense for Billie or Ariana to win. But Sabrina’s breakout status in pop, with her explosive streaming success and critical fanfare in the past year made her to hard to ignore, far surpassing her predicted 25% winning rate.

Sabrina actually had a higher chart score than Ariana’s Eternal Sunshine and higher chart positioning than Billie’s Hit me Hard and Soft.

With all that being said, this was not a surprise to me all! Sabrina is my sister, we have been through thick and thin since the Singular albums, so I’m extremely happy for her and this win.

album of the year: cowboy carter!!!

Embed from Getty Images

Out of all my losses within my model, Cowboy Carter won at the lowest recorded probability rate, with a whopping 4% predicted win rate according to my model.

How could this have happened?

Snub city, is why. Beyonce is known for being THEE most decorated artist in Grammy history, yet somehow she has always found herself on the losing end of the stick when it comes to the Album of the Year category.

Her wins are also found amongst a higher range of categories, more so than her contemporaries (e.g Cowboy Carter also taking home the Best Country Album win or Renaissance taking home Best Dance/Electronic Album in 2023). Her Big 5 weighted Grammy’s score takes a hit because of this, and her 4th place score of 25 is a tumble from Billie Eilish’s second place score of 39.5.

Cowboy Carter edged out Hit me Hard and Soft, with a 2% cultural impact score against Billie’s 1% cultural impact score (Please be aware here these scores are much lower than the cultural impact scores we investigated earlier, as these are looking at the album itself, and is comparison with the artists as well). Beyonce also took the crown in longevity scores and obviously a higher artist chart’s score as she has been an artist for much longer.

The legacy of Beyonce should not have been discounted as it did in her Grammy’s Weighted Score. That, in turn with a lower album Chart Score of 65.67 against Eilish’s winning 77.91 score, made her take a hit that made my model think she was going to be snubbed once again.

SMH, model. Have hope.

why didn’t the prediction model work?

https://twitter.com/myralists/status/1886276138487427516?s=46&t=X6DZ3zt3VBxQ6PPl3NT6Tg

The first thing that I noticed with my model and the outcomes that we got tonight, is the fact that it relied on prestige and the “grammy darling” factor a lot more than it should have. I honestly think that in previous years, this model could have proved to have better results, with some slight differences.

But there are some other factors that I was able to pinpoint to see what exactly was going on! Was my model just flat out miscalculating things? Or was there more I could have taken into account?? Let’s find out.

The Novelty Factor

Embed from Getty Images

My model wanted to give Billie Eilish and all her offerings to come out of Hit Me Hard and Soft very, very, badly it seems due to her past Grammy acclaim and explosive continuing chart performance.

But what I did not account for was innovation, and that seemed to be the buzz word for this year’s awarding criteria.

There was no way I could have expected that, not with the history I have seen with Grammy’s selection. One of my biggest complaints with the 2024 Grammy’s was actually for Album of the Year, and for the same reasons that I think the selection committee took note of and applied it to this year’s win.

Embed from Getty Images

When Taylor Swift’s Midnights took the crown, I was thoroughly confused. I had 100% believed that the win would go to SZA’s groundbreaking SOS, as the innovation and new sounds found in the album were lauded and celebrated all throughout the year. For me, it felt like Midnights was a more familiar album, both in Taylor’s discography and in general.

I think the same could be described as Hit Me hard and Soft. While performing extremely well, and having a good premise, I didn’t feel the same shift and cultural impact that I saw with Cowboy Carter, especially with its boundary-less direction and a new sound for Beyoncé.

I could say HMHAS, albeit good I want to keep emphasizing that, wasn’t anything completely new. It wasn’t a complete departure from Happier Than Ever. When Billie took the sweep for When We All Fall Asleep, Where Do We Go?, it was a celebration for bringing a new sound the pop landscape.

Enough is enough (Billie, it might just be your turn)!

The Grammy’s have had more than enough history of excessively rewarding an artist, before moving on and heading to the best new big thing. This isn’t even Billie’s first time getting curved like this; she was in the headlines for the 2022 Grammy’s for the same thing, not getting any wins for Happier Than Ever.

Billie Eilish Shut Out at the 2022 Grammys.

The same thing is happening right now the Big 4 Grammy Darling, Taylor Swift. With a weighted Grammy Score at the max 46.5, you would expect her to take something home at every Grammy’s. But this year, she was also notably shut out of awards. I dont think that there was any specific reason for that, there were just too many good projects this year and Swift’s most recent album was not one her best, OBJECTIVELY.

Enough is also enough when it comes to the (lack of) celebration for minority artists. The Grammy’s has been called out by spectators and even artists themselves (namely, The Weeknd) for not taking artists of color seriously, and not being in touch with current trends.

“The award voters are not in touch with the current pop and youth culture”.

Rob Kenner, a former Grammy voting member

Hate Me Now: What It’s Like To Be A Grammy Voter

As mentioned earlier, Hip Hop Artists and Black artists in general are rarely seen to be awarded in the Big 4 categories, oftentimes relegated to the hip-hop and R&B ones. Even Album of the Year took 25 years to award itself to a black woman again, the last being Lauryn Hill’s The Miseducation of Lauryn Hill.

A hip-hop reckoning? Why Drake is sitting out the Grammys

Remember how I just said The Weeknd was one of the artists to call out the Grammy’s on this? Well, during this year’s show Harvey Mason Jr., the CEO of the Grammy’s committee issued a public apology to the artist, as well as made promises to expand on the diversity of the voting committee, to ensure fair practices in the voting process when it comes to these awards.

This is why we saw this push for innovation and quality this year in the awards, as opposed to keeping it safe as the Grammy voters were continually doing.This push, went noticed by many, as social media sentiment expressed glee, and confusion, at why the Grammy’s were actually “fun” in the way that they were rewarding artists.

They have also expanded their voters to 13,000, as the show liked to remind us over and over.

The UnderDog Effect v Legacy Status

Beyonce only has ONE WIN in these 5 grammy categories. For Put a Ring on it. That discounted her here. But Beyonce is a legacy artist, has been on the scene as a solo artist for 23 years, and is THE most decorated Grammy artist of all time.

Her continual album losses put here at a detriment here. The model looked at all of her losses, and failed to account for all her wins, all her nominations, and all of her legacy.

As mentioned earlier, my model wasn’t really well adept at knowing when to account for legacy status or the underdog/new girl in town effects. Which is fine, I didn’t think to tell it to. But that’s what caused issues in Best Pop Vocal and Album of the Year specifically.

Best Pop Vocal welcomes novelty, as seen in the forms of its previous winners, and if that had been recognized in the model, Sabrina’s probability score would have given here a little boost here.

Virality over Reputation

One thing that my model could not account for with its prediction was virality. This was the most apparent in the Kendrick sweep, and the overtaking of Billie’s predicted wins in both the Song and Record of the Year categories.

The viral shift that came in the form of Not Like Us, was definitely taken of notice this year. The viral lyricism of the song boosted its acclaim, even surpassing what the Grammy’s could have predicted.

“A MINORRRRRRRRRRR!”

Not Like Us, Kendrick Lamar

Birds of a Feather was also a huge and successful hit, finding its way through the TikTok circuit for months on end. But Not Like Us surpassed boundaries, and was a hit with all communities, and even all ages.

Embed from Getty Images

I also wanted to mention here that it’s so interesting that Kendrick got 5 Grammy awards off of a diss track, that wasn’t deemed too serious, the same amount of Grammy’s he had in total walking into the ceremony. A viral hit can really take you far in this awards’ space, and can even overtake “Grammy Darling” reputation.

Soooo, what now?

I actually really enjoyed doing this and hope to conduct more investigations like this in the future. I think with this prediction model creation and interpretation, there needs to be a good balance. In this case, a balance between legacy, innovation, and cultural relevance, as well as looking at history is important in creating a robust model that can account for “upsets” or “surprises.

Some things I can take into account with a prediction model like this can be:

• Incorporate novelty and more creative metrics
• Find a better way at looking at social media and sentimental data
• Adjust the weighting of past wins
  • Expand the Grammy history past the Big 5
• Give underdogs more credit
• Balance virality and reputation

Embed from Getty Images

ALSO: WATCH DOECHII’S PERFORMANCE AT THE GRAMMY’S PLEASEEE IT WAS TOO GOOD!

Watch Doechii Serve “Catfish” & “Denial Is A River” In An Unforgettable Performance After Winning Best Rap Album | 2025 GRAMMYs