On February 13th, the Rams and Bengals will battle it out for the Super Bowl title. This is certainly the sporting event all of us in the football stratosphere have been waiting for, but there is something to settle before then.
That would be the 2021 award voting. Will Ja’Marr Chase be the offensive rookie of the year? Will TJ Watt translate his sack record into his first defensive player of the year award? This is all to be settled on Thursday, February 10th, along with one another critical award: the 2021 NFL Most Valuable Player (MVP) award.
Honestly, what does MVP really mean? Really, it comes down to your definition of “most valuable”. Does that mean the player who performs the best? Or can we accurately decide which player meant the most to his respective team? There is a lot of subjectiveness that plays into it, but I’d like to take a look from a more objective basis.
Today, we’ll be utilizing historical MVP voting data through 2011. After finding out which metrics matter the most when it comes to getting an MVP vote, we’ll take a look at which player passes the bill as this year’s deserving 2021 MVP winner. There are a lot of ways to vote, but using how voters have decided in the past allows us to start to have an accurate blueprint of what an MVP winner should do in order to get the award. So, who should be the 2021 MVP? Let us take a closer look!
What Metrics Matter Most In The MVP Voting?
Since the MVP award has essentially been a quarterback award, we are going to focus exclusively on that position In my opinion, there are much stronger metrics to evaluate quarterbacks than year’s past. Statistics such as PFF grade and completion rate over expectation give us a strong idea of how well a quarterback played independent of his supporting cast, while expected points added (EPA) per play illustrates how effective a team’s passing attack was. That being said, voters still rely on older metrics, and that’s okay- as long as it’s unbiased with a clear amount of objectiveness in the process, it can be defended. With that in mind, there are three metrics that stand out above the rest:
Metric #1: Total Touchdowns
In a landslide, total touchdowns are the statistic that is rewarded the most, just like in fantasy football. Why is this the case? Well, I can think of a few reasons. For starters, scoring touchdowns is obviously correlated with leading an efficient, high-scoring offense. Meanwhile, the main way you can contribute to your team is by scoring points, so it makes sense the voters see this as a way of identifying value. Now, as we know, not all touchdowns are created equal; a quarterback sneak from the one-yard line isn’t the same as a 70-yard touchdown. Plus, does who scores the touchdown really accurately define the most important player on the drive- really, it’s just opportunity driven. Regardless, this is what the voters have latched onto.
Metric #2: Wins
Trust me; this frustrates me to no end. Wins should not be considered a quarterback stat, and for obvious reasons. It takes a complete team effort at all phases of the game to get a win, and a quarterback should not be penalized for poor roster construction or poor luck. Let me put it this way; was Jimmy Garoppolo a better quarterback than Justin Herbert because the former won more games than the latter? On the bright side, we know that the quarterback position, based on Wins Above Replacement and just anecdotally, is by far the most important position in football, so they do contribute to wins and losses a lot. Since 2011, no quarterback with fewer than 10 wins has finished in the top-three of MVP voting, with the average being 12.78 wins in a 16-game season. Remember, that’s just to be considered for MVP! If you’re looking for the MVP winner, they probably are going to need to be at the top of the league standings.
Metric #3: Total Yards
Total yards do factor in the MVP voting equation, albeit significantly less than total touchdowns. I can see why – total yardage is less indicative of an offense’s overall success than points scored obviously – though one can wonder if the gap should be as extreme as it is. Yes, yards can be accumulated in garbage time and other less-important scenarios, but so can touchdowns! Also, if a quarterback gets the team down the field, but the running back scores the touchdown, who was more important on that drive? Moving forward, if we continue to utilize surface-level metrics, I’d like to see total yardage take a greater weight in the voting.
Also, every game counts! Though players sometimes can miss games due to rest, season-long statistics correlated better with votes than per-game statistics. This is something that needs to be addressed on a case-by-case basis, but for our research, we are going to rely on accumulation statistics.
Creating The MVP Index
Just like that, we have everything we need to create the MVP Index! Using these three metrics, we can create a score to determine who should be the MVP, based on traditional voting criteria. This might seem overly simplistic, but, hey, simple can be good. As I mentioned previously, it would be nice to see more consideration for some of the advanced metrics we now have access to. However, until that is the case, we need to project the voting results with what we know correlates with it.
The score will be on a 1-100 scale, based on their percentile ranking among top-three MVP candidates since 2011. Here are the average statistics for each of the quarterback MVP winners:
| MVP Index | Total TD | Total Yards | Wins |
|---|---|---|---|
| 59.78 | 45 | 4809 | 13.11 |
With that, let us take a look at how each top-three candidate has rated in the MVP index since 2011:
Each of the MVP winners are highlighted in yellow. Meanwhile, using the correlation between MVP index and voting share (r= .66), we can predicted the estimated share of the votes the quarterback should get in an average year. As you can see, Aaron Rodgers (2014), Matt Ryan (2016), and Tom Brady (2017) benefit from weak competition, although you can make the case Ryan should not have won the award in 2016 based on traditional voting trends.
Then, there is poor Drew Brees in 2011, who absolutely was deserving of an MVP award. His 5476 total yards ranked first on this list, but, unfortunately for him, he fell just short to Aaron Rodgers in touchdowns and wins. The voting should have been closer regardless, and considering Brees never won an MVP award, it is extremely unlucky that his best season came at the absolute worst time. Josh Allen (2020) also finished above the traditional score for an MVP winner, though he was, just like Brees, beaten by Rodgers. What is it with Rodgers timing his best seasons with other elite quarterback years as well? Luckily for Allen, he has plenty of time to win his first MVP award.
Who Should Be The 2021 MVP?
Now, it is time to answer the question we’ve all been waiting for! Who should be the 2021 MVP? Let us find out. There are a number of ways we are going to attack this question. That being said, we have to start with the MVP Index!
First off, let me note that this has been adjusted to account for the fact it was a 17-game season this year.Based on the MVP index, Tom Brady definitely comes out as the clear favorite. He leads the qualifiers or is tied for the lead in all three major categories, so if voters are consistent with how they have conducted business previously, he should end his career with an MVP award. What a great ending to a career that would be!
However, I think it is fair to question if the voters’ past process has been correct. After all, they are using surface-level statistics with several flaws, such as quarterback wins, leading me to question whether this is truly defining who the most valuable player in 2021 was. I referenced our access to advanced metrics previously, and it’s time to put them to use. Using the quarterback’s percentile ranking in both PFF grade and completion percentage above expectation + EPA/play composite score, we can better identify which quarterback performed the best:
Here, we get a much different result. Both metrics here are efficiency statistics, which hurt Brady in multiple ways. For one, he was a part of a much more pass-heavy offense than Rodgers and Burrow, meaning he had more opportunities to accumulate touchdowns and yards; he also played one more game than both Rodgers and Burrow.Now, at the same time, this also meant it was harder for Brady to be more efficient than Rodgers and Burrow; there was more on his plate, and it is naturally harder to maintain efficiency with more volume.
You see how there are different ways to frame each argument? Thus, to make this process as objective as possible, combining what voters have done with what they should do may be the best way to crown the deserving the most valuable player:
Now, when you combined the two indexes, we have a neck and neck race! Really, it appears that who you pick to be this year’s MVP should come down to personal preference.
If you believe in efficiency statistics as the best mark of an MVP, then Rodgers likely gets your vote. He led the league in EPA/play and had more yards/pass attempt (7.8 vs 7.3), and also had a significantly higher completion rate above expectation. At the same time, if you have favor traditional accumulation statistics, Brady should come out in front. He attempted 188 more passes than Rodgers, threw for 1201 more yards, and also threw for six extra touchdowns. That’s much more of a difference than just the one game that Rodgers missed due to being on the COVID-19 list.
Burrow finds his way into third place, while the rest of the candidates, even if they show up strong on the traditional MVP index, likely should not be in the consideration to finish in the top three this year. He led the league in both PFF grade and yards/pass attempt, so I’m sure he’s going to have his supporters for MVP. The season-long dominance wasn’t there for me to see him as this year’s MVP, but he should be the runaway favorite for Comeback Player of the Year.
Overview
So, who should be this year’s MVP?
If I had to pick, I would slightly lean Brady. The production of the offense can be tied to him in terms of producing air yards, while he was tasked with maintaining his efficiency in a very high-volume passing attack. Using traditional metrics, he runs away with this award, and isn’t as though he also didn’t finish third in EPA/play as well.
That being said, if someone wanted to vote for Rodgers, I wouldn’t fault them. After being selected as the first-team All-Pro quarterback, it’s almost a lock that the current Packers quarterback is going to win his fourth MVP award, and that is totally fair. That’s the beauty of this year’s race; you cannot go wrong with the top-two candidates.
Is there a best way when it comes to deciding the MVP winner? I don’t think so. In the end, some subjectiveness is needed to make a final vote, and all variables should be considered. I might lean more on what many would consider a “new-school” approach when it comes to dismissing quarterback wins and some surface-level statistics, but context is key. Really, we want to know which quarterback performed the best in his current situation, and that means factoring in statistics that are both independent and dependent on team play. It’s such like the CY Young vote in baseball- we tend to value ERA over ERA estimators, because what happened on the field is what truly matters. That’s what I argue we focus on here.
Hopefully, this research, as well as the MVP Index, can give you a better idea on how voters have decided the award recently. This is certainly something that can be useful in the betting sphere, as those projected to score a lot of touchdowns should be valued over those who are accumulating a lot of yards or overall efficiency. Meanwhile, if you’re not at the top of the league in terms of wins, it’s going to be hard to be a serious candidate, thus narrowing the pool further. That would change slightly with a Rodgers win this year, but a larger sample size indicates that this has been rather predictive of the final MVP winner.
Is it going to be Rodgers or Brady? Find out on December 10th. Until then, let’s sit back and enjoy the debates!
Photos by Icon Sportswire | Adapted by Justin Paradis (@JustParaDesigns on Twitter)