Tom Brady Is Not The G.O.A.T.

By Michael Gertz
Thursday, August 3, 2017

Following his 5th Super Bowl win, the media seems to have reached a consensus opinion on Tom Brady: he's the greatest quarterback of all time. There is also agreement as to why this is the case. After all, 5 Super Bowl wins is more than Joe Montana's 4. And it's pretty tough to defend any more nuanced position than that in a 20 second segment on TV. Fortunately, we don't have such restraints here, so let's attempt a more thorough analysis.

Judging Quarterbacks on Championships

Using championships as an absolute proxy for player greatness is a faulty approach that persists in all sports, but it seems particularly pervasive in football due to the unique importance of the quarterback position. The oddest part of this simple approach is that it stands in direct contradiction with many other similar simple status quo approaches to football analysis like rushing and defense being the most important aspects, or games being decided primarily by line play.

We all know on some level that judging quarterbacks by championships is a bit flawed. Not many people consider Eli Manning a greater QB than Dan Marino despite having 2 more Super Bowl rings. And no one even mentions that Terry Bradshaw was also tied with Brady and Montana at 4 championships. Yet when it comes to separating quarterbacks with similar resumes, championships become a tiebreaker that trumps all else to such an extent that it goes unquestioned.

One of the simplest sources of contradiction with this type of analysis can be seen by looking at Tom Brady's career progression. In the first 7 seasons of his career, Brady was a good QB who managed to win 3 Super Bowls by playing on very good (and lucky/rule bending) teams despite finishing in the 5th-10th range in QB rating and never being named league MVP. From 2007 on, Brady played at a much higher level, but has won less Super Bowls in more seasons. No one would argue that pre-2007 Brady was better, but if the two were different players, the early Brady could easily be considered by pundits to be the better player.

A little bit more reasonable take on things instead gives Brady credit for his more general playoff success, including 2 Super Bowl losses and 4 AFC championship losses in addition to his 5 Super Bowl wins. While this consideration mitigates some of the randomness of championships alone, it still gives team credit to an individual, which is just a rounding error that doesn't need to be committed when we have more precise measurements of how to divide credit among players or units.

Using Advanced Stats

To avoid using championships as the only way to separate any quarterbacks who routinely make Pro Bowls, let's delve deeper into some advanced stats that are a bit more precise than conventional ones. Our Expected Points Added stats go back to 2009, and can pinpoint how much different specific types of plays contributed positively or negatively to a team's net point differential in each game.

Below is a table comparing Brady to some of his recent Hall of Fame level peers, including Peyton Manning, Aaron Rodgers, and Drew Brees. Each value is for expected points added per game above the league average for that season, and includes all seasons by each QB from 2009-2016 apart from Rodgers' 2013 and Manning's 2015 which were half seasons due to injury. All values are for the team rather than the QBs themselves alone, but to help single out the QB's contributions, different values are listed each including only specific play types.

Average Team Expected Points Added
Play Types	T. Brady	P. Manning	A. Rodgers	D. Brees
Pass Plays	5.9	7.3	6.7	6.3
& Run Plays	7.1	7.6	6.3	7.2
& Unusual Plays	7.9	8.0	6.6	7.5
& Defense	8.8	9.4	8.2	4.6
& Special Teams	10.4	8.4	7.4	3.8

Each row includes all play types listed in rows above it, so the bottom row is equal to each player's team's total net point differential per game. The unusual plays category includes several highly random play categories we track like fumbled snaps, pre-snap penalties, lost fumbles on runs or receptions, kneel downs, and the last play of each half.

Starting with the top row being the most QB dependent and including only pass plays, Tom Brady comes in last of the 4 contemporaries. This isn't entirely surprising though, as he also doesn't stand out in some more conventional QB stats like QB rating. Where the advanced stats really come in handy is in detailing how Brady has had so much more success despite his passing game being unspectacular relative to the others in this group.

By the time we include rushing plays, Brady makes up considerable ground and stands in the middle of the pack. While by no means should a QB receive full credit (or even a majority of credit) for success on run plays, there is an argument that they should receive partial credit. Defenses often back off some when facing strong passing attacks, which can open up running lanes. Unfortunately for Brady, at this point we have dished out about as much credit as is reasonable to give a QB for a team's success, and he still doesn't stand out among this elite group of quarterbacks.

He gains a little more ground on the rest of the group when considering the unusual plays that have little to do with a QB. At this point we have considered all the plays a QB is actually on the field for, and now Brady is near the top but we still have yet to explain the Patriots' unique success during his reign.

The Patriots defense was probably Brady's biggest advantage in his first 3 championships (prior to this data). Since 2009, it has given the Patriots about a 1 point advantage per game compared to league average, but Manning and Rodgers have actually also benefitted from defenses even slightly better than that. Those figures are pretty remarkable given the amount of cap space that needs to be spent on an elite quarterback. The one team that hasn't been able to work with the limited cap space is Brees' Saints, which have been 3 points per game worse on defense than league average in those 8 years. That's somewhat equivalent to Brees being stuck playing only road games over the whole timespan, which explains why the Saints have often struggled just to make the playoffs while the other 3 QBs have made the playoffs in every single season considered. It also explains why Brees has received much less credit from the media in recent years than he has deserved.

The final piece of the puzzle is special teams, which don't usually play a large role over this large a sample size, but do in this case. While Manning and Rodgers still have had the benefit of good defense, the cap limitations cost them almost the same amount of expected points back on special teams. Drew Brees has suffered through similarly poor special teams, costing each of the three QBs almost a point per game. Brady, on the other hand, has seen a massive 1.6 point net positive edge over the league average on special teams in this time frame. That big difference alone has given Brady's teams the edge over Manning's, and resulted in his Patriots being at least 2 points per game better overall than those of the other 3 great QBs.

In fact, on defense and special teams alone the Patriots have been about 6 points better per game than the Saints. That 6 points alone explains the difference between the Patriots annually going deep in the playoffs and the Saints often not even making them. It's hard to imagine that simply swapping the 2 QBs to the other team over their careers wouldn't have resulted in G.O.A.T. consideration for Brees and no more than a couple Super Bowl rings for Brady. And harder yet to think that at very least in that scenario Brees wouldn't be considered the better QB.

Clutch in the Playoffs?

Since 2009, Brady's teams overall have been about 2 points per game better than Manning's and 3 points per game better than Rodgers', with that edge coming essentially entirely on plays where none of them were on the field. But should a couple points per game really result in the different playoff outcomes? After all, in just those 8 years, Brady went to 6 AFC championships, won 3 of them, and won 2 Super Bowls. And in their combined 12 seasons considered, Manning and Rodgers managed just 5 conference championship appearances, winning 3 of them and 1 Super Bowl. Granted, this all excludes Manning's unimpressive final half season and surprise Super Bowl win, which itself should tell us something about using championships to judge QBs.

Typically, 2 or 3 points only increase the odds of winning a game by around 5-10%, which multiplied by 16 games results in about 1.2 more wins per season. The actual win totals since 2009 have Manning matching Brady's 12.4 average, with Rodgers at 11.3. But for Aaron Rodgers those small differences were magnified by the playoff seeding system, and the NFC being more competitive than the AFC in recent years.

As a result, Brady managed to be a 1-seed 4 times, a 2-seed 3 times, and a 3-seed once. While Manning managed a similar 3 1-seeds, 1 2-seed, and 1 3-seed, Rodgers was stuck with 2 5-seeds and 1 of each of the rest of the seeds from 1 to 6. Therefore, despite winning only a game less per season, Rodgers was left with 2 byes and 10 road playoff games compared to Brady's 7 byes and just 2 road playoff games since 2009.

Turning back to the advanced stats to analyze Brady's 17 playoff games over this span, the Patriots' pass offense has been 1.1 expected points per game worse in the playoffs than their regular season levels for each season. That is probably around what is to be expected, as despite most of the games being at home, the level of defense faced in the playoffs is probably a couple points better than in the regular season. But it certainly doesn't point to Brady being clutch or playing any better in the playoffs. Therefore the key to the Patriots playoff success (at least since 2009) is more in line with their ability to get great playoff seeds, which has required great play from all 3 phases of the game rather than simply G.O.A.T. QB performances alone.

And going back to his first 3 Super Bowl runs, Brady didn't exactly tear it up with clutch playoff performances. He improved over the years from unimpressive following the 2001 season, to solid after 2003, to very good after 2004.

Conclusion

The Patriots' dynasty has largely been the result of consistently great playoff seeds, even prior to 2009. And those seeds have only been possible by producing well rounded teams that have talent far beyond just a great pass offense.

Some in the media have argued that at least within the pass offense, Brady hasn't had as much support from great receivers that other QBs have. But while Brady has survived a few seasons without many options, he's generally had a good receiver like Wes Welker or Julian Edelman, a very good pass blocking offensive line, solid pass catching running back options, and has happened to play with one of the best tight ends of all time in Rob Gronkowski.

The final knock against Brady is Bill Belichick. Whether deserved or not, Belichick is widely regarded as the G.O.A.T. of coaches in his own right, and by many of the same pundits that refer to Brady as the greatest. This begs the question: how can two people both be given singular credit for the same team accomplishments? If either of them needed the other to win the 5 Super Bowls, then how many would they have won without the other, and would that be enough to both be the G.O.A.T. on their own? Evaluating coaches is extremely difficult because it is always plagued by this sort of chicken-and-egg circular reasoning, but there has to be some discounting of Brady's accomplishments if we are to assign any credit to Belichick.

The biggest key to Brady's success, which has also gone largely under the radar, is actually his willingness to take salaries of about half his market value for much of his career. While many elite quarterbacks have generally settled for a bit less money than they could get from free agency in order to stay on their team and free up cap space, Brady has taken it to a whole other level (perhaps thanks to his wife's income).

Taking giant pay cuts has allowed Brady to be really the only great quarterback of the salary cap era to have consistently played with good defenses and special teams. But even here Brady doesn't receive all the credit. Clearly, Belichick and the Patriots front office have made well above average personnel decisions over the course of the dynasty that have given Brady a much better supporting cast than even his pay cuts alone would grant on an average team.

The class of Tom Brady's peers that we have compared him to are Hall of Fame caliber, but their greatness has also been exaggerated due to The Changing Landscape of the NFL inflating passing stats to the point where the current QBs will always appear to be the best. So even if Brady's peers have been a uniquely great group, it's difficult to consider any player for G.O.A.T. when they don't even stand out among their own generation of quarterbacks in terms of on field influence rather than team accomplishment.

The bottom line is that the evidence does not point to Brady contributing more to his offense (even in the playoffs) than Manning, Rodgers, or Brees. Therefore, to rate him above them, the accomplishments of Brady's defenses and special teams must be attributed to him. And no sane person would bypass a team's front office and coaching in order to credit a quarterback's leadership for the success of a defense and special teams in any other context. Brady's low salaries are part of the equation, but also play no role in accessing his on field greatness.