Assuming Brees breaks the record, we can expect a four-hour telecast devoted to the greatness of Drew Brees, which is largely warranted. Brees is a future Hall of Famer and one of the most accurate quarterbacks in the history of the game. And he’ll be breaking one of the oldest records in football, one currently held by the standard bearer at the position.

If you’ve been at Football Perspective for long, you probably know where this is going. How impressive will it be for Brees to break this record? The short answer is, probably not as impressive as you might think.

What are the odds of throwing a touchdown in 47 straight games^{1}? Brees deserves all of the credit and praise he gets for being an elite quarterback, and a Blaine Gabbert-type is obviously not going to be the one to break this record. But the real question we want to ask is what are the odds of a star quarterback throwing a touchdown in 47 consecutive games. We can get a pretty good estimate of that.

In 44 games from 2002 to 2005, Marc Bulger threw a touchdown in 93% of his games, or 41 of 44 games. And in two of the games where he did not throw a touchdown, he threw fewer than five passes. In 73 games from 2000 to 2004, Daunte Culpepper threw a touchdown in 86% of his games. Brett Favre, from 2001 to 2004, threw a touchdown in 95% of his games. Eli Manning, from 2005 to 2011, threw a touchdown in 86% of his starts and was booed by Giants fans in just as many. Peyton Manning, excluding his rookie year, threw a touchdown in 87% of his games with the Colts. Philip Rivers once threw a touchdown in 50 of 54 straight games. Aaron Rodgers has thrown a touchdown in 50 of his last 53 games, with one of his zeroes coming in a partial game against the Lions. Tony Romo has thrown a touchdown in 90% of his games since 2007, and 92% of those games if you exclude two games he did not finish. Matt Ryan has thrown a touchdown in 30 of his last 31 games. Matthew Stafford has thrown a touchdown in 28 of his last 30 games, with one of his shutouts coming in a game he did not finish due to injury. From ’99 to ’01, Kurt Warner threw a touchdown in 93% of his games.

And then there’s Tom Brady. Since 2007, Brady has thrown a touchdown in 92% of his starts, or 94% if you exclude his game against the Chiefs when he tore his ACL in the first quarter. Brady has also thrown a touchdown in each regular season game the past two seasons, which means he could also break Unitas’ mark in 2012.

There is obviously an upper limit to the question ‘what is the likelihood of an elite quarterback playing at an elite level throwing a touchdown in any given game?’ Last year, I speculated that Brees’ likelihood was around 89-91%, which in retrospect, might be a little low. The upper limit is probably closer to 96 or 97%, although obviously very few quarterbacks could get there. If we assume complete games — i.e., that the quarterback won’t get injured or get benched or rested — maybe a star quarterback in today’s game has a 94% chance of throwing a touchdown in any given game.

In that case, such a quarterback has a 5% chance of throwing a touchdown in 48 consecutive games. In some ways, of course, this is a “what are the odds of that” sort of question. Yes, Brees is at 43 in a row, but he’s not alone. Brady has thrown a touchdown in every game the last two seasons. Stafford has done it in 18 straight games, Rodgers for 17, and Ryan is at 15. And, of course, players like Kurt Warner and Brett Favre and Peyton Manning have played at elite levels for stretches just like Brees.Perhaps the better question is, assuming 14 elite quarterbacks playing at elite levels play in 48 straight games, and each has a 94% chance of throwing a touchdown in any given game, what are the odds that *none *of them go 48/48? The answer to that: 48%. In other words, it is more likely than not that *some* quarterback would break the record.

Brees deserves all the praise in the world for essentially putting himself at the upper limit of elite quarterback play. He deserves credit for having a quick release and excelling at pre-snap coverage, which limits the amount of hits he takes. On the other hand, he’s fortunate to have almost entirely avoided playing in poor weather. He’s fortunate to have avoided injury on the hits he has taken, and to have not played for a coach that chose to bench him for a meaningless game after a drive or two (in fact, he missed week 17 of the 2009 season entirely, keeping his streak alive). He’s also fortunate that he’s thrown a touchdown in 43 games and not 41 or 42 out of 43, like many other elite quarterbacks. Brees has had bad games during this streak — in 7 of them, he’s averaged 4.8 AY/A or fewer — but he always managed to throw at least one touchdown. That’s less skill than luck, and you can read about some of Brees’ near misses here.

The skill involved for Brees is getting himself to that upper limit. Given enough quarterbacks playing at elite levels for enough years, Unitas’ record was bound to fall. Brees happens to be one of those quarterbacks.

- Assuming independence and a consistent rate per game [↩]

{ 7 comments… read them below or add one }

Thank you for a cogent and sane analysis of this topic. I’m getting rather tired of sports analysts claiming that when one of the modern QBs breaks a record, of course he’s better than the guy whose record he broke. In reality, most records are quite meaningless and have not as much to do with skill as one might think, as evidenced by the number of single game and single season records held by players who are clearly not amongst the best at their position. Non-sensationalist sports journalism is a real breath of fresh air. Now I’m gonna go yell at Skip Bayless and everyone else on First Take.

Thanks, Andrew. Appreciate the love.

“Given enough quarterbacks playing at elite levels for enough years, Unitas’ record was bound to fall.”

“assuming 14 elite quarterbacks playing at elite levels play in 48 straight games, and each has a 94% chance of throwing a touchdown in any given game, what are the odds that none of them go 48/48?”

One good statement, and one good question. The only problem is that you failed to take into account the ability of elite quarterbacks to be able to stay healthy for long enough to accomplish the feat.

If a qb is able to pass for 25-30 touchdowns in a season he would have a chance to pass for a td in all the games of that given season. Obviously, the higher the number, the more likely he’s able to meet this first qualifier. The equation also needs to take into account how many times he’ll have to carry the team, overcome elite pass defenses, and even face historically inept defenses. A bad defense might allow 4-6 tds in one game. If the qb only throws for 25, then he’ll have to complete a td pass in 15 other games with only 21 chances.

Passing for a touchdown in 16 games isn’t as easy as it seems. You just pointed out four qbs that have thrown a td for 16+ games, which is a lot lower than the 14 that you cited in your question. I feel that you’re starting with too many “elite” qbs. You spoke of 14 qbs. Well, that means that there would be four qbs from each conference that had had pro bowl/all-pro/hall of fame seasons that didn’t make the pro bowl. Your statement suggests that 44% of the NFL not only have “elite” qbs, but that they also performed at that level.

The way that I’d like to see this set up would be the likeliness of a qb to make it through a full season, for two in a row, then for three and four in a row, also. From there I’d like to see the percentage of NFL qbs that are considered elite. Take the percentage of elite qbs from the number of qbs that are likely to make it through a season, then multiply that number by the number of qbs that have thrown a td in every game of a given season. That would be the starting point to see if an elite qb could accomplish this feat.

I don’t have a problem with your article’s premise, just the way that you went about stating how it’s more likely to happen than not. If that were the case, then this record wouldn’t have stood for 51 years, or greater than half the history of the NFL.

Thanks for stopping by, Jim. Sorry for the confusion: I didn’t mean to imply that there are 14 elite QBs playing today. Just that over the last few decades there have been (at least) 14 elite quarterbacks who played for three straight years and *could* have done this. Fouts or Marino or Young or Montana or Manning or Favre or Warner or whomever. Now it’s probably easier today than it was 15 years ago because of changes to the game, but my point was that with enough super elite quarterbacks getting close to that upper limit, it would be more likely than not that some QB would break it.

It’s great that it’s Brees, but it doesn’t make him better than Peyton Manning or Tom Brady or Aaron Rodgers. Now he might be better than those quarterbacks, but if so, it’s not because of this streak.

Thanks, Chase.

I thought about all the qbs that could’ve done it over the last few decades, I just thought that you menat now.

There’ve been so many, and Dan Marino and Dan Fouts were the first two that came to mind. Then the Bill Walsh proteges of Ken Anderson, Joe Montana, and Steve Young. Honestly, I thought that one of those five had a great chance to break his record. It’s just amazing that it’s lasted this long.

Anyway, good article. It was definately thought provoking.

Drew Brees has managed to avoid injury? The dude ripped his shoulder to pieces. It’s a miracle he is even playing football let alone leading the single greatest sports comeback story (the Saints) in history.

What was not brought out was the likelihood of a quarterback passing for 47 straight games from 1956-60, when the rules did not favor the offense and quarterbacks could be killed on every play legally. The statistics quoted in the article are one of current probabilities-but what were they during Unitas’s era? It makes the Golden Arm’s record that much more golden and incredible. Perhaps akin to Dimaggio’s 56 game hitting streak being about 100 games today?

{ 2 trackbacks }