Worth the Wait

Turnovers don't really fall until Year 6

A rookie quarterback gives the ball away by interception or a fumble he loses about 10.9 times every 100 drives. For the next four seasons that statistic barely moves: 10.4, 10.6, 10.7 and 10.7. None of those years is statistically different from the rookie year (every p-value is above 0.2).

Year 6 is the first real drop, to 9.5 per 100 drives (p = 0.002 against the rookie year). That is about 12% fewer giveaways, and it takes five full seasons to get there. It stays below the rookie level after that, though Years 8 and 9 drift back toward 10.

Most quarterbacks who had their best seasons later were written off early or had simply been waiting their turn

A quarterback's best season is the one where his EPA per play was furthest above the league average (both are explained below the chart). Among the 73 quarterbacks whose careers lasted at least eight seasons, 30 had their best season before Year 5, including Patrick Mahomes, Lamar Jackson and Cam Newton in Year 2. Early stars are real, but where do the rest fall?

43 had their best season in Year 5 or later. Longer careers have more seasons to peak in, so the count is not the point; who they were is. 15 were "write-offs": they started in Years 1–3 and played below average, like Drew Brees, Matthew Stafford, Ryan Tannehill, Alex Smith, Sam Darnold and Geno Smith. People looked at their early production and labeled them as busts, failures, or "write-offs". 15 more were understudies: they were not starting yet, usually because they were behind an established veteran, like Aaron Rodgers behind Brett Favre or Kirk Cousins behind Robert Griffin III.

Most good quarterbacks show it early, but a third are late bloomers

Of the 104 quarterbacks with at least one starting season, 63 eventually had an above-average season. Most of them, 41, did it by Year 3. Teams are right that good quarterbacks often show it early. But 22 of the 63 (35%) did not get there until Year 4 or later, and 15 not until Year 5 or later: Aaron Rodgers, Drew Brees, Tony Romo and Kirk Cousins in Year 4, Alex Smith and Sam Darnold in Year 7, Geno Smith in Year 10. A team that judges its quarterback after three seasons gives up on the analytical evidence that quarterbacks need time.

Rookie to Year 2 is the biggest jump, but it isn't the peak

Let's look at a group of 58 quarterbacks. These are the quarterbacks from the data who started 8+ games in Year 5 through their careers, measured by EPA per play against the league average. The jump from rookie year to Year 2 is by far the largest: from 0.069 below average per play to 0.035 above. Across all quarterbacks with back-to-back starting seasons, 74% improve from Year 1 to Year 2, more than in any other year.

But the group keeps climbing after that, dips in Years 4 and 5, and is at its best in Year 8 (+0.061). Judging a quarterback by his second season catches him at the start of the climb.

Most teams give up on their quarterback by Year 4

For the 88 quarterbacks who started for their original team and are no longer its starter or retired elsewhere, I found the last season they played there. The median is Year 4. 61% were gone by Year 4 and 73% by Year 5. The single most common last season with their original team is Year 3.

In other words, most teams moved on before the quarterback's turnovers came down (Finding 1) and before the late bloomers arrived (Findings 2 and 3).

The most common number of starting seasons a team gives its quarterback is one

Of those same 88 quarterbacks, 30 (34%) got exactly one starting season from their original team before it moved on. Another 15 got two and 7 got three: 59% got three seasons or fewer.

A handful got more than a decade, but they are the exceptions. The typical quarterback gets a short run to prove himself, much shorter than our findings say he needs.

The money forces the decision before the learning curve plays out

The clock is ticking. A first-round quarterback's rookie deal costs his team about 3.5% of the salary cap each year (a second-rounder about 1%). Keeping him after that costs far more. Of the 24 quarterbacks who have entered the league since 2011 and have been given a large second contract from their own team, 16 signed it before Year 4, after only three seasons, and none after Year 6. Those deals have continued to grow from a median 14% of the cap (2014–16) to about 22% (2020–25), a fifth of the whole roster's budget on one player. But who gets paid? 17 of 22 quarterbacks who were above average in Years 1–3 got one, but only 4 of the 29 who were below league average early got one. 12 teams never gave a big second contract to a quarterback they started with.

But why 2011? In 2011 the NFL and the NFL Players Association signed a collective bargaining agreement which reformed rookie and free-agent contracts. Drafted rookies now received a standard 4 year contract and first-round picks had a team option attached in the fifth year. Put together: teams have to decide whether to commit a fifth of their cap space after only four seasons at the latest, before the Year 5–6 improvement in Part 1, and they almost never pay a quarterback who struggled early. As much as the statistics disagree, it is a strong reason to move on quickly, and it is simply about money, not about how good the quarterback could become.

21 quarterbacks had their best season after their first team gave up on them

Of the 44 quarterbacks who started for their original team and at least one other, 21 had their personal best season for a team other than their original, and 18 of those seasons were better than the league average.

Drew Brees left the Chargers after Year 5 and had his best season in New Orleans in Year 9 (13–2). Jared Goff left the Rams for Detroit and had his best year there in Year 9 (15–2). Ryan Tannehill went from slightly above average in Miami to a +0.20 per play in Tennessee. Sam Darnold and Geno Smith were below average with the Jets, and both later had above-average seasons in Seattle, including Darnold's 14–3 Super Bowl winning season in 2025.

Given time, the early strugglers' box scores caught up

25 quarterbacks were below average in Years 1–3 but kept getting starts afterward, including Drew Brees, Matthew Stafford, Alex Smith, Ryan Tannehill, Derek Carr and Sam Darnold. In Years 1–3 they completed 55.1% of their passes for 203 yards and 1.12 touchdown passes per game. From Year 4 on: 61.2%, 241 yards and 1.49 touchdowns per game.

The 29 quarterbacks who were good early improved only a little over the same span (58.5% to 60.3%, 233 to 250 yards, 1.56 to 1.64 touchdowns), so the strugglers closed most of the gap, it just took time.

Most early strugglers never become above average, but first-round "busts" often did

Of the 53 quarterbacks who were below average in Years 1–3, only 18 (34%) ever had an above-average season after that. Of the 32 who were above average early, 25 (78%) went on to have an above-average season after that. Early play does seem to predict later play, but draft position sharpens it. Among the early strugglers, only 2 of the 11 quarterbacks taken in Round 4 or later (or undrafted) became above average, and 9 of them never started another season. A late pick who struggles early is usually what he looks like.

First-round picks are different. 11 of the 25 first-rounders who struggled early later became above average, including Sam Darnold, Matthew Stafford, Ryan Tannehill, Alex Smith, Mitchell Trubisky and Teddy Bridgewater. They were drafted high for a reason, but after early struggles were called busts despite still having more to give.