Advertisement

Jameis Winston vs. Marcus Mariota: The stats that matter

Traditional football stats are far too simplistic, lack context and can be easily manipulated. These numbers cut through the noise and tell you what’s really happening.
 
[Winston: Times; Mariota: Associated Press]
[Winston: Times; Mariota: Associated Press]
Published Aug. 17, 2018|Updated Aug. 18, 2018

Yeah, but …

Every Jameis Winston-Marcus Mariota debate is full of them.

Winston has completed more passes (939), gained more passing yards (11,636) and thrown more passing touchdowns (69) before his age 24 season than any quarterback in NFL history.

Yeah, but …

No one has played in more games (45) before his age 24 season, so of course Winston would lead in completions, yards and touchdowns.

Here’s what happens when we use games played as the qualifier: In 45 games, Ryan Tannehill had more completions. In 45 games, Kirk Cousins had more passing yards. In 45 games, Andy Dalton had more passing touchdowns.

As for Mariota, he has led more fourth-quarter comebacks and won more games than Winston. He’s even won a playoff game, leading the Titans back from a 21-3 deficit to beat the Chiefs in January. That’s the most important thing, isn’t it? A quarterback who wins?

Yeah, but …

The Titans played one of the league’s softest schedules. They beat the Andrew Luck-less Colts (twice), the Deshaun Watson-less Texans, the winless Browns and the Joe Flacco-full Ravens. Yes, they beat the Jaguars in Week 17 to clinch a wild-card berth, but Jacksonville had clinched the AFC South a week earlier.

And the playoff win against Kansas City? Give Mariota credit, but give Chiefs coach Andy Reid some, too. Postseason heartbreak is his calling card.

Part of the reason there’s a debate three years after Winston and Mariota went 1-2 in the NFL draft is that neither quarterback has significantly separated himself from the other. We’re used to declaring a winner by now. Drew Bledsoe was better than Rick Mirer. Peyton Manning was better than Ryan Leaf. Donovan McNabb was better than Tim Couch. Andrew Luck was better than Robert Griffin III.

RELATED STORY: Mariota won 2016. Here’s why. (January 2017)

Which side of the debate you fall likely comes down to the team you support. No one in Tennessee is posting essays on Reddit blaming everyone but Winston for his seven lost fumbles last season. No one in Tampa Bay is reviewing All-22 coaches film and arguing that then-Titans coach Mike Mularkey should have called more play-action passes for Mariota.

The traditional stats that populate these debates don’t add much clarity, either. They’re numbers without context. You can pick and choose the ones that make your guy look good, and people do.

Let’s cut through the noise. Forget counting stats like passing yards that bear almost no relationship to wins. The stats below are the stats that matter. They tell us not only what happened but also what might happen. And right now, they show that one quarterback holds a slight edge over the other.

RELATED STORY: Tom Jones' answer to the Winston-Mariota debate

Passer rating when not under pressure

Why it matters: While there’s value in producing when plays break down, performance in such situations tends to be volatile. Because plays under pressure rarely look alike, it’s difficult to use them as a basis for predicting the future. A quarterback’s performance when passing from a clean pocket is more consistent from season to season. If a quarterback struggles when he isn’t under pressure, how can you expect him to succeed when he is under pressure?

Stay updated on Tampa Bay’s sports scene

Stay updated on Tampa Bay’s sports scene

Subscribe to our free Sports Today newsletter

We’ll send you news and analysis on the Bucs, Lightning, Rays and Florida’s college football teams every day.

You’re all signed up!

Want more of our free, weekly newsletters in your inbox? Let’s get started.

Explore all your options

Advantage: Winston

Winston's rating Mariota's rating
2015 95.0 95.2
2016 94.4 103.3
2017 95.2 78.2

Winston’s rating when not under pressure has remained stable, but Mariota’s dropped about 25 percent last season. Oddly enough, he posted a higher rating (82.7) when under pressure. This is a troubling sign for the Titans, but one for which there might be a few explanations. Among them: the hamstring and ankle injuries throughout the season, Mularkey’s power run-based offense and the lack of a quality receiving corps. Mariota’s healthy now and has a new offensive coordinator, but the receiving corps looks the same. The Titans are counting on a pair of 2017 draft picks (Corey Davis and Taywan Taylor) to make major gains.

Follow Thomas Bassinger on Twitter

• • •

Turnover-worthy play percentage

Why it matters: Turnovers are catastrophic. According to ESPN’s Mike Clay, 43 percent of a team’s win total can be determined solely by its turnover margin.

But not all turnovers are the result of “turnover-worthy” plays and not all “turnover-worthy” plays result in turnovers. For example, sometimes a quarterback correctly diagnoses the coverage and an accurate pass bounces off a receiver’s hands and to a defender. And sometimes he misreads the coverage and throws a pass to a defender who drops the interception. The box score assigns 100 percent of the blame on the tipped interception to the quarterback but assigns no penalty for the dropped interception.

To better reflect the quality of quarterback play, the analysts at Pro Football Focus grade each dropback and not just the results of each dropback. If a quarterback puts the ball in harm’s way, he’s going to have a higher turnover-worthy play percentage, regardless of whether the defense actually intercepts it or forces a fumble.

Advantage: Mariota

Winston's TWP% Mariota's TWP%
2015 4.70% 4.73%
2016 3.96% 3.80%
2017 5.20% 2.38%

By now, you’ve heard Winston fans/Mariota critics slam the Titans quarterback because he threw more interceptions than touchdowns last season. The implication is that Mariota was more turnover-prone than Winston. As PFF’s turnover-worthy play percentage shows, this is simply not the case.

Such criticism (conveniently) doesn’t take into account the value Mariota added as a runner (five touchdowns) or the cost of Winston’s fumbles. Winston fumbled 15 times last season and lost seven of them; Mariota fumbled twice, losing one.

It also doesn’t take into account tipped or dropped interceptions. According to Football Outsiders research, Winston’s opponents dropped eight potential interceptions. Mariota’s opponents dropped two. (Before you dismiss dropped interceptions as a made-up stat, consider that they’re a better predictor of future interceptions than actual interceptions.)

When we study each quarterback’s entire body of work, as PFF is attempting to do with its turnover-worthy play analysis, we find that Mariota was much better at protecting the ball last season than Winston.

RELATED STORY: Watch the eight Winston interceptions that defenders dropped

• • •

Adjusted net yards per pass attempt

Why it matters: Passer rating can give us a quick snapshot of a quarterback’s effectiveness, but it can be misleading. It’s old (the NFL adopted it in 1973), gives too much weight to completion percentage and ignores sacks, which people often blame on the offensive line but generally have more to do with quarterback play.

Case in point: Sam Bradford in 2016. By passer rating, he was the NFL’s sixth-best quarterback. Why? Because he led the league in completion percentage. Most of his passes, however, were short and safe. By another measure — adjusted net yards per pass attempt (ANY/A) — Bradford was about average.

ANY/A is basically a souped-up version of yards per pass attempt (Y/A). It rewards quarterbacks for touchdowns and penalizes them for interceptions and sacks. Of the easy-to-calculate passing statistics out there, it is the most strongly correlated to wins. Here’s the formula: (passing yards + 20*(passing touchdowns) - 45*(interceptions thrown) - sack yards) / (pass attempts + sacks).

Advantage: Winston

Winston's ANY/A Mariota's ANY/A
2015 6.44 6.10
2016 5.98 7.14
2017 6.70 5.51

Both Winston and Mariota were aggressive passers in 2016, with Winston finishing second in average pass length and Mariota finishing third. While Winston maintained the aggressiveness last season, Mariota was slightly more conservative. He threw more passes than he had in 2016 and completed a higher percentage of them, but he did not gain as many yards or increase his rate of touchdowns.

RELATED STORY: Catching up with the guy who’s been on both sides of Winston-Mariota

• • •

Defense-adjusted value over average

Why it matters: Not all completions are the same. A 5-yard completion on third and 5 in a close game is different from a 5-yard completion on third and 15 in a blowout. Football Outsiders’ passing efficiency statistic — defense-adjusted value over average (DVOA) — takes this into account, as well as strength of opponent (hence, “defense-adjusted”). DVOA, which is expressed as a percentage, represents the value a quarterback adds per play over an average quarterback. A positive percentage indicates above-average quarterback play; a negative percentage indicates below-average quarterback play.

Advantage: Winston

Winston's DVOA Mariota's DVOA
2015 2.1% -13.2%
2016 3.6% 11.1%
2017 14.3% -3.3%

By DVOA, Winston’s 2017 was his best season yet. In fact, his 14.3 percent DVOA ranked 12th in the NFL, just behind Matthew Stafford (14.9 percent). After Mariota leapfrogged Winston in 2016, he crashed last season. His -3.3 percent DVOA ranked 20th.

• • •

Conclusion

To some, the Winston-Mariota debate was over the moment the NFL handed Winston a three-game suspension because its investigation concluded that he violated the league’s conduct policy by touching an Uber driver “in an inappropriate and sexual manner without her consent.”

The Bucs will press on with Winston as their starting quarterback. From a football standpoint, it’s hard to deny that he was a more efficient and productive passer than Mariota last season.

Yeah, but …

The debate is hardly over. Mariota was better in 2016, and maybe he’ll bounce back in 2018. A lot can change in a few months.

Contact Thomas Bassinger at tbassinger@tampabay.com. Follow @tometrics.