news-engine

This is the landing page for QBERT. As with our other landing pages, it’s best viewed on the web rather than over email or in the Substack App. The charts and tables will be updated regularly, and some of the text will change, too. Last year, we built QBERT, our quarterback rating system. It serves a dual purpose: Rating every QB since 1950 on a level playing field, taking advantage of advanced stats data for recent QBs while being as fair as possible to historical QBs for whom this data isn’t available; Projecting future QB performance to feed into ELWAY, our NFL projection model. The 2026 version of ELWAY will launch very soon. We don’t think there’s anything quite like QBERT1 out there. Still, quarterbacks typically round into form in their sophomore seasons. QBERT and ELWAY have, too, with extensive improvements and new features. For instance, this page now includes a mini-interactive that lets you compare quarterbacks from throughout NFL history and see how their QBERT ratings have evolved over time. Here’s Josh Allen versus Patrick Mahomes: Or Joe Montana, Dan Marino and John Elway himself. Note Marino’s insane, early peak. We’ve also put a lot of work into QBERT on the backend. Last year, I argued that Drake Maye ought to have been the NFL MVP rather than Matthew Stafford. But the new version of QBERT thinks that was a bad take and Stafford had the better season. (Note, though, that the chart below combines regular-season and playoff data, so Maye’s mediocre playoffs are a factor here.) What’s changed in the system? There are several refinements: Drops are not the QB’s fault. This is probably the most consequential change. We are quite confident that drops should be attributed to the WR, not the quarterback.2 Not only does QBERT now give a QB credit for a completion on a drop; we also estimate the additional yards and touchdowns (and even the avoided interceptions3) that would have resulted from the caught pass. QBs who had fewer drops than expected, like Maye last year, suffer a slight penalty.4 Trevor Lawrence and Bo Nix had outlier-high drop rates, on the other hand, and benefit substantially from this new treatment. Yards gained on pass interference calls deserve credit. Last year’s model ignored penalties. But having collected penalty data, we now simply treat defensive pass interference (DPI) calls as completed passes, giving a QB credit for the completion, the first down, and the air yards associated with the “catch”. This benefits QBs like Dak Prescott, who had 268 yards essentially erased from his stat line last year on DPIs, by far the highest in the league. Furthermore, intentional grounding calls are now treated as incompletions.5 Other QB-related penalties are still ignored because we’ve found they don’t provide much signal. Sneaks are cheap and not that valuable. QBERT likes rushing QBs more than the consensus, but not all QB rushes are created equal. Sneaks are almost purely a result of the offensive line and/or opportunities. We now give far less credit to TDs from sneak plays versus other rushing TDs and also apply a penalty term to sneaks to ensure that they don’t have positive expected value relative to other QB plays. Over the past three years, two QBs have benefited disproportionately from sneaks: Jalen Hurts (30 sneak TDs) and Josh Allen (21). Isolating sneaks does result in slightly more credit to non-sneak rushing TDs, however.6 Pocket passers were being undervalued. Last year, the coefficients used in QBERT were fitted on data from 2016-2024. We’ve now collected advanced-stat data going back to 20067 and refitted QBERT over a larger sample. This matters because 2016-2024 was a highly favorable era for mobile QBs like Lamar Jackson, Allen and Mahomes, whereas the late 2000s and early ‘10s were more favorable to pocket passers (Brady, Brees, Manning). This makes QBERT more robust across eras and also brings it more in line with the conventional wisdom, helping modern pocket passers like Stafford. Touchdowns now matter less, and the process that produces TDs matters more. Last year’s version of QBERT was essentially pegged to ESPN QBR. QBR is a smart system, much better than traditional NFL passer ratings, and still serves as an important scaffold for our system. But QBERT is now less wedded to it. Our reverse-engineering of QBR suggests that it places an extremely high value on touchdowns, equivalent to something like 45 yards per passing TD.8 Other systems like ANY/A assign less than half as much credit to TDs. While there is some persistent skill in red zone play, TDs are both noisy and opportunity-dependent, and TD rates don’t “travel” all that well when a QB changes teams. Long story short: the reward given to TDs has been reduced by 50 percent, with credit instead flowing into categories like first downs that better predict touchdowns in the long run. No more bonus credit for being on a good team. The 2025 version of QBERT accounted for the game script, giving more credit to quarterbacks who establish early leads. The rationale was that offenses are less efficient with a lead late in the game because it’s rational for them to play conservatively and run out the clock.9 However, upon review, essentially all of the difference stems from replacing quarterback plays with handoffs to running backs. Offenses are less efficient in the 4Q with a lead, but quarterbacks aren’t. Thus, we’ve removed the game-script variables. Mostly, they were giving QBs extra credit for being on good teams10, undermining QBERT’s goal of separating QB performance from team performance. We found that these combined changes increase the amount of signal that QBERT provides to ELWAY by 10-15 percent. In other words, QBERT is now much smarter at isolating the impact of the QB from the team context. There are also two changes that don’t technically affect QBERT’s historical grades, but do impact QBERT and ELWAY’s forward-looking projections. Interceptions are important but noisy. QBERT is both a descriptive and predictive system. How do interceptions figure into this? While INTs are the worst possible outcome from a passing play, and there is almost certainly some intrinsic INT-avoidance skill, they are relatively rare events in the modern NFL. It literally takes several years of data to reliably benchmark a QB’s interception propensity. So in projecting future performance, QBERT now regresses interception rates toward an expected rate11 until there are several years of data to empirically establish a reliable INT rate for a particular QB. QB injury risk is now modeled rather than based on arbitrary grades. We’ve also invested a lot of time in improving QBERT/ELWAY’s handling of injuries.12 In 2025, we relied on hand-curated (i.e., subjective) injury grades for QBs. Injuries are now modeled statistically instead. The rules are relatively intuitive: (1) past injuries predict future injuries; (2) career longevity provides evidence that a QB avoids injuries; (3) age increases injury risk; and (4) a high sack rate increases risk. We also found that the 2025 build of QBERT/ELWAY was understating QB injury risk by almost half. In any given week, 4-5 percent of healthy starting QBs are injured to the point where they can’t make their next start, whereas we’d been assuming a rate of closer to 2-3 percent. This page will update every week.13 We’ll move some of the text you see above to the methodology page once the NFL season is underway. So here’s what you can expect to see in each week’s update: QBERT ratings for each QB so far in 2026 Projected QBERT ratings for each QB’s next start QB depth charts with injury and point-spread ratings A breakdown of each QB’s weekly ratings for 2026, allocating credit between his passing and rushing performance and adjusting for the difficulty of the conditions he faces14 An updating lifetime leaderboard of the top QBs in NFL history. (Yes, Tom Brady remains #1.) And a mini-interactive chart where you can compare any two QBs since 1950. Who are the best QBs in the NFL? Up next, though, is our current QB leaderboard. Once the season starts, we’ll update this with 2026 data. For now, it’s based on 2025 data plus our most highly-rated QBs for Week 1 in 2026. While Allen is our #1 … yes, QBERT is fond of Brock Purdy. It would be boring if it perfectly aligned with the conventional wisdom. The next chart will eventually show our QBERT ratings for every QB to have thrown a pass15 in 2026. But since the season hasn’t started yet, we’ll display the 2025 data instead. Our adjusted QBERT ratings are lower across the board than our raw QBERT ratings because the modern NFL is very pass/QB-friendly: an average 2025 season would have been considered extraordinary by 1985 standards. However, QBERT operates on two tracks. Our historical QBERT ratings are one thing, but which QB would you most want to start for your team next Sunday? These projections account for career trajectories in a highly detailed way — young QBs tend to improve very rapidly once they gain more experience. They also account for injuries and other long-term absences. The recovery from a serious injury can take a little zip off of even the best quarterback’s fastball. We’re not anti-Chiefs here at Silver Bulletin, but QBERT thinks Mahomes is going to have to re-establish his worth after an ACL tear last year.