Why is IPV Valuable?

The best proof of concept for IPV is its ability to accurately compare IPV scores between NFL and college QBs who played in very different situations. I charted 14 top prospects from the 2025 NFL Draft class the exact same way I chart NFL QBs — no artificial adjustments. Every single draft prospect recorded a negative IPV value , placing them below the average NFL starter.

By any other metric, these players performed exceptionally in college — impressive yardage, completion percentage, TD/INT ratios, and passer ratings. 12 of them posted a QBR higher than every single NFL QB in 2024. All 14 were above the average NFL starter in QBR, and 13 had a PFF grade above 80.

The IPV scores those prospects produced in college closely mirror their actual 2025 NFL performance. Cam Ward, Jaxson Dart, and Tyler Shough were the highest IPV scorers and the only prospects to fall into the NFL starter range — they all showed promise as rookies but ranked in the bottom half of the NFL in QBR, validating their below-average IPV scores. Shedeur Sanders struggled more, posting an 18.9 QBR — exactly in line with his NFL-backup IPV tier. The two lowest-IPV QBs, Max Brosmer and Brady Cook , performed abysmally: a combined 2 TDs and 11 INTs with QBRs of 13.9 and 11.7 respectively.

Traditional stats like passing yards, TDs, INTs, completion percentage, and passer rating are called "individual stats" — but in reality, there is nothing individual about them. They reflect the entire offense's passing game and are heavily dependent on factors outside the QB's control:

• Receiver performance — getting open, catching the ball, yards after catch
• Offensive line — blocking ability, pocket cleanliness, time to throw
• Opposing defense — player talent and coaching scheme
• Offensive scheme — mental and physical demands placed on the QB
• Game situation — down & distance, score, time, timeouts

As a result, QBs see major year-to-year swings as the situation around them changes. Advanced stats like EPA per dropback and QBR control for some of these factors — but they still fundamentally represent an offense's overall performance. For example, an incomplete pass is always a negative EPA play — even when throwing the ball away is the smartest possible decision.

PFF grades attempt to isolate individual performance but have key limitations IPV improves upon. PFF grades each play on a -2 to +2 scale in 0.5 increments. IPV gives every play an exact value in points added above or below the average NFL starter — to the nearest hundredth of a point.

PFF's concept of "zeroing out" plays — awarding a 0 for a QB making the right decision to throw it away — is an improvement over penalizing the play, but still not accurate. QBs in tough situations can only ever reach a 0 on many plays. Meanwhile, QBs in simple offenses surrounded by talent can easily rack up +0.5s, inflating their grades. PFF grades don't fully separate the QB from the players around him.

Most critically: PFF cannot directly compare college and NFL QBs. 13 of the 14 draft prospects I analyzed had a PFF grade above 80 in 2024 — ranking in the top half of NFL starters on that scale — despite all posting negative IPV scores and struggling as NFL rookies. Additionally, IPV breaks performance into four distinct categories: Decision Making, Handling Pressure, Creativity, and Execution — something PFF grades do not offer.

When I use IPV to call a throw special, I'm not citing the eye test or a stat contaminated by surrounding talent. I'm saying: based on data from every passing play of the 2024-25 NFL regular season, this play is worth exactly X more points than what the average NFL starter produces in the same situation.

IPV evaluates the outcomes a QB generates given the difficulty of his circumstances — not traits like arm talent, accuracy, or mechanics. It's descriptive, not predictive. Used alongside scouting, IPV can expose sleepers and busts, and force scouts to ground their evaluations in something more objective than "productive" — a label often derived from stats that don't account for factors outside his control.

IPV's main blind spot is the pre-snap process. It evaluates everything a QB does after the snap but does not record whether he changed a protection or audibled based on the defensive look. Determining whether protections are set correctly — or whether a QB even has the freedom to audible — requires context I can't confidently assess.

There are also the standard challenges of evaluating QB play: we don't always know the playcall or how a QB is being coached, and decision-making analysis involves subjectivity. But I don't find this a compelling critique — those pushing it never offer a legitimate alternative. Decision making is an integral part of the quarterback position. Any metric that refuses to evaluate it cannot claim to describe QB performance.

Yes, analyzing QB play is inherently subjective. But IPV does so as objectively as possible — while still accurately describing a quarterback's individual value.