Let's see what ChatGPT has to say:
Which is the better measurement of quarterback performance, QBR or Passer Rating?
ChatGPT said:
In short:
QBR is the better measurement of quarterback performance — but the full answer depends on
what you want to measure. Here’s a clear breakdown.
- Total QBR (ESPN) is a more complete, context-adjusted measure of a QB’s overall impact on winning.
- NFL Passer Rating is a very limited, outdated, and context-free statistic that only measures passing efficiency.
For evaluating
overall quarterback play,
QBR is better.
Total QBR was designed to fix the limitations of Passer Rating. It incorporates:
✔ 1.
QBR weighs plays based on:
- Down & distance
- Game situation
- Score
- Opponent strength
A 12-yard completion on 3rd-and-10 counts far more than a 12-yard completion on 3rd-and-20 during garbage time.
✔ 2.
It separates:
- Receiver YAC
- Protection quality
- Play design
This means it attempts to isolate what the
quarterback actually contributed.
✔ 3.
QBR includes:
- Scrambles
- Designed QB runs
- Sacks
- Fumbles
- Penalties attributable to the QB
Passer Rating ignores all of these.
✔ 4.
- QBR: 0 to 100
- Passer Rating: 0 to 158.3, a weird historical artifact with no intuitive meaning.
Passer Rating only looks at:
- Completion %
- Yards per attempt
- TDs
- INTs
It ignores:
- Running
- Sacks
- Fumbles
- Pressures
- Game context
- Garbage time padding
- Difficulty of throws
- Expected points
So a QB can get a high passer rating by throwing screens all day.
Passer Rating is:
- Consistent across eras (sort of)
- Simple to compute
- Widely understood historically
For
quick comparisons within the passing game only, it still has some value.
If you want to evaluate
how well a QB actually played and
how much they contributed to winning, use
QBR.
If you want a simple snapshot of
pure passing efficiency, use
Passer Rating.
If you want, I can also break down
EPA/play, which many analysts (especially in analytics circles) consider even
better than QBR for evaluating QB performance.