BearsTalk Analytics
Metric Glossary
Every number we publish, in plain English — what it measures, how to read it, and which direction is actually good. Nothing here requires a statistics background.
About BearsTalk Grades
Impact and Efficiency grades are calculated using BearsTalk’s proprietary grading methodology, combining position-specific performance, production and efficiency data. Grades are scored on a 0–100 scale relative to other qualifying players at the same position.
The individual components, weighting and calculation methodology used to produce BearsTalk Grades are proprietary to BearsTalk.
Search, or jump to a category.
General
How effectively a player performs within the opportunities he receives, using position-appropriate performance and efficiency data. It is designed to separate quality of performance from pure workload.
How to read it: Scored 0–100 against every other graded player at the same position. A smaller-role player can grade well here while grading lower on Impact — the two measure different things.
Every play starts with an expected point value based on down, distance and field position. EPA is how much that expectation changed because of the play.
How to read it: Positive means the play helped the offense score; negative means it hurt. Totals reward volume, per-play figures reward quality.
BearsTalk does not attribute EPA to individual defenders, so defensive players are measured on charted defensive results instead.
The overall production and impact a player has generated within his role and position, using BearsTalk's proprietary grading methodology.
How to read it: Scored 0–100 against every other graded player at the same position. Unlike Efficiency, volume and workload matter: sustained production over a larger role can produce a stronger Impact grade.
Offensive, defensive and special-teams snaps charted for the player this season.
How to read it: Workload, not quality. It tells you how large a role the player actually has.
The share of plays that gained enough yardage to keep the offense on schedule.
How to read it: A simple consistency measure. Roughly 45% is average for an NFL offense.
Passing
The share of third-down pass plays that converted.
How to read it: Higher is better, but third-down samples are small early in a season.
Average expected-points change on third-down pass plays.
How to read it: Higher is better. Very small samples early in a season.
How far past the line of scrimmage the average throw travelled.
How to read it: Style rather than quality. A low figure means a short, quick passing game; a high figure means downfield aggression.
Passes actually thrown, excluding sacks and scrambles.
How to read it: Volume. Pair it with efficiency numbers before drawing conclusions.
The share of attempts charted as uncatchable for reasons other than a defensive play.
How to read it: Lower is better; it isolates accuracy from receiver performance.
The share of dropbacks the defense sent extra rushers on.
How to read it: How aggressively defenses have attacked him. Context, not quality.
Completions divided by attempts.
How to read it: Around 65% is league average. It is heavily influenced by how far downfield a quarterback throws, so read it alongside average depth of target.
Passes caught by the intended receiver.
How to read it: Volume measure; completion percentage is the meaningful version.
How much more often a quarterback completes passes than an average quarterback would, given the difficulty of each throw.
How to read it: Expressed in percentage points. +5 is excellent accuracy, 0 is exactly as expected, −5 is poor. It already accounts for throw depth and situation.
The share of attempts thrown 20 or more yards downfield.
How to read it: Aggression, not quality.
Completion percentage on throws of 20 or more yards.
How to read it: Higher is better, but deep attempts are few, so this moves sharply week to week.
Expected points added on attempts of 20 or more yards.
How to read it: How much the downfield passing game has actually been worth.
Every play the quarterback dropped to pass, including sacks and scrambles.
How to read it: The denominator behind most quarterback rate statistics.
Passing expected points added divided by dropbacks.
How to read it: The single best per-play measure of quarterback play. Above +0.15 is strong, below 0 is poor.
The share of attempts that produced a gain of 20 or more yards.
How to read it: Higher is better. Explosive plays are the strongest single driver of scoring.
The share of attempts producing a first down or touchdown.
How to read it: Higher is better; it rewards throws that actually move the chains.
Pass plays that produced a first down or a touchdown.
How to read it: Volume production that is far less noisy than touchdown totals.
Passes intercepted by the defense.
How to read it: Fewer is better, but interception totals swing wildly over short stretches.
The sum of the expected-points change on every one of the quarterback's pass plays.
How to read it: The single best volume measure of quarterback production.
Touchdowns thrown.
How to read it: Volume, and noisy in small samples.
Total yards gained on completed passes, including yards after the catch.
How to read it: A volume number. High totals usually mean a high-volume passing offense.
The share of dropbacks on which the quarterback was pressured.
How to read it: Largely a measure of protection and play design, not of the quarterback.
Dropbacks on which the quarterback was pressured.
How to read it: Context for his other numbers rather than a measure of his play.
Average expected-points change on pass plays inside the opponent's 20.
How to read it: Higher is better. Red-zone snap counts are small, so treat it as directional.
Sacks divided by dropbacks.
How to read it: Lower is better. Around 6–7% is typical.
Plays where the quarterback was tackled behind the line on a pass attempt.
How to read it: A shared responsibility between quarterback, line and play design.
The share of pass plays that kept the offense on schedule.
How to read it: A consistency measure that is far less volatile than yards per attempt.
The share of dropbacks that ended in an interception or lost fumble.
How to read it: Lower is better.
How many yards receivers added after catching his passes.
How to read it: More a measure of the receivers and the scheme than of the quarterback.
Passing yards divided by attempts.
How to read it: The oldest efficiency measure in football. About 7.0 is average.
Rushing
Charted tackle attempts the runner defeated.
How to read it: Higher is better; volume dependent.
Broken tackles divided by carries.
How to read it: Higher is better, and independent of workload.
Designed runs plus quarterback scrambles charted as runs.
How to read it: Workload.
Rushing expected points added divided by carries.
How to read it: Above 0 is genuinely good for a running back; the position average sits below zero.
The share of carries gaining 10 or more yards.
How to read it: Higher is better. Around 10% is typical for a starting back.
The share of carries producing a first down or touchdown.
How to read it: Higher is better; partly a function of down and distance.
Runs that produced a first down or touchdown.
How to read it: Chain-moving volume.
How often a carry ended with the defense recovering the ball.
How to read it: Lower is better. Because fumbles are relatively rare, this metric can move significantly over smaller samples.
Fumbles recovered by the defense.
How to read it: Lower is better, and very noisy in small samples.
Carries taken inside the opponent's 5-yard line.
How to read it: Who the coaching staff trusts to finish drives.
Carries taken inside the opponent's 20-yard line.
How to read it: Scoring-role indicator.
The sum of the expected-points change on the player's runs.
How to read it: Running plays start at a small expected-points disadvantage, so even good backs post modest totals.
Touchdowns scored on the ground.
How to read it: Volume, and driven heavily by how often a player is used near the goal line.
Total yards gained on the ground.
How to read it: Volume production.
Runs that began as pass plays and became quarterback runs.
How to read it: Charted to the quarterback as rushing production alongside designed runs.
Quarterbacks only.
The share of carries stopped at or behind the line of scrimmage.
How to read it: Lower is better. Around 18% is typical, and blocking drives much of it.
The share of runs that kept the offense on schedule.
How to read it: The steadiest way to separate consistent backs from boom-and-bust ones.
Rushing yards divided by carries.
How to read it: About 4.3 is average. A handful of long runs can carry the figure, so read it with success rate and stuffed-run rate.
How many yards the runner gained once a defender got to him.
How to read it: Higher is better, and it is one of the cleanest measures of the runner himself.
How much room the blocking created before the first defender arrived.
How to read it: Mostly a measure of the offensive line and scheme.
Receiving
Targets on third down.
How to read it: Trust in must-convert situations.
How far past the line of scrimmage his average target travelled.
How to read it: Role, not quality. Slot and check-down roles sit low; field-stretchers sit high.
The total downfield distance of every pass thrown his way, caught or not.
How to read it: Opportunity in terms of field territory rather than catches.
Charted tackle attempts he defeated after catching the ball.
How to read it: Higher is better.
Receptions divided by targets.
How to read it: Higher is better, but deep threats will always catch a lower share than a player worked underneath.
The share of his targets thrown 20 or more yards downfield.
How to read it: Role, not quality.
Drops divided by catchable targets.
How to read it: Lower is better. Around 5% is typical.
Catchable passes charted as dropped.
How to read it: Lower is better; drop charting is a judgement call and totals are small.
Receiving expected points added divided by targets.
How to read it: Above +0.4 is excellent; below 0 means his targets have cost the offense.
The share of targets that produced a gain of 20 or more yards.
How to read it: Higher is better.
The share of targets producing a first down or touchdown.
How to read it: Higher is better, and less depth-dependent than yards per catch.
Catches that produced a first down or touchdown.
How to read it: Chain-moving volume.
The sum of the expected-points change on plays he was targeted.
How to read it: The strongest volume measure of receiving production.
Touchdowns caught.
How to read it: Volume, and highly variable over short stretches.
Total yards gained on catches.
How to read it: Volume production.
Passes caught.
How to read it: Volume.
Targets inside the opponent's 20-yard line.
How to read it: Scoring-role indicator.
The share of his targets that kept the offense on schedule.
How to read it: A consistency measure that is steadier than yards per target.
Passes thrown in the player's direction.
How to read it: The core measure of receiving workload.
Yards gained after the ball arrived.
How to read it: Higher is better; it captures what he does with the ball in his hands.
Receiving yards divided by catches.
How to read it: Mostly a reflection of how far downfield he is used.
Receiving yards divided by targets.
How to read it: The best simple efficiency measure for a receiver, because it charges him for incompletions as well.
Offensive Line
Share of carries that went for ten yards or more.
How to read it: Higher is better. Explosive runs usually need a second-level block, so they credit the line as well.
A 0–100 BearsTalk grade for an offensive line as a unit, using position-appropriate pass-protection and run-blocking performance data and comparing the unit against the other NFL offensive lines.
How to read it: Higher is better. Because public play-by-play never records what an individual blocker did, BearsTalk grades the line as a unit and does not publish individual lineman grades.
Penalties are excluded: public data does not say which lineman a flag was on.
Share of the team's dropbacks on which the quarterback was pressured.
How to read it: Lower is better. It is the single best public signal of protection, but a quarterback who holds the ball adds to it and a quick game lowers it.
Measured for the whole line, not one blocker.
Share of dropbacks on which the quarterback was hit as or after he threw.
How to read it: Lower is better. It separates lines that give up contact from lines that only give up movement.
Share of the team's dropbacks that ended in a sack.
How to read it: Lower is better. Sacks are partly on the quarterback, so pressure rate is usually the steadier read on protection.
Share of carries stopped at or behind the line of scrimmage.
How to read it: Lower is better. High stuff rates point to losses at the point of attack rather than bad running.
Average yards a ball carrier gains before a defender reaches him.
How to read it: Higher is better, and it is the cleanest run-blocking number in public data: yards after contact belong to the runner.
Designed carries by ball carriers only — quarterback scrambles are excluded.
Pass Rush
Plays where the rusher forced the quarterback off his spot or his timing.
How to read it: The most common form of pressure.
Sacks, quarterback hits and hurries combined.
How to read it: The best volume measure of a pass rusher, and far steadier than sack totals alone.
Pressures divided by defensive snaps played.
How to read it: Removes workload, so a rotational rusher can be compared with a full-time starter.
Plays where the rusher hit the quarterback without recording a sack.
How to read it: Near-misses that show a rush is winning.
Tackles of the quarterback behind the line on a pass play.
How to read it: The headline number, but a noisy one — pressures describe rush quality far better.
Sacks divided by defensive snaps played.
How to read it: Workload-independent finishing rate. Small numbers by nature.
Coverage
Interceptions and pass break-ups divided by targets thrown at him.
How to read it: Higher is better, and it does not punish a defender for being targeted rarely.
The share of passes thrown at him that were caught.
How to read it: Lower is better, though a defender playing deep will naturally allow fewer.
Catches made against him.
How to read it: Volume; the rate version is the meaningful one.
Passes he caught himself.
How to read it: Higher is better, and the noisiest statistic in football.
The passer rating quarterbacks have produced throwing at him.
How to read it: Lower is better. Under 70 is excellent, over 100 is poor. Coverage samples are small, so early figures swing hard.
Passes he broke up, including interceptions.
How to read it: The best volume measure of ball production in coverage.
Passes thrown at the receiver he was charted covering.
How to read it: How often quarterbacks have been willing to test him.
Touchdown catches given up against him.
How to read it: Lower is better, and extremely noisy.
Share of throws at him that ended in a touchdown catch.
How to read it: Lower is better. Touchdowns allowed are relatively rare, so this rate can fluctuate considerably over smaller samples.
Coverage yards given up divided by targets thrown at him.
How to read it: Lower is better, and it is the cleanest single coverage measure available.
Receiving yards given up on passes thrown at him.
How to read it: Volume measure.
Tackling
Tackles shared with a teammate.
How to read it: Volume measure.
Snaps played on defense.
How to read it: Workload and role size, not quality.
Fumbles he knocked loose.
How to read it: Higher is better, and very noisy.
Missed tackles divided by total tackle attempts.
How to read it: Lower is better. Around 10% is typical; above 20% is a real problem.
Charted tackle attempts he failed to complete.
How to read it: Lower is better; volume dependent.
Tackles made without help.
How to read it: Volume measure.
Solo tackles plus assists.
How to read it: A volume measure driven heavily by snaps and by how often the ball comes his way, not purely by quality.
Tackles divided by defensive snaps.
How to read it: Removes workload so part-time players can be compared with starters.
Tackles made behind the line of scrimmage.
How to read it: The best measure of run disruption.
Interceptions, forced fumbles and fumble recoveries combined.
How to read it: Higher is better. Takeaway totals do not repeat reliably from season to season.
Tackles for loss divided by defensive snaps.
How to read it: Workload-independent disruption rate.
When a grade reads N/A
BearsTalk grades draw on position-appropriate performance data, and public play-by-play does not chart every measurement for every player.
When too little of what a grade is meant to capture is available for a player, we publish N/A — insufficient grade coverage, with no rank, rather than a number we do not trust. It is a statement about our data, never a judgement of the player: a player can be N/A on one grade and ranked on the other.
Ready to use them? Browse the player grades.