A matchup field-goal percentage describes recorded shots in a particular matchup sample. It does not isolate how much the defender caused those shots to go in or miss. Start by reading the season, shot count and matchup possessions, then examine the possessions that produced the number.
This matters when you use a report to prepare a guard plan. A small sample can suggest a question worth investigating. It should not turn into a confident instruction to copy a defender's approach.
Read the denominator before the percentage
Consider two hypothetical samples. A player makes two of five attempts in one matchup and eight of twenty in another. Both are 40 percent. The samples do not contain the same amount of evidence, and neither tells you the shot locations, clock pressure or help behind the defender.
Write the underlying makes and attempts alongside every percentage. If the report shows partial matchup possessions, preserve that unit. Those possessions are not interchangeable with minutes, field-goal attempts, or a count of complete offensive possessions defended alone.
A possession can include several defenders and several actions. The recorded matchup is a tracking attribution, not proof that one player controlled every part of the possession. Consult the HoopBrief methodology before comparing fields with different definitions.
Ask which shots reached the sample
Watch for selection effects. A defender might prevent a drive and force a pass without creating a recorded shot attempt in the sample you are reading. Another might contest a difficult late-clock attempt after an earlier breakdown. Shot results alone do not distinguish those sequences.
Use this checklist beside Matchup Lab:
| Check | Question | What it prevents |
|---|---|---|
| Season | Are both samples from the same season? | Mixing roles and rosters across years |
| Meetings | Did the sample come from several games? | Treating one night as a stable pattern |
| Attempts | How many shots produced the percentage? | Hiding a tiny denominator |
| Context | What help and shot types were involved? | Giving one defender credit for the whole scheme |
| Baseline | What comparison is actually available? | Comparing unlike rates |
Do not fill a missing field with a guess. If a shot-type breakdown is unavailable, say that and keep the conclusion narrow.
Separate a descriptive difference from a cause
Suppose a hypothetical scorer shoots worse in a matchup than in a broader season sample. Several explanations remain possible: the defender, the team coverage, a different shot mix, an injury, a changed role, or ordinary variation. The difference itself does not choose among them.
Use language that preserves the evidence. “The recorded sample is below the scorer's comparison rate” describes the table. “This defender shuts the scorer down” adds a causal claim the table does not establish.
When you review film, look for mechanisms you can describe: early pickup, screen navigation, denied passing angles, or help positioning. Even then, distinguish the clips you observed from a season-wide conclusion.
Convert the table into a scouting question
A useful next question might be: “In these possessions, did the defender influence the first driving direction, or did the help defense remove the second option?” That question leads to something you can watch and teach.
A weak question is: “Who is the best defender?” It asks one aggregate field to settle a much broader evaluation. Your report may help narrow the film review, but it cannot replace the definition of the task.
Use the free NBA scout for a supported player pairing. Compare the proposed guard plan with the available evidence. Ask for uncertainty and the counter the offense could use. If you use HoopBrief Ask, name the season and the specific decision you need to prepare.
Keep the conclusion useful
Finish with three lines: what the sample says, what it cannot say, and what you will inspect next. For example: “The observed shooting rate differs from the comparison sample. The table does not isolate individual defensive impact. I will review the first screen and the help behind the defender.”
That is a stronger preparation note than an unsupported ranking. It can guide a workout or staff discussion without pretending to resolve every part of player evaluation.
Is a larger sample automatically unbiased? No. More observations can reduce some uncertainty while preserving selection effects, tracking limitations and contextual differences.
Can I compare matchup possessions with shot attempts? They measure different things. Keep the labels and units separate, and do not divide them into a new rate without checking what each field includes.
When should I upgrade to a paid plan? When the report and follow-up workflow repeatedly help with a task you can evaluate. Read the current plan details; a subscription does not remove the limits of the underlying data.