How Matchup Grades Work
Every prop in DMP gets a matchup grade from A+ to F. The grade measures how soft or tough the opposing defense is for that specific stat and for that player's role.
The Core Idea
Not all defenses are equally bad at everything. A team might be elite at stopping points but terrible at preventing rebounds. A team might lock down primary ball handlers but give up huge numbers to stretch bigs.
DMP's matchup grade answers one question: how does this specific defense perform against this specific role for this specific stat?
The grade is stat-specific. The same player can have an A for rebounds and a D for assists in the same game, because the opponent defends those stats differently.
Grade Scale
Grades are derived from a softness score (0-100 percentile). Higher = softer defense = more favorable matchup.
| Grade | Softness Score | What It Means |
|---|---|---|
| A+ | 90-100 | Elite soft matchup. Defense is among the worst in the league at defending this stat for this role. |
| A | 80-89 | Very favorable matchup. Clear defensive weakness you can exploit. |
| B+ | 70-79 | Above-average matchup. Slight tailwind, but not a primary reason to bet. |
| B | 55-69 | Neutral-to-positive. No real edge or obstacle from the matchup alone. |
| C | 40-54 | Average. Matchup is neither helping nor hurting. Other signals should drive the decision. |
| D | 25-39 | Tough matchup. Defense is above-average at suppressing this stat. Adds headwind. |
| F | 0-24 | Elite tough matchup. One of the best defenses in the league at stopping this stat. Strong reason to pass. |
NBANBA: Archetype-Based DVP
NBA matchup grades use a 10-archetype Defense vs Position (DVP) system. Instead of grouping players into 5 generic positions, DMP classifies every player into one of 10 archetypes based on usage patterns, assist rate, and play style.
How the NBA softness score is calculated:
- DMP looks at how many points, rebounds, assists, and 3PM each team allows specifically to players of this archetype over the season.
- Each stat gets its own softness score (0-100 percentile) ranked against all 30 teams.
- For the prop you're looking at, DMP pulls the matching stat's softness score. Points prop? You get the points softness score for that archetype.
- For combo props (PRA, P+A, etc.), scores are blended with weights — e.g., PRA uses 45% points, 30% rebounds, 25% assists.
- The softness score maps to a letter grade (see table above).
Why archetypes matter: A team might be great at defending primary ball handlers but terrible against stretch bigs. Generic "PG/SG/SF/PF/C" positions miss this. A scoring wing and a three-and-D wing play the same "position" but stress defenses differently.
CBBNCAAB: 3-Bucket Position DVP
College basketball uses a simpler model because roster data is less rich. Players are grouped into 3 position buckets — Guard, Forward, and Center — and DVP is calculated per-bucket, per-stat.
Key differences from NBA:
- 3 position buckets instead of 10 archetypes
- No injury adjustments (college injury reporting is unreliable)
- Requires at least 8 games of data before showing a grade
- 350+ teams means more variance in DVP quality — grades at the extremes (A+, F) are more meaningful
The softness score calculation is the same: percentile rank of how many stats this team allows to this position bucket, mapped to the same A+ through F scale.
NFLNFL: Team Defense vs Position
NFL matchup grades use team-level defensive stats mapped to the player's position (QB, RB, WR, TE). For example, a WR's receiving yards grade is based on how many passing yards the opposing defense allows per game.
The defensive stats are ranked across all 32 teams and converted to a softness percentile, then mapped to the same grade scale.
MLBMLB: A Different Model
Baseball matchups work fundamentally differently from basketball and football. There is no team-level "Defense vs Position" because the matchup is pitcher vs batter, not team defense vs player role.
Instead of a DVP-based grade, MLB projections factor matchup quality directly into the projection pipeline through:
Platoon splits
Left-handed batter vs right-handed pitcher (or vice versa) is the #1 contextual factor. All rate stats are adjusted for the handedness matchup.
Opposing pitcher quality
The starter's ERA, K%, and WHIP directly adjust the batter's expected production. A batter facing a Cy Young candidate has a very different projection than the same batter facing a spot starter.
Park factors
Venue-specific adjustments for each stat type. Coors Field inflates offense; Oracle Park suppresses it. HR park factor differs from hits park factor.
Weather
Wind speed, direction, temperature, and humidity all affect ball flight. These adjustments don't exist in basketball.
Why no letter grade for MLB? A letter grade implies a single percentile that captures the matchup. In baseball, the matchup is multi-dimensional (pitcher quality + platoon + park + weather) and baked directly into the projection number rather than displayed as a separate grade. The projection itself is the matchup-adjusted output.
How to Use Matchup Grades
Do
- Use the grade as one signal among many — it's most powerful when it confirms other signals (projection edge, line movement, sharp action)
- Pay attention to the specific stat grade, not just the "overall" matchup reputation
- Check the injury adjustment — a tough matchup can become favorable if the best defender is out
- Look at the "Role Hit Rate" section below the grade to see how players in this role have actually performed against this team recently
Don't
- Bet solely on matchup grade — an A+ grade with no projection edge is not a bet
- Treat the grade as a prediction — it measures defensive tendency, not what will happen
- Ignore a C or D grade if every other signal aligns — elite players overcome tough matchups regularly
- Compare grades across sports — an NBA "A" and an NFL "A" are computed differently