We are reader supported. When you purchase through links on our site, we may earn an affiliate commission. Also, as an Amazon affiliate, we earn from qualifying purchases.
Pitch evaluation changed fast. Public data moved beyond radar-gun readings and simple spin rate leaderboards. Teams now judge pitch quality with models that convert movement, speed, and release into one scale that anyone can read. That scale is Stuff+. If you have felt unsure about what the number really means, how it is built, or why it sometimes disagrees with ERA, this guide will walk you through it in clear, beginner-friendly steps.
Introduction
Stuff+ compresses complex pitch physics into a single rating. You can scan a leaderboard and know, at a glance, who brings elite raw pitch quality and who survives on skill outside raw stuff. But a single number can mislead when used without context. You need to know what the metric includes, what it leaves out, how quickly it stabilizes, and how to pair it with other stats to get the full picture. Read on if you want simple rules you can apply the next time you open a box score, make a fantasy move, or break down a new call-up.
What is Stuff+
Stuff+ is a model-based rating of the physical quality of a pitcher’s pitches, scaled so 100 is league average. The metric estimates how a pitch would perform based strictly on its speed, movement, release, and related physical traits after removing location and context. A higher number signals stronger raw stuff. A lower number signals below-average raw stuff.
The purpose of Stuff+
The goal is isolation. Stuff+ aims to measure what the pitcher brings to the mound before we consider where he puts the ball, how he sequences pitches, how the catcher frames, or how the defense fields behind him. It focuses on the ingredients of a pitch rather than how the pitch was used in a given at-bat.
How Stuff+ is built
There are variations across providers, but the backbone is consistent. The model consumes tracking data for every pitch and produces an expected quality score tied to the pitch’s physical characteristics. Then it scales those scores so that 100 equals league average for that season. The final pitcher-level number is typically a usage-weighted blend of the pitcher’s individual pitch scores.
Key inputs the models often use
Velocity remains central. More speed shortens reaction time and increases the margin for error. But velocity alone does not define stuff. Movement profile matters just as much, often more.
Movement shape measured by induced vertical break and horizontal break drives the models. A fastball with strong induced vertical break holds plane better at the top of the zone. A slider with lateral sweep creates a different swing path challenge. Shape that separates pitches from each other also helps, because it makes recognition harder for hitters.
Spin rate and spin efficiency affect how the ball moves. Spin without movement is less valuable than efficient spin that produces useful break. The same raw spin can create ride, run, or sweep depending on axis and efficiency.
Release point and extension change the geometry of the pitch. More extension shortens the flight distance, making the same velocity play faster. Release height and side angle affect approach angle and how the ball enters the zone. Small release tweaks can change how two pitches tunnel off each other.
Pitch axis, seam orientation effects, and gyro components influence both magnitude and direction of movement. The model treats all of these as signals about how the ball will behave in flight and how much deception the raw flight pattern provides independent of aim.
Standardization choices
To keep the metric focused on raw quality, most Stuff+ methods remove location and count effects. That means the same slider thrown middle-middle or backfooted under the hands gets the same raw stuff score. The idea is to grade the pitch, not the command. Models also adjust for handedness matchups and normalize by pitch type, since a changeup and a cutter solve different problems and should not be judged on identical axes.
From pitch-level to pitcher-level
Pitch-level Stuff+ reflects the quality of a single pitch. Pitcher-level Stuff+ aggregates those scores across all pitch types, usually weighted by usage. A pitcher with a dominant slider but a weak fastball might still grade well if he leans on the slider often. A starter often posts a lower number than a short-burst reliever who throws only his nastiest two pitches.
Reading the scale
Common ranges
85 to 95 suggests below-average raw stuff. These pitchers can still succeed with strong command, ground-ball profiles, and sequencing, but they live on a smaller margin for error.
95 to 105 suggests near-average raw stuff. Results will depend greatly on command, mix, and health. This range contains many back-end starters and bulk relievers.
105 to 120 suggests above-average raw stuff. This band includes many late-inning relievers and a good share of mid-rotation starters with out pitches.
120 and above suggests elite raw stuff more often seen in leverage relievers and frontline starters. This group can miss bats without perfect locations because the ball flight itself is difficult to square.
Single pitch interpretation
It is common to break out Stuff+ by pitch type. A 120 slider with a 90 changeup says the breaking ball is a weapon while the changeup lacks bite or separation. The best pitchers either carry one monster pitch that can be spammed in many counts or run a balanced mix where no pitch sinks the overall profile. Always look for both the quality of the top pitch and the floor of the weakest pitch, since hitters will hunt the weak link.
What Stuff+ does well
Stuff+ is strong at predicting swing and miss skill and at identifying pitches that age well if velocity holds. When a pitcher adds a new shape or velocity jump, Stuff+ often picks up the breakout weeks before ERA moves. It is excellent for filtering who can beat hitters in the zone without perfect locations and who has to nibble. It also helps explain role changes. A pitcher with a mediocre fastball and an elite sweeper might perform better in shorter stints where he can feature the breaking ball aggressively.
Where Stuff+ falls short
Command is not the target of this metric. That means walks, bad miss locations, and predictable sequences can bury a pitcher who still carries an excellent Stuff+ number. Deception outside of ball flight also matters. Glove position, tempo, and repeatability can raise or lower real outcomes without moving the score. Health and fatigue matter too. A pitcher might hold a great number on paper while losing feel or velocity start to start. Sample size can bite. Early season or small role samples can swing the rating before stabilizing.
Using Stuff+ in practice
Development and front office use
Stuff+ guides pitch design. If the model reads a sinker as low quality because it lacks separation from a four-seamer, a coach can chase more run or ride, shift release, or alter seam orientation. If a slider has good sweep but poor velocity, a subtle grip tweak can buy speed while keeping shape. If a changeup floats too close to the fastball, adding drop or reducing spin can help. The end goal is separation and efficiency that the model will reward and hitters will feel.
Fantasy and betting decisions
Treat Stuff+ as a forward-looking flag. A spike after a grip change or a new pitch can precede a strikeout jump. A fall after velocity loss can foreshadow struggles even if ERA still looks clean. Pair it with strikeout and walk rates, pitch usage, role, and a location or command metric to build a balanced view instead of relying on one number. Avoid reacting to a one-week swing, especially for relievers who might throw only a handful of pitches.
Broadcast and fan takeaways
Use Stuff+ to understand why certain pitches dominate. High-ride fastballs win up. Sweepers and gyro sliders attack different swing paths. Splitters play off a firm fastball because of sharp late drop. Seeing a number aligned with the eye test adds confidence. Seeing a mismatch prompts a deeper look at location, sequencing, and mix.
Case patterns you will see
High Stuff+, poor results
A reliever sits in the 120s but carries a walk problem and prone fastball misses over the heart. Stuff+ does not grade command or predict whether the bad miss lands in the middle on 2-0. Walks and barrels can erase the advantage from elite movement. The fix often involves fastball aim changes, a secondary adjustment, or simpler sequencing that keeps the edge pitches in play.
Low Stuff+, strong results
A control artist lives around 90 to 95 but pounds edges, changes speeds, and steals strikes early in counts. He gets weak contact by avoiding the middle and staying unpredictable. This profile can work for long stretches, but the margin is thin against patient lineups. Watch for loss of command or speed late in the year.
Midseason jump after a design tweak
A starter adds two inches of induced vertical break on his four-seamer, keeps the same release, and shifts to more high-zone usage. Stuff+ pops into the 110s, strikeouts rise, and walk rate stays steady. That is the cleanest form of breakout and the most stable. If velocity and shape hold, the gains often stick.
Improving Stuff+: practical levers
Velocity: modest gains move the needle across the board. Even one extra tick increases both swing and miss and weak contact potential because of reduced reaction time. Strength, mechanics, and recovery processes support this lever.
Spin efficiency and axis: refine grips to convert raw spin into useful movement. Aim for more ride on four-seamers, more sweep or depth on breaking balls, and distinct spin signatures across the mix.
Movement separation: ensure each pitch occupies its own lane. Too much overlap makes recognition easier. Distance pitches along both vertical and horizontal axes so that hitters must cover multiple planes.
Release and extension: small posture or stride changes can create more extension or alter approach angle. These tweaks help the same raw velocity play faster and change how pitches tunnel.
Usage and count strategy: while Stuff+ itself removes location and count, usage indirectly shapes outcomes. Leaning into the best pitch and trimming the weakest can raise the pitcher-level score and performance in tandem.
Sampling and stability
Per-pitch models stabilize faster than full outcomes like ERA because the inputs are physical. Still, patience is necessary. Single-game shifts rarely reflect true changes. Look for trends over weeks, not days. Check whether the underlying drivers moved. If velocity, movement, or release did not change, a Stuff+ blip is likely noise. Once a pitcher logs steady shape and speed over a meaningful sample, the rating becomes more trustworthy.
Stuff+ and the strike zone
Because location is removed, Stuff+ answers a different question than a command metric. A good framework is this: Stuff+ tells you what happens when the pitch is thrown for a strike or competes in the zone. A command metric tells you how often the pitcher hits quality targets. You want both. High stuff with below-average command still wins in short stints. Average stuff with elite command can anchor a rotation. When both are above average, you get sustained dominance.
Pairing with other metrics
Strikeout and walk rates quantify actual plate appearance outcomes. Use them to validate how much of the raw stuff translated to results. Ground-ball rate and barrel rate add contact detail. Run estimators like FIP and expected ERA connect contact, strikeouts, and walks to future run prevention. Combine these with Stuff+ to answer three questions: how nasty are the pitches, how well are they being located, and what should we expect next month.
Role and workload impact
Relievers can push Stuff+ higher because they throw at max effort and focus on their best shapes. Starters manage energy across six innings, carry a broader mix, and often pitch through fatigue. Expect a reliever to run 5 to 10 points higher than his starter version. When a pitcher moves roles, track whether velocity and shape change enough to alter the rating. Do not assume a starter’s Stuff+ will copy over to leverage relief without adjustment.
Environment and context
Most models normalize across the league, but day-to-day conditions still affect movement and velocity. Ballparks with different weather profiles can make ride or break play slightly differently game to game. Keep your focus on sustained shape, speed, and release rather than chasing single-game highs or lows. The model scale already accounts for league-wide run environments.
Data and transparency
Different providers produce different numbers because they weight inputs differently and train on different targets. One model might focus on whiff expectation. Another might tie closer to run value per pitch. The important rule is consistency. Use the same source for comparisons. Treat the gaps as perspective rather than contradiction. All are trying to answer the same core question through slightly different lenses.
Common mistakes to avoid
Do not treat Stuff+ as a ranking of current ERA skill. ERA bounces with context the model does not capture. Do not project big leaps from tiny samples or spring exhibitions. Do not ignore role stability, health, and schedule. Above all, do not skip the pitch-type detail. A great overall number can hide a weak link that opponents will attack in longer looks.
Checklist for evaluating a pitcher with Stuff+
Ask these questions
Which pitch drives the number? If it is one elite pitch and two weak ones, can the pitcher lean heavily on the strength without exposure?
Is the number stable? Did velocity, movement, or release change recently? If not, treat sudden jumps or dips as noise until you see confirmation.
Does a location or command metric support the story? High raw quality with poor command requires a different forecast than balanced skills.
What is the role and expected workload? Starters and relievers sit on different baselines. Multi-inning relievers often land between the two.
Is the schedule favorable? Even great stuff runs into lineups that do not chase. Context matters for short-term projections.
Why Stuff+ matters now
Hitter swing decisions and contact quality are better captured today than ever before. To pitch at the top of the zone and avoid barrels in a launch-angle world, you need either force or precision. Stuff+ tells you who brings the force. It helps front offices target arms, helps coaches design pitches, and helps fans and analysts sort noise from substance. In a data-rich era, it gives you a clean starting point.
Conclusion
Stuff+ distills pitch physics into a practical scale. It shines when used to flag future strikeout growth, to verify midseason changes in shape or velocity, and to guide pitch design. It does not score command, sequencing, or deception. It will not replace run prevention metrics. Use it with strikeout and walk rates, role, usage, and a location measure. If you do, you will read pitchers with more confidence, spot breakouts earlier, and avoid betting on small-sample mirages.
FAQ
Q: What is Stuff+?
A: Stuff+ is a model-based rating of the physical quality of a pitchers pitches, scaled so 100 is league average.
Q: What does a 110 Stuff+ mean?
A: A 110 Stuff+ means the pitchers pitch quality projects about 10 percent better than league average after removing location and context.
Q: Which inputs drive Stuff+ most?
A: Velocity, movement shape measured by induced vertical break and horizontal break, spin rate and spin efficiency, release point and extension, and pitch axis tend to drive most of the signal.
Q: Why can a pitcher have high Stuff+ but poor results?
A: Because Stuff+ does not grade command, sequencing, deception, or health, so walks, bad locations, or predictability can overwhelm raw pitch quality.
Q: How should fans use Stuff+ with other stats?
A: Pair it with strikeout and walk rates, pitch usage, role, and a location or command metric to build a balanced view instead of relying on one number.