Answer in brief
A 2014 ESPN feature describes Curry's release at about 0.4 seconds from the start of his shooting motion; it links the speed to foot placement, a compact body alignment and an early release while still rising, though the measurement method is undisclosed and the catch-and-shoot version is documented far more thinly.
What the sources confirm about Curry's shot, and what stays off-camera
In April 2014, ESPN The Magazine published a detailed breakdown of one specific Stephen Curry three-pointer, mapped to the game clock of a game against the Clippers. David Fleming's longform remains the most detailed source on what happens before Curry's release, and this article is built on it: the piece was read in full, not summarized, and its central measurement was cross-checked against an independent mention of the same segment in a 2015 article by Grantland analyst Kirk Goldsberry. What follows honestly separates what is documented in detail — the shot after a dribble — from what open sources support only with performance statistics, not biomechanics — the shot after a pass.
The 0.4-second release: what ESPN's 2014 breakdown actually recorded
In the ESPN The Magazine longform about Curry's three-pointer against the Clippers, David Fleming writes that 0.4 seconds pass between the start of Curry's shooting motion and release, versus nearly 0.6 seconds for an average NBA player. The piece ran on 1-2 April 2014; it does not disclose how that pair of numbers was measured, and no methodology or lab is cited alongside them. The ESPN Sport Science project is mentioned once in the same longform, but for a different point: releasing while still rising, Sport Science found, gives Curry "19% more space inside the rim" for the ball's flight, a claim unrelated to the 0.4-versus-0.6 figures. Separately, in a June 2015 analysis for Grantland, Kirk Goldsberry mentions a beautiful Sports Science piece from the previous season showing that Curry gathers the ball after a dribble and releases in roughly 0.4 seconds, calling his release one of the fastest in the league. That definition is slightly different from Fleming's (from gathering after the dribble, not from the start of the shot), and the Sport Science video itself was not accessible while preparing this piece, so the 0.4-second figure is given on the strength of two independent journalistic accounts, not our own review of the underlying footage. None of the sources used here calls 0.4 seconds a certified NBA record; it is a journalistic measurement, not official league statistics. As of this writing, Curry remains an active player for the Golden State Warriors, wearing number 30.
The pull-up: the sequence that got mapped to the game clock
The most thoroughly documented shot is not a catch-and-shoot but a pull-up, a shot after one or more dribbles. That is exactly the sequence Fleming breaks down in ESPN The Magazine: a specific possession against the Clippers in which Curry receives the ball from Draymond Green, dribbles twice, then shoots, with key points in the sequence marked against the game clock at 6:35, 6:31, 6:30, 6:29 and 6:28. Splitting shots into catch-and-shoot and pull-up is not an editorial simplification: official NBA statistics separately track Catch and Shoot and Pull Up categories, and the tracking-data provider for that statistic is Second Spectrum, as confirmed by the shooting-stats page on nba.com. The asymmetry below is not hidden: a breakdown of Curry's catch-and-shoot with the same level of detail and attribution could not be found; that shot type gets its own section, built on performance statistics rather than biomechanics.
Foot placement: the last dribble step, not a separate motion
By Fleming's account, before he rises, Curry sets his feet softly and precisely toward the rim, an image the author compares to stepping carefully into slippers. This happens on the last dribble step, right at the three-point line, as Curry cancels his body's lateral and forward momentum. In our reading, since the feet are already turned toward the target by this point, there is little room left in the sequence for a separate, visible turn of the torso afterward, but the source itself does not state that conclusion; it is our interpretation, not a direct claim from the report.
The body compresses into one vertical line before the final bounce
The second element is the body locking into one vertical line. Fleming describes it this way: Curry compresses his right knee, hip, torso, elbow and wrist into a perfectly stacked vertical spring, and, per the source, this happens before the ball's final bounce, not right before release. That last dribble bounce gets a bit of extra force and immediately triggers the rise, leaving no pause between the end of the dribble and the start of the shooting motion. From there, per the same breakdown, the ball rises into the shot pocket just below the chest, the middle and index fingers, called the shooting fork, find the seam, and the right wrist cocks back until the skin on the wrist pulls taut. In our reading, it is the absence of a pause between the end of the dribble and the rise, rather than any single, unusually fast body part, that accounts for much of why the shot looks instantaneous; the source itself does not frame this as a separate conclusion.
Why the release happens before the top of the jump
The third element, and the one the source treats as the most unusual, is the timing of the release itself. For players with classic two-phase mechanics, Fleming's example is Ray Allen, he describes the process as more athletic and two-staged: the player rises to the peak first, then shoots on a flatter trajectory, around 45 degrees. For Curry, per the same breakdown, the phases merge: he releases the ball while still rising, his toes barely leaving the floor, which adds roughly 10 degrees of arc to the shot. The source describes this as one motion rather than several deliberate, sequential steps, not as the result of a faster hand; the report makes no such comparison.
The elbow and the guide hand: what stays constant under defensive pressure
Fleming identifies one element that does not change when a defender is close: the shooting elbow. It stays tucked near the ribs, without the sideways flare coaches call chicken-winging, and it is the integrity of that elbow, in the author's phrasing, that keeps Curry's shot accurate under pressure. Fleming writes that as a child Curry drilled this discipline lying in bed, tossing a balled-up sock toward the ceiling hundreds of times an evening and trying not to touch it; this is the journalist's own description, without quotation marks or a direct Curry quote, not a line from an interview. Separately, the source describes the non-shooting left hand: for nearly the entire motion it stays, in the author's words, paralyzed, serving only to hold and steady the ball, not to push or direct it. It opens not at the moment of release itself but slightly earlier, once the arms are already extended overhead in the elbow-at-eye-level position. The source ties defensive pressure explicitly only to the elbow; the author does not mention the guide hand in that context.
The catch-and-shoot: what league statistics record, and what we cannot claim
The NBA's official stats glossary confirms that Catch and Shoot and Pull Up are two separately tracked shot types, not a distinction invented by journalists. The glossary's own definitions are circular: Catch Shoot FG%, for instance, is described as the field-goal percentage on catch-and-shoot attempts, with no stated thresholds for distance or number of touches; Second Spectrum is named as the tracking-data provider for the category, as confirmed by a separate shooting-stats page on nba.com. A breakdown of Curry's technique specifically at the moment he receives the ball before a no-dribble shot, footwork and shoulder position on the catch, with comparable detail and attribution could not be found. But the statistical context exists: according to Kirk Goldsberry's reporting for Grantland (June 2015), in the 2014/15 season Curry made 136 no-dribble three-pointers at 47%, tied for ninth in the league by makes that season; league-wide, no-dribble threes went in at 37% versus 32% after a dribble, and 76% of all league three-point attempts in that period were no-dribble shots. These figures are about performance and frequency, not about Curry's catch mechanics, and they reflect the 2014/15 season, not the present one. The general coaching model for jump-shot preparation is described by the acronym BEEF, for balance, eyes, elbow, follow-through, but that is a general theory, not a Curry-specific breakdown.
A related but different case: what free-throw kinematics say about preparation
A separate line of research covers free-throw kinematics, not the jump shot and not Curry personally. A peer-reviewed 2021 study by Cabarkapa and colleagues, published in the Central European Journal of Sport Sciences and Medicine, compared the preparatory phase of the free throw between more and less accurate shooters in a sample of 17 recreational basketball players. A discriminant analysis using five kinematic variables correctly separated accurate from inaccurate shooters in 89.5% of cases; individual variables contributed unevenly: elbow angle accounted for 36.7%, forearm angle 23.9%, elbow height 19.2%, shoulder angle 6.8%, and knee flexion just 2.9%, which the authors describe outright as minimal or nearly absent. Separately, within the group of accurate shooters, the authors single out forearm angle, how close to vertical it is during the preparatory phase, as the key variable distinguishing makes from misses, and they attribute to it the same 23.9% of overall variance that appears in the discriminant model above; the authors do not report a separate variance share for makes versus misses specifically, only that the differences in forearm and shoulder angle were statistically significant (p<0.05). This study did not measure Curry, does not concern the jump shot, and was conducted on recreational players rather than NBA athletes; its findings cannot be applied to Curry's shot. It is included here only as an example of how this phase of the motion gets studied in concrete, numerical terms.
The limits of this analysis: one primary source, and no original video review
The limits of what is documented here should be stated plainly. The entire detailed movement breakdown and the 0.4-second figure trace back to one thoroughly read 2014 primary source; a second independent technical breakdown with its own separately measured figure, rather than a retelling, could not be found among sources located as of September 2026, and the closest related mention, Grantland in 2015, appears to describe the same ESPN Sport Science segment rather than an independent measurement. The methodology behind the 0.4-versus-0.6 figures is not disclosed in the accessible text. No review of match footage was carried out for this piece: everything described above comes from written, attributed reporting, not from the editors' own frame-by-frame analysis. An independent technical breakdown of a more recent Curry shot, or a source on the catch-and-shoot of comparable depth, would be reason to revisit this topic in a separate piece.
Practical checklist
- Watch where the feet point on the last dribble before the rise.
- Check whether there is a visible pause between the final dribble and the upward motion.
- Track the shooting elbow relative to the torso.
- Note when the guide hand opens relative to full arm extension.
- Check whether release happens while rising or only at the top of the jump.
Questions and answers
Is 0.4 seconds an official league record?
No. It is a figure from a journalistic breakdown by ESPN The Magazine (David Fleming, April 2014), not a certified NBA record; the text does not disclose the measurement method. The ESPN Sport Science segment that Grantland (2015) refers to describes a slightly differently defined interval, from gathering the ball after a dribble to release. The league's official stats glossary has no release-speed metric, though only the glossary was checked, not the league's entire statistical infrastructure.
What is the difference between a catch-and-shoot and a pull-up in NBA statistics?
A catch-and-shoot is a shot taken without a dribble right after receiving the ball; a pull-up comes after the shooter's own dribble has touched the floor at least once. The NBA tracks them as separate statistical categories, Catch and Shoot and Pull Up. Second Spectrum supplies the tracking data behind both categories, as stated on the nba.com shooting-stats page. The league's public glossary gives no exact threshold definitions by distance or number of touches.
Can an amateur train to the same release speed in a few sessions?
There is no basis for that claim, and this article makes no such promise. The described sequence, foot placement, compressing the body into one line, merging the rise and release phases, is the result of years of professional practice by one specific player; the sources used for this article contain no training program for reproducing that speed as an amateur.
Why doesn't the text give exact time splits for each movement?
Because the primary source used here does not break the motion down that way, and the editors did not conduct their own frame-by-frame video review for this piece. Precise intervals between sub-phases of the motion would require footage with a verified frame rate, a separate task not undertaken for this article.
What does chicken-winging mean?
It's a coaching term for when the shooting elbow flares out to the side during the rise instead of staying on a vertical line to the rim. In Fleming's breakdown, that flare is named as the main threat to accuracy, and keeping the elbow tucked is described as something Curry controls even under defensive pressure.