Twenty fastballs, then twenty curveballs, with a few changeups at the end. It is the default bullpen on most teams, and it feels productive: each pitch gets a chance to settle in, and by the end of each block the pitcher is usually hitting spots.
Motor learning research has studied this exact question for almost fifty years under the name contextual interference. The findings are useful for pitchers, with a twist that matters a lot for younger players.
What is blocked vs random practice?
Blocked practice means repeating one skill many times before moving to the next. In pitching, that is a set of fastballs followed by a set of sliders. Random practice means mixing skills so the next rep is rarely the same as the last. In pitching, the catcher calls pitch type (and often location) for every throw, as in a game. Serial practice sits in between, cycling through pitches in a fixed order.
What the classic research found
In 1979, Shea and Morgan had 72 college students learn three different movement patterns, either in blocks or in random order [1]. During practice, the blocked group looked better. On the retention tests, ten minutes and ten days later, the random group performed better, and it also did better on transfer to new versions of the task. This pattern of worse practice performance and better learning became one of the most cited findings in motor learning.
Baseball has its own version. In a six week study of 30 college hitters, a group that faced fastballs, curveballs and changeups in random order improved 56.7% on a game-like test, compared with 24.8% for a group that saw each pitch type in blocks [2]. Our post on why pitchers throw strikes in the bullpen but not in games covers that study and the learning vs performance gap in more depth.
Is there research on pitchers?
One study tested this directly with pitchers. Tsutsui, Satoh and Yamamoto had 20 high school and college pitchers (ages 15 to 22) throw straight balls and breaking balls at a 40 by 40 cm target from a mound [3]. Each pitcher threw 40 practice pitches, 20 of each type, in either blocked or random order. During practice and on an immediate test, the groups were the same.
The next day, on a retention test, the random group had better control. The effect came from the higher skill pitchers. Among the lower skill pitchers, practice order made no difference. The study had only five pitchers in each subgroup and a single practice session, so it is a small piece of evidence. It does line up with theory, as you will see below.
Why mixing pitches helps learning
Researchers have offered a couple of explanations, and both make intuitive sense on a mound [4]. In a blocked set, the plan for the next pitch is already loaded. You copy the last throw with small tweaks. In a random set, every pitch forces you to rebuild the plan: pick the pitch, find the grip, find the release and then read the result. A game asks for exactly that on every pitch. Mixing also lets you compare pitches back to back, which may sharpen the feel for what makes each one different.
How big is the benefit, really?
This is where honest coaching matters. Brady's meta-analysis of 61 studies found an average random practice advantage of 0.57 in lab studies and only 0.19 in applied sport studies, roughly three times smaller [5]. The advantage was also much smaller in younger learners. Czyż and colleagues' 2024 meta-analysis of 54 studies confirmed a retention benefit overall, while finding it almost negligible in applied settings and not significant in young participants [6]. A 2023 meta-analysis of sport studies by Ammar and colleagues found small, non-significant effects, with age influencing the results, though its search methods have drawn criticism [7].
The fair summary: random practice is a good bet for experienced pitchers working on pitches they can already throw. Expect a modest edge. For a pitcher who cannot yet find the zone, other fixes come first.
Plan the work, then do it
Put bullpens and Command days on a schedule
Pitch AI schedules your bullpens and Command days, so the work actually happens. Harder command work gets a set place in your week, alongside the rest of your throwing.
How to order your bullpen by stage
The challenge point framework explains why skill level matters so much [8]. Random practice adds difficulty. For a pitcher who already owns a pitch, that difficulty creates useful learning. For a pitcher who is still figuring a pitch out, the task is already hard, and adding more simply overwhelms. So match the order to where each pitch is:
| Where the pitch is | Bullpen order | What it looks like |
|---|---|---|
| Brand new grip | Mostly blocked | Short sets of the new pitch, with feedback on shape after each set |
| Lands for a strike about half the time | Small mixed sets | Pairs or triplets, such as fastball, new pitch, fastball, with the catcher calling the order |
| Game-ready | Fully random | Catcher calls pitch and location every throw, including in hitter's counts |
Driveline has used a similar progression in its weighted ball command days. In Hacking the Kinetic Chain, "Command A" days use a structured four sets of five throws so athletes can adjust to each ball, while "Command B" days present the same implements in random order, forcing an adjustment on every throw [9]. That is a program design, untested in a published study, and it mirrors the research well.
For pitches still in development, our pitch design workflow covers how to test grips and decide when a pitch is ready.
A sample random bullpen
For an experienced pitcher with three reliable pitches: 35 pitches. The catcher signals pitch type and location before each throw. No pitch type more than two in a row (Tsutsui's study allowed no more than three [3]). Start with a few from the stretch, finish with a few in two-strike counts. Chart miss distance. The next day, throw ten random pitches at the start of your session as a retention check.
When blocked practice is the right call
Blocked practice has its own uses. It builds early performance faster, which matters in at least four situations:
- A brand new pitch. Find the shape before asking it to compete with your other pitches.
- Young or very wild pitchers. The research benefit is smallest here, and early success keeps kids engaged. Our guide to youth pitcher development covers the broader picture.
- Returning from time off. Rebuild feel with simple sets before adding complexity.
- Right before a start. On the Rubicon podcast, the host separates learning days from compete days and argues that a short blocked pen before a start is fine because that day is about confidence [10]. That is practitioner opinion, and it matches the PoST skill periodization framework, which recommends simpler, confidence-building work close to competition [11].
Gray makes a similar point about matching variability to the athlete: the more variable a pitcher already is, the less extra variety they need [4]. For more on the science of variability, read part 2 of our skill series on dexterity in pitching, and continue this series with how pitchers train to perform under pressure. If you want a ready-made base, the free 4-week throwing and command program is a good place to start.
Frequently asked questions
Should pitchers throw random or blocked bullpens?
It depends on the pitch and the pitcher. Experienced pitchers working on pitches they can already throw for strikes tend to benefit from random bullpens. Brand new pitches, young pitchers and very wild pitchers usually do better with blocked or lightly mixed work until the pitch is reliable.
What is contextual interference?
Contextual interference is the finding that mixing different skills during practice makes practice performance worse while often improving retention and transfer. It is the scientific term behind the random vs blocked practice debate.
Is there a study on random practice for pitchers?
Yes. In a study of 20 high school and college pitchers, random practice of fastballs and breaking balls led to better control the next day, but only among higher skill pitchers. For lower skill pitchers, order made no difference. The study was small and covered one session.
Should I throw a blocked bullpen before a start?
Many coaches say yes. A short, simple blocked pen before a start is aimed at confidence and readiness. Save random, harder bullpens for learning days earlier in the week.
A bullpen for every stage
Let the plan carry the structure
Pitch AI schedules your bullpens and Command days, so the work actually happens, and Pitch Lab logs each pitch's shape so you can tell when a new pitch is ready for random work.
Sources and further reading
- [1] Shea JB, Morgan RL (1979). Contextual interference effects on the acquisition, retention, and transfer of a motor skill. Journal of Experimental Psychology: Human Learning and Memory, 5(2), 179-187. https://doi.org/10.1037/0278-7393.5.2.179
- [2] Hall KG, Domingues DA, Cavazos R (1994). Contextual interference effects with skilled baseball players. Perceptual and Motor Skills, 78(3), 835-841. https://doi.org/10.1177/003151259407800331
- [3] Tsutsui S, Satoh M, Yamamoto K (2013). Contextual interference modulated by pitcher skill level. International Journal of Sport and Health Science, 11, 68-75. https://doi.org/10.5432/ijshs.201211
- [4] Gray R (2021). How We Learn to Move: A Revolution in the Way We Coach and Practice Sports Skills. Perception Action Consulting and Education (chapter 8).
- [5] Brady F (2004). Contextual interference: a meta-analytic study. Perceptual and Motor Skills, 99(1), 116-126. https://doi.org/10.2466/pms.99.1.116-126
- [6] Czyż SH, Wójcik AM, Solarská P, Kiper P (2024). High contextual interference improves retention in motor learning: systematic review and meta-analysis. Scientific Reports, 14, 15974. https://doi.org/10.1038/s41598-024-65753-3
- [7] Ammar A, Trabelsi K, Boujelbane MA, et al. (2023). The myth of contextual interference learning benefit in sports practice: a systematic review and meta-analysis. Educational Research Review, 39, 100537. https://doi.org/10.1016/j.edurev.2023.100537
- [8] Guadagnoli MA, Lee TD (2004). Challenge point: a framework for conceptualizing the effects of various practice conditions in motor learning. Journal of Motor Behavior, 36(2), 212-224. https://doi.org/10.3200/JMBR.36.2.212-224
- [9] Boddy K, Daniels M (2017). Hacking the Kinetic Chain: Advanced Pitching. Driveline Baseball (Weighted-Ball Command A and B days, p. 169).
- [10] Rubicon (2026). Casting The Die, Episode 2, with Steffen Simmons (podcast; practitioner opinion). https://youtu.be/PkRCz3dhYn0
- [11] Otte FW, Millar SK, Klatt S (2019). Skill training periodization in "specialist" sports coaching: an introduction of the "PoST" framework for skill development. Frontiers in Sports and Active Living, 1, 61. https://doi.org/10.3389/fspor.2019.00061