Why your volleyball practices should be as game-like as possible

This article is a free sample of the sort of longer-form content available to patrons who sign up at the Avid Fan level or higher.

Note: If you prefer this via PDF, download the attached file (link at the bottom).

There was a women’s volleyball coach symposium in December 2018 in Buenos Aires. Jose Benini from Boca Juniors invited me to attend, but unfortunately I could not. Apparently, I missed a big debate about which type of training approach is better – one that works more on skills in isolation or one that is more game-like. 

This is not a new debate. Very experienced and successful coaches have argued on each side for many years. Jose asked me to share my perspective on the topic. I will do that below, but first I need to take some time to share the science.

Motor learning

On one end of the training spectrum is what’s referred to as “blocked” exercises. This is where the player does the same thing over and over with little variation. Think about a hitter attacking from a toss, or a defender digging balls hit at them by a coach. Everything is very controlled and very isolated. The player need only focus on executing the specific skill. Nothing else.

At the other end of the spectrum is “random” practice (you may also see it called “distributed”). This is where the player performs a variety of skills with no repetition of pattern. Think of how they work in a game. They perform all the skills in an ever-changing sequence.

Which is best for developing skills?

According to research into motor learning – which is the neurological process of developing physical skills – random training is superior to blocked training. Here is a graph I saw during a presentation by current Dutch Women’s National Team coach Jamie Morrison at a USA Volleyball High Performance Coaches Clinic in 2015 (Jamie was the USA assistant at the time). It shows a comparison of training in the two different methods.


There are three parts to the graph. The first is an initial test (Pre-Test). Then there is the training period (Acquisition Tests). Finally, there is a competition period (Transfer Test). The blue line is the blocked training group, while the yellow line is the random training group (white is the control group). Notice how the blocked group is better during the first two stages, but then drops very sharply in the final stage while the random group continues to rise the whole time. This is a function of transfer, which is a key concept in motor learning. 

Basically, the reason we see the blue line drop while the yellow line continues to rise is that there is far less transfer from blocked training to playing than from random training. Have you ever seen a team that looks very good in training but doesn’t look so good when they actually play? This is because their training isn’t producing very much transfer.

Let’s look closer at this idea of transfer.

Transfer and specificity

Transfer is the degree to which what we train then is applied when we play. The amount of transfer that happens is based on the specificity of our training. The closer we get to training a skill the way it will be used in competition, the more transfer we get.

Specificity is an important concept to understand because motor pathways (the patterns our brain develops to execute a skill) are very specific. Two things that look the same can have very different motor pathways. 

The late Carl McGown is probably the person most responsible for bringing the concept of specificity of training into our sport thanks to his 40+ years working with the USA national team and his years teaching those concepts to other coaches. In the book Volleyball Coaching Wizards – Wizard Wisdom there is a quote from Carl that helps explain this specificity concept. 

“Hitting balls off of a tee and hitting a moving baseball are grossly very much alike, but the neurological commands that you need to put in place to do hitting a ball off a tee are just not the same as hitting a baseball. We would expect transfer to be very, very small.”

Why do you think that might be?

Read – Plan - Execute

Consider the difference between the two skills. Hitting a ball off a tee means striking something that is fixed in space. Hitting a pitched ball is much more complex. It involves timing the incoming ball – judging when to swing - and also recognizing its movement – judging where to swing. In other words, hitting a ball off a tee is simply about skill execution – swing at the ball. Hitting a pitched ball, however, requires the batter to read the speed and direction (and possibly curve) of the ball, and also plan their swing (timing and location), as well as executing the swing.

Because hitting off a pitcher involves the read and planning elements it requires a more complex motor pathway. You could think of it as three pathways merged into one – time the pitch based on its speed, judge its future location based on its path, and execute the swing. Compare that to having to execute each of those individual pathways separately. Sounds less efficient, right?

Of course, the Read-Plan-Execute sequence is not something we do consciously. In fact, our training is intended to make it all automatic and unconscious. As we train a skill we alter brain function to make the motor pathway for that skill more efficient. That’s why we want to train the whole skill (swinging at a pitched ball). If we only train part of the skill (swinging at a ball on a tee), then while that particular motor pathway might become more efficient, it’s separate from the other elements of the whole skill and how they work together. Thus, the whole skill is not developed nearly as effectively. This is why you will sometimes hear that “whole” training is better than “part” training.

Bringing it to volleyball

Let’s make a volleyball comparison.

Machines are sometimes used to “serve” balls to receivers. Normally, it is the machine sending balls to a single receiver (even if there are multiple receivers on the court, they usually know where the ball is going). Let’s compare that to players receiving a real serve in a 3-person reception.

The machine usually is pretty consistent in direction and speed. There may be small differences, but the player knows basically where the ball will go and how fast it will arrive. They simply have to put their passing platform in the correct position. There is thus a small amount of read and plan, but it is much more about execution.

In the case of 3-person reception against a server it is much more complex. The receiver reads the ball’s direction and path from the server’s contact (and pre-contact movement). This is much more reading because there is far more variability with a server than a machine. Then, based on the ball path read, they must judge if it is their ball or whether to let their teammate take it. That’s is a second read. If it’s theirs, they have to determine how to play the ball (plan). Then they execute.

The part training element of blocked training is why the graph above shows a drop when the blocked training group moves into the competitive phase. They have not developed an efficient pathway for the full skill as it is used in game situations. They’ve learned the physical execution of the skill, but not the reading or planning elements – at least not nearly to the same degree as did the random group. The lack of specificity of training results in very limited transfer.

The bottom line is that training in a game situation is superior to isolated skill training as it leads to better transfer from practice to competition. This is where the concept of “the game teaches the game” comes from.

But we need to give them lots of repetitions

Yes, players need repetitions to learn and develop. Not all repetitions have the same value, though. Game-like repetitions have a higher value when it comes to making our players better at playing the game than blocked repetitions. That means you need fewer game-like repetitions to get the same training impact. And I bet you can get just as many reps in game-like situations as you would probably do in a blocked drill, especially if you set things up to encourage it.

Do this exercise some time. Have your team play a winners-stay-on game (often called King or Queen of the Court). Usually there is a serve to start each rally. I want you to count how many reception attempts there are in the game (so total serves minus service errors). If you want to go even deeper, count how many attempts each of your passers get. Let the game go on for the same amount of time you would normally spend working on reception in a more blocked exercise. Then compare the totals between the two different activities.

My guess is you’ll find get totals closer than you might have thought. It’s easy for us to forget to count repetitions that happen in games as repetitions, but that’s exactly what they are. 

Now, add on to that a couple of extra considerations. First is obviously that the game repetitions produce more transfer than the blocked ones, as outlined above. Second, and perhaps equally importantly, the reception attempts in the game are probably more challenging than they would be in a blocked drill. Why? Because servers in blocked passing drills are generally just trying to make it go as quickly and easily as possible.

Let me add a final thing to think about on this subject of repetitions. During blocked drills the focus tends to be on just one or a small number of players. There may be others involved – servers and targets in a serve reception drill, for example – but they are usually just there to support the ones actually doing the learning. In the game, though, many more players are involved. They are getting good repetitions in more than just one skill.

Keeping to the winners-stay-on example, obviously all the skills of the game are used, but let’s just think about the serving. Because each rally starts with a serve that means the players are practicing serves that are meant to help their team win the rally at the same time the receivers are working on passing those types of serves. The game therefore is doing double duty when just considering those two skills. That’s more efficient for you as a coach because it means you can get more training in the same amount of time.

But we can’t just play!

The most common argument against fully random training is that we cannot possibly just play all the time in training. This is obviously true in the case where we don’t have enough players to play, such as when doing individual lessons. This is something I will circle back to later. For now, though, I want to address the situation where we actually do have enough players.

Yes, we really can just play – sort of

The first thing I will say is that we actually can just play games all training long. I don’t mean simply rolling the balls out and letting the players play. Also, I’m not saying they can only play 6 v 6. Let me explain.

First, our most important job during training is providing the players with feedback. The argument many coaches who favor more isolated skill (blocked) training make is that they cannot provide feedback during games. This isn’t actually true, but it does take more concentration and focus. 

Most coaches have no problem giving feedback in a blocked drill because there is a natural stop after each repetition to do so and it’s very easy to focus. Many don’t do as well when the players are working in a game for a couple of reasons. One is because there is no natural stop after each repetition. Another is because those repetitions are more randomly spaced. Further, we tend to lose focus during games. Instead of concentrating on one skill we get distracted by other things. Rather than just looking at attacker approach footwork, for example, we find ourselves looking at reception, or setting, or any number of other things.

But we can – and should – stop a game when we see a problem with the thing we are focusing on. There’s no rule that says we have to let them keep playing without providing feedback and addressing the issue. And we can definitely train ourselves to maintain the proper focus. So from that perspective we can absolutely train in games.

Second, games need not only be 6 v 6. We can, and should, use small-sided games in our training. Any game that includes the thing we want to work on will serve our purpose. While we may not be able to work on full offensive or blocking tactics while playing 4 v 4, for example, we can work on most individual skills as they are all used. And they are also useful in working on specific situations, such as OH/MB vs. OPP/MB (essentially left side vs. right side).

The point is that small-sided games offer lots of opportunities to work on skills in a very game centered way. They have the advantage of usually providing the players with more skill repetitions than full 6 v 6 games do as well, so more opportunities for feedback.

Also keep in mind that you need not use standard point-per-rally scoring in your games. You can easily set the scoring to encourage something specific, or to put more focus on what you’re working on. Maybe you give bonus points for running a middle attack. If you’re working on good attack approaches, you could give points for every correct one, regardless of the outcome of the attack. Having these different types of scoring systems not only can help encourage players to do what you want, they can help you focus better on what you should be watching.

Third, you can make adjustments to games to get more repetitions of whatever you want to work on. Let’s say you want your right side hitter (Position 2) getting a lot of attacks. You can make a rule that says the first set of each rally must go to them. This might mean you have to change how you initiate each rally, though, depending on your target skill or tactic. For example, it takes good passing to execute a quick attack, so you may need to start each rally with a free ball, an easy serve, or a controlled attack to get more first contacts to target. 

But we probably shouldn’t ONLY train in games

Having said that it is entirely possible to only train in games, let me now suggest that’s probably not the best way to go. Even those who are the strongest advocates of random training acknowledge that some limited amount of blocked training is valuable. Even Carl McGown, who was very strong in his support of game-like training, admitted that blocked training had value.

Notice, however, the emphasis on “limited” there. Basically, the idea is that we use blocked training to introduce something, or perhaps to reinforce it. The players go through a small number of repetitions so we know they understand what is expected of them. Once they have that understanding, we move on to the more random training – while keeping our focus on that thing we’re developing.

Here’s an example. You’re teaching a group of players proper attacking footwork. You start with them doing several approaches without a ball to make sure they have it right. Once they do, you immediately move to hitting a live ball. The key here, though, is that you keep a focus on the footwork and don’t worry about bad timing, hitting the ball in the net, etc. They are executing the whole skill – attacking the ball – as the motor learning research suggests, but you are focusing your feedback on only one part of it.

As I suggested above, you can do the same thing in terms of reinforcement of what you’ve been working on. Have the players do a handful of approach repetitions, perhaps as part of their warm-up process, then jump right in to hitting a live ball.

Yes, new players as well

It might be easy to think in terms of making things more game-like for players with some experience. After all, they can probably play the game, and you probably already incorporate games into your training. The big pushback against game-like training usually comes from coaches working with newer players – players who don’t have the basic skills yet.

The science is the science, though. It applies equally to everyone.

I would actually argue that making things game-like for beginners could be even more important than for experience players. Why? Because they have no context. When you work with an experienced player on a skill, they can envision how it will be used in a game. A beginner cannot, so they don’t really know the overall objective of what you’re teaching them.

Let’s use passing as an example. You might do something like toss a ball to a player and have them pass it back to you. Then you may have them pass back and forth between two players. But when will a player ever pass the ball back where it came from in a game? Basically never.

Yes, you can use a simple linear passing exercise to introduce the basic idea, but that’s as far as it should go. The players do a handful of repetitions to show that they understand the skill, then you make things more realistic. Have them pass to a target. Then, when you see they understand that idea, have the target pass the ball back to them so they can then pass the third contact over the net. 

Look! They are playing volleyball!

Will it be ugly? Of course! That’s what you want, though. It will gradually get back if you keep them focused on what you want them to learn. If it all looks nice and neat then the players are not being challenged enough. Karch Kiraly once said about training the USA Women that he looks for them to only be successful about two repetitions out of three. That means there are lots of mistakes in the USA gym!

Making things more game-like with only one or a few players

Sometimes we don’t have enough players to actually play. We still want things to be as game-like as we can get them, though, to produce the maximum possible amount of transfer. This is accomplished by ensuring real game elements are included in each repetition.

Here’s an example of how we can make a highly blocked exercise much more game like.

Let’s say you’re doing setter training. The simplest thing to do is to stand in the middle of the court and toss balls to a setter who is standing in the target position. This is extremely isolated skill training. How can we make that more game-like? Well, we can vary where our tosses come from. We can vary where the tosses go. We can have the player come from Position 1 or block first. Combining toss variability and pre-set movement makes things vastly more game-like than simply tossing from one fixed spot to another. Even better, you can add in an actual passer or digger to make anticipating the ball much more realistic. And, of course, you can make the setter do a hitter coverage move after they set to add yet another element.

Adding the things the player would normally do before and after the focus skill is something you can do when training any position to make things more game-like.

Adjusting the game

The reverse of making things more game-like is actually making them slightly less game-like. In a real game rallies always start with a serve. There’s a big random factor to that, though. There are service errors. There are aces and bad passes. Sometimes there are specific things we want to work on that are best accomplished with a more controlled first ball. For example, we might want the first set to go to our middle hitter. Very hard to do if you don’t get many balls going to target. So rather than beginning with a serve, we can start with a free ball or a controlled coach’s attack.

You can also force a certain kind of second contact, either directly or indirectly. For example, if you want to work on non-setters taking the second ball you can set up your game, or game-like activity, such that the setter has to take the first ball most or all of the time. You can hit the ball to them to start the rally as one option. Even better is having the opposing team always attempt to attack the ball at the setter. That way the setter is forced to play the sorts of balls they would get in games.

Keep in mind when you’re thinking about these sorts of options, though, that you still want to keep things as close to real game situations as possible. And very importantly, you want a strong random element to whatever it is you’re focusing on. For example, if you’re working on team serve reception you don’t want to toss the ball over the net. That’s a long way from realistic.

Final thoughts

The bottom line is that training in the most game-like fashion you can is strongly favored over training skills in isolation. The latter simply does not produce as much transfer. This is a major consideration when you have limited training time, as is the case for most of us. We need to be as efficient as we can, and that means getting the most out of every repetition. There is room for so-called blocked training, but only in a limited fashion to ensure the players understand the expectations. Everything else should be as game-like as you can possibly make it. Just make sure you incorporate lots of feedback in those game-like activities.