For at least the first half of the season, a team should only scrimmage in rare instances and for very short periods of time. The problem with scrimmaging is that it’s totally uncontrolled. Sure, you can tell people what they should focus on, but as the points drag on, they’ll usually revert to their former habits. The whole point of practice is to control specific aspects of the game context, to work on those things specifically. This is all with the aim of performing Deliberate Practice.
Deliberate Practice is a scientifically supported method for improving at a skill. There’s a lot of interesting research on this, if you’re interested in an accessible read, try K. Anders Ericsson’s Peak: Secrets from the New Science of Expertise. I won’t go into detail about why this works, but the important thing to note is that it differs from regular practice in its requirement of tight feedback loops: do something, get feedback, do the same thing again taking feedback in mind, repeat.
Scrimmaging is about as far as you can get from Deliberate Practice. It’s uncontrolled and usually features players relying on whatever skills they already have rather than working on new ones. Of course, skill-specific drills are important and part of every team’s practice plan, and while those should also be informed by Deliberate Practice, they are not what I want to focus on right now.
At some point, a team needs to bring all the little individually practiced pieces together and play a cohesive game of ultimate. This is where the idea of scrimmage comes in. Scrimmaging is the game we all know and love in its purest form, but beyond that, we love it because it lets us do the things were good at–in other words it’s not uncomfortable–but discomfort is what makes us better. Discomfort is the reception of feedback that we have done something poorly and need to improve. If we are not uncomfortable, we will never improve. There is a caveat to this however. Being too uncomfortable can lead to frustration and negative attitudes. Frustration on the ultimate field does not come from making mistakes per se, but from making the same mistake many times. N-pull is a nice antidote because it lets the coach take a small actionable pattern and fit it into a context that the players are comfortable and confident with, while maintaining the focus that’s necessary for improvement. That comfortable context allows the players to get into a game flow that is conspicuously absent from focused drills. It also provides enough randomness in the game to avoid seeing the exact same situation over and over, which can be a precusor to frustration.
Another downside of scrimmaging is a false sense of accomplishment. It feels great to win a long point, to outlast your defender, to know that with every step you push yourself, you’re getting faster. But this all masks the fact that in tournament play, long points shouldn’t be seen as anything but bad. They drain your legs and they frustrate, especially when you don’t win them. I argue that at practice, long points are even worse. They group hundreds of actions together and potentially dozens of mistakes, and offer very little specific feedback, except vapid lines like “let’s be chilly”, “swing the disc”, “less hucks”, etc. Scrimmages also reinforce the mistakes that are made by not nipping them in the bud–“perfect practice makes perfect” is not just an empty adage, it’s a core tennet of Deliberate Practice.
The obvious alternative is n-pull, which is advantageous in all the arguments supra. In n-pull, the game is controlled. If the rest of the practice was focused on getting the disc off the sideline, then you can continue to practice that by just starting n-pull with the disc on the sideline. Obviously you can do this in scrimmage too, but that requires disrupting the flow of the game, the fact that n-pull already has clear stopping points makes these readjustments much more natural. Other examples of fine-grained game control abound.
This may seem counterintuitive because mistakes seem to be a bigger deal in n-pull, but n-pull actually encourages players to step outside of their comfort zone more than a scrimmage. This is because a player at practice already has an intrinsic motivation to improve (most of the time at least, and if they don’t then it doesn’t really matter how you practice anyway), but that will to improve only goes as far as a player is conscious of it. As long points wear on, players tend to go into survival mode and try to finish the point as quickly as possible which they do by reverting to their habits and the things they’re good at (or at least things their brains think there good at, which is what a habit is in a sense). Short points keep players fresh, and allow them to remain conscious of what they should be working on.
N-pull’s counter to the false sense of accomplishment is actually a double whammy. It keeps the perspective of players more accurate, because scores are valued exactly as much as they are worth–there is no false positive feedback after fighting your way back in a long point. Remember, the fact that you scored a long point is only very slightly better than having lost that point, it’s important not to lose sight of the fact that your team made enough mistakes to keep that point long. The second advantage is that when players finish practice and are not totally wiped, they’re more likely to keep thinking about ultimate, about what they can do to improve. They’ll talk about ultimate more, they’ll read ultiworld, they’ll watch videos, and they’ll make time to throw and workout. This desire to get better outside of practice, I believe, is the biggest determinent of how good a player will get, but that extra fire can be snuffed out by being too drained after practice.
All of these advantages of n-pull over scrimmaging come down to the fact that the feedback loop is short. If you turn the disc over, you have time to think about it and talk it out. It also gives your teammates time to think about if they could have done something different to avoid it as well. It gets the entire team thinking about the point as a whole, rather than just their active part in it.
One argument I’ve heard against n-pull is that it’s too short. Some players don’t touch the disc enough, have time to work out nerves, or get into a groove. The flaw in this argument is that it suggests that there’s something unrealistic in n-pull (at least as compared to full, tournament-play ultimate). If some players aren’t getting many touches in n-pull, they’re also probably not going to get enough touches in a full scrimmage. It’s true that in terms of pure cardinality, they may get more touches as a result of the points being longer, but by that logic, they could get equally many touches by just playing more n-pull points, and at least in that case, they will be fresh and practicing as the best version of themselves. W/r/t getting nerves out, this is not something afforded to a team in tournament play, so prevention for it should not, de facto, be built into the drill as a preventative measure. Only if nerves have proved to be an actual problem should they be addressed.
The truth about scrimmaging is not that it should never be done, but that it should only be done situationally. A great time for scrimmaging is in the final weeks of the season. For the team, having practiced in top form for a few months, it can become advantageous to prepare for long points. It doesn’t mean that long, grinding points are any less bad, only that they are an unavoidable truth in ultimate and with the series approaching, the importance of winning those points outweighs the purist, academic desire to play Good Ultimate. The value of practicing long points, however, only has value if you have previously set a high bar for quality that you not only aspire to, but have actually consistently played at before.