The best offense is no offense
The "be good at frisbee" theory of offensive 'systems'
When you think of the best frisbee offenses of the past 5-ish years, who do you think of? At least if you’re someone who makes frisbee content online, you probably think of Washington DC Truck Stop and Brown University Brownian Motion. To list a few examples:
Pod Practice: A look into the best offenses in the country — Brown BMo and DC Truck Stop
Matt Fazzalaro of Ultiworld: Truck Stop's Glorious Offense (paywalled)
Let’s discuss a somewhat unique aspect of these two offenses that the above analyses hint at but never quite fully appreciate: both of these teams claim they have no offense.
And I don't even mean that in some metaphorical or mystical way. They've both said it, quite literally and publicly. AJ Merriman went on the TOBU Fitness podcast in January 2024 and had this to say about Truck Stop's offense:
You could get all 8, 9, 10 guys that play offense for [Truck Stop] in here, and give them a whiteboard, and they could not teach you the Truck Stop offense. There is no "Truck Stop offense".
People always want to run some system or some cuts...they're, I think, looking at it the wrong way. It's very much a chemistry thing...it's just players knowing where each other are going, and knowing how to counter off of those things.
Brownian Motion coach Jake Smart was interviewed on Pod Practice in January 2025 and said this about BMo's offense:
Secret time: we don't have a scripted set. At all. Each group finds their own way to play. I didn't touch a whiteboard last year. It's not written down, there's no place it's written down.
Everyone hates whiteboards. Maybe you should, too.
But, seriously: can we stop for a moment and appreciate that two of the best offenses in frisbee both explicitly claim there's no offense? As I mentioned in a recent article: if you're going to copy what other teams are doing on offense, shouldn't you at least copy what the best teams are doing? How long is it going to take before mainstream frisbee culture fully embraces what Truck Stop and Brownian Motion are doing?
Be good at frisbee
How could it be possible that the best offense is no offense? I'd make two main points: first, teams underrate individual skills and overrate "systems".
In a post a few weeks ago, I led off my offensive theorizing by talking about individual skill:
Good offense comes from good frisbee players. The perfect horizontal stack cutting pattern won't make up for a team's lack of individual skills...In some ways, this bullet point is as important as all the others combined. You won't be a Nationals-level handler if you don't have a Nationals-level handler's throwing skills.
When I posted that article, I knew that I wanted to find some time to discuss that bullet point more thoroughly. Now is that time.
A coach emailed me a number of months ago and asked me about my offensive philosophy. My reply, in part, was:
I think one of the overall themes of my blog is that one of the best offensive strategies is to develop players who have better skills, better field sense, better everything... It's better to have smart, skilled, athletic players running a bad offense than having weaker players running a "theoretically correct" offense. So, my offensive strategy is individual skill development? ... I think that aside from a few of the really elite teams, there's a lot of advantage to be gained by motivating players to step up their individual game.
Another (national-championship-winning) coach told me this recently in an email discussion:
In general, I'm not a fan of most "3rd party analyst" frisbee articles I come across. Mostly because they often project waaay more complexity or thought into the actions being taken relative to what actually is going through the minds of the team they are commenting on.
...I know first hand that people usually assume waaay more complexity or forethought into much of what they're seeing. More often than not, Occam’s Razor applies... e.g. they threw that one pass simply because it was the first person they saw open, and that's it.
Good frisbee players aren't wasting time on the field over-thinking how the system works. They just see open spaces and cut into them, see open teammates and throw to them, see poaches and work around them, see opportunities and take them. And the more time your team spends diagramming the system, the less time players have to get the reps and the feedback required to build those intuitions.
To put it another way: I’d be curious to watch a frisbee game and track how many turnovers happen “because of the system”, and how many happen because of individual skill deficits—whether inaccurate throws, bad decision-making, or simply not having the field sense and throws required to get oneself out of a tight situation. I don’t have that data for you today, but I personally would wager that imperfect skills cause turnovers more often than imperfect systems.
Play random
The second reason no offense is good offense: predictability is bad for an offense (as I’ve said seemingly 1000 times). Effective modern offenses embrace spontaneity and an ability to read-and-react to the defense in real time.
In the NBA, former Milwaukee Bucks coach Mike Budenholzer famously told his team to "play random".
The YouTube video's title makes light of his comments, calling it an "unmotivating speech". But the Bucks won the NBA championship a few games later—play random is serious business. A great Reddit thread titled In defense of "play random" fills in some of the background knowledge:
In 2007, Jeff Van Gundy and first-year assistant general manager Daryl Morey collaborated on a joint project to mold Morey’s data-driven approach with Van Gundy’s seasoned pragmatism...Two-thirds of the way through the project, Morey discovered that the play at the top of the list was unfamiliar to him. Confused, he asked Van Gundy what it was called.
[Van Gundy said] "It’s called random. That’s when the play breaks down, and we just set a random screen." [Morey replied], "Well, that might tell us something right there. Our best play is ‘Random.’"
Neither man believed that “something” was to have no offensive structure at all. Still, the data was sending a powerful message with wide-ranging implications that few grasped in 2007, more understood in 2014, and almost everyone in the game takes for granted in 2022. The best way to score isn’t to diagram an elaborate sequence that beats a set defense. It’s to sew and exploit chaos. To play instead of calling a play. To be “random.”
A comment later in the thread points out the Boston Celtics—who won a championship themselves a few years later with a league-best offense—play a similar way:
Tatum described Boston’s offense as randomness, built on trust and reads, rather than any calls from the sideline. Ime Udoka focused on letting the Celtics play free-flowing basketball last season by trying to let his stars become natural playmakers.
Ultimate star Manuela Cardenas—a world- and US national- champion—also says in an interview on Pod Practice that her philosophy is to "just play frisbee":
Everyone knows horizontal...everyone knows vertical...now [b/c those systems are so familiar] it's kind of easy to stop those setups, so we just gotta get the frisbee, and play, and flow.
So, to recap: The two (three if you're including Manu) best frisbee offenses claim they have no offense. Two of the last four NBA champions won with offenses based on "randomness". All four emphasize reading the defense and trusting/knowing your teammates.
What do you do at practice?
One reason I wanted to write this article is that, from what I've seen, a number of club teams spend much of their limited practice time on "teaching the system".
But to me, that's always felt wrong. For almost every team in the club division, the reason they don't make Nationals isn't because they don't have a Nationals-level system—it's because they don't have Nationals-level players. Sectionals teams that miss Regionals, by and large, are missing it because they don't have Regionals-level players. Not because their cutting systems aren’t crisp enough. Frisbee players spend so much time learning the system that they don't get the chance to develop their individual abilities and chemistry.
In his interview, Jake Smart mentions that his team:
plays a lot of 3v3 at practice
and engages in honest communication about what is and isn't working
How many other teams are playing 3v3 extensively at practice? From what I've heard, not many. One reason BMo succeeds is they simply spend so much time honing their skills, calibrating their knowledge of their own skills, and learning how to work together with their teammates.
Other teams feel like they "don't have time" to work on individual skills at practice, and expect players to build skills on their own time. But, in my opinion, that's completely backwards: you don't have time not to develop your players' individual skills.
And to be clear, I'm not saying there's anything wrong with the other analyses I linked above. Even if the teams themselves don't have a system written down, there can still be consistent patterns in their play that emerge on film. (I'm also not sure I fully believe these teams have literally nothing written down.)
But, even given that these patterns exist, if you copy their patterns, you're missing the point (at least partly). You should copy their philosophy of developing good frisbee players. Players who are good at playing frisbee together—and good at the problem-solving required to get good at playing frisbee together.
Worry less about offensive systems. Develop individual skills—be good at frisbee. Embrace and exploit chaos. Build skill, build chemistry, then play "read-and-react random".
This is directionally right, and definitely expresses the new meta -- the buzz around town -- about offense. But it's overstated IMO because the correct point isn't about lack of models, it's about pedagogy. Players don't learn well through chalk talks, they learn through motion, through experience, through trial and error, and through the formulation of explanations around what works and what doesn't and why. They learn by playing together and deriving mental models together, and this creates a sophisticated shared understanding.
That is why it is possible to sketch out the "rules" (though they are highly contextual) of, say, Truck's offense. (Tbh no one has done this publicly... the ultiworld articles and ultiworld discord posts are not really sufficient... they don't get to the core of matter IMO... but you *can* do it.) But players don't learn by reading rules, they learn by deriving them.
What's maybe most important is that self-derived rules will be extremely well-adapted to the particular abilities of the players on the team, because they're discovered through trial and error, and that process necessarily will account for the particular skills and dispositions of the players on the team.
I think some of the comments below and your response to them get at all of this - writing this out as my first reaction on reading the post.
This might be my all-time favorite post of yours. It's definitely a point I agree with (it's basically the theme of "the ultimate resource"!) but "the best offense is no offense" is such a beautifully dramatic way to put it, and the NBA connections are super interesting and new to me.
I think I would just want to reinforce that a huge and essential component of the "individual skills" teams should probably be focusing more on (instead of scripted "systems") are not physical, but mental/cognitive. Players need to learn (through trial-and-error and with the help of constructive feedback from coaches) to face and look the right direction, see the right things, see more things, read the game, recognize opportunities to interact with teammates, recognize opportunities to stay out of teammates' way, and generally make good decisions. So about 90% of practice time should be spent on small-sided, not-fully-scripted activities (like the 3v3 you mention, and of course I also love keepaway variants) where players are actually practicing and developing these crucial cognitive skills (along with, of course, their physical skills). If practice time usually looks like that, then "just go play good frisbee" is 100% sufficient as an offensive system. Whereas if practice time is instead 90% scripted drills (working on physical skills, but with no decision-making element), forget about it. "Just go play" will be horribly inadequate because the players won't actually know how to play... and no "system" will save them.
Anyway, thanks for the spectacular post!