Skip to main content
Menu
A Plea For More Frisbee Data From A U.S. Ultimate Coach

It’s midway through the first half against Great Britain and they are scoring too easily. As a coach of the U.S. under-23 men’s Ultimate Frisbee team, I’d scouted the British team in an earlier game. Now, after watching the first 5 points of our semifinal match at this summer’s world championships, my fellow coaches and I gathered the team’s defensive line in a huddle: Their handlers are really comfortable throwing the around breaks, so let’s shift our marks to the backfield and make them throw the inside flick to a tight window. (For the 99 percent of you who didn’t get that: Basically, let’s position our defense in such a way that the only option for their throwers is a difficult forehand throw to a well-covered receiver.)

The strategy happened to work: Deprived of easy, short passes, the British team began to take riskier and riskier long throws. Eventually, the percentages tilted in our favor, and we won by 3 points. We were on our way to a gold medal.

That strategy, though, was basically put together on improvisation and a hunch. As my colleague Carl Bialik writes elsewhere on FiveThirtyEight, there isn’t much data in ultimate to help coaches like me.

I can imagine a day, maybe at the 2025 world championships, when a brilliant coaching insight1 would emerge from a data set, displayed in a crisp chart on my iPad 12 (hologram edition). It would show which spots on the field the other team’s main throwers have trouble completing passes to. Another would reveal, say, that one of our players has a much higher completion rate along the forehand side than the backhand side. “Let’s run plays to that side of the field for him,” we, the brilliant coaches, would say.

But having more data in my coaching arsenal is only the first step. Players don’t always absorb data-driven feedback easily.

This isn’t unique to ultimate. Even in basketball, the players like Shane Battier who explicitly embrace analytics and can probably speak the language of usage rate and points per possession are rare.

So coaches have to adjust their language. Players can better process “he doesn’t want to go right” than execute on “when positioned within 3 feet of the left elbow, your matchup’s shooting percentage decreases by 32 percentage points.” When I was playing top-level ultimate, I was much more effective on the field when the only thing running through my head was “screw these guys — let’s beat ’em.”

But the job of a coach is to bridge the gap between a player’s lizard brain and the stat sheet. I’d love the challenge of having to translate analytics into simple language that gets my players out-performing, not over-thinking.

Screen Shot 2015-12-15 at 6.47.53 AM

The few dabbles with this kind of data are intriguing. The chart next to this paragraph, which was put together by Sean Childers and is also in Carl’s piece, shows the chances of scoring from different parts of the field. It’s no surprise that the closer you are to the end zone, the better your chances of scoring are. But see that bow in the 40 percent range? Sean’s data — limited as it is — suggests that a team is just as likely to score from 50 yards outside the end zone in the middle of the field as it is from the sidelines 35 yards away. That’s actionable intelligence and confirms what I’ve gathered from years of experience — when you’re stuck on the sideline, bad things happen. (That’s why my advice to our defense against Great Britain was so focused on sideline strategy.)

I would love to have other hunches of mine confirmed or disproved in this way. Am I justified in thinking that passes that “break the mark,”2 no matter how small, have a cumulative effect of loosening up the defense, the way that Barcelona’s incessant tiki-taka style creates enough small cracks that eventually a big scoring opportunity emerges?

Quick movement from FC Barcelona and Seattle Sockeye

Quick movement from FC Barcelona and Seattle Sockeye

What about my intuition that after five or six passes, an offense is usually so tired that it’s better to try to score in one long pass than to continue grinding away 3 yards at a time?

huck

You may notice that those two hunches might actually be at odds with each other. This is why we need the data! Send in the nerds.

CORRECTION (Dec. 16, 9:22 a.m.): An earlier version of this article incorrectly referred to Great Britain’s ultimate team as the English team throughout.


Read more: Ultimate Frisbee Is In The Dark Ages Of Analytics — And It Wants To Escape

Footnotes

  1. And/or lucky guess.
  2. Usually a defense tries to funnel offensive flow to one side of the field. Breaking the mark means that the offense is throwing passes to where the defense doesn’t want it to.

Jody Avirgan hosts and produces podcasts for FiveThirtyEight.

Comments