Rajiv Maheswaran: The math behind basketball's wildest moves
Rajiv Maheswaran: Die Formeln hinter den kompliziertesten Spielzügen im Basketball
Using advanced data analysis tools, Rajiv Maheswaran and Second Spectrum help make basketball teams smarter. Full bio
Double-click the English transcript below to play the video.
die Wissenschaft bewegter Punkte.
by the science of moving dots.
in our offices, as we shop and travel
im Büro, während wir in unseren Städten
and around the world.
einkaufen und reisen.
if we could understand all this movement?
Bewegung verstehen könnten --
and insight in it.
Erkenntnisse gewinnen könnten.
at capturing information about ourselves.
sensors or videos, or apps,
with incredibly fine detail.
sehr detailliert zu erfassen.
where we have the best data about movement
für diese Datensammlung
or football or the other football,
Football oder Fußball
and our players to track their movements
Spieler Bewegung erfassen --
is turning our athletes into --
and like most raw data,
schwer zu bearbeiten und langweilig.
and not that interesting.
basketball coaches want to know.
aber bestimmte Dinge wissen.
because they'd have to watch every second
the game with the eye of a coach.
nicht mit den Augen eines Trainers.
shots and rebounds.
slightly more complicated.
and pick-and-rolls, and isolations.
Most casual players probably do.
Den meisten Gelegenheitsspielern schon.
the machine understands complex events
komplexe Ereignisse,
with the eyes of a coach.
mit den Augen eines Trainers zu sehen.
something like a pick-and-roll,
Pick-and-Roll erklären,
it would be terrible.
wäre es fürchterlich.
in basketball between four players,
zwischen zwei Offensiv-Spielern
without the ball
in der Offensive ohne Ball.
guarding the guy with the ball,
Gegenspieler des Ballführenden
and ta-da, it's a pick-and-roll.
und plötzlich ist es Pick-and-Roll.
of a terrible algorithm.
he's called the screener --
in den Weg stellt -- der Blocksteller --
but he doesn't stop close enough,
and he does stop
it's probably not a pick-and-roll.
ist es kein Pick-and-Roll.
they could all be pick-and-rolls.
und alles ist Pick-and-Roll.
the distances, the locations,
we can go beyond our own ability
lernen über unsere Grenzen hinaus,
Well, it's by example.
"Good morning, machine.
"Guten Morgen, Maschine.
and here are some things that are not.
features that enable it to separate.
die Unterscheidung zu finden.
to teach it the difference
use color or shape?"
what are those things?
was das für Dinge sind.
the world of moving dots?
der bewegten Punkte steuern?
with relative and absolute location,
und absolutem Ort, Abstand, Timing
of moving dots, or as we like to call it,
wie wir es in akademischer Sprache nennen:
in academic vernacular.
you have to make it sound hard --
Es muss schwierig klingen.
it's not that they want to know
how it happened.
So here's a little insight.
Hier ein kleiner Einblick:
the most important play.
der wichtigste Spielzug.
and knowing how to defend it,
spielt oder verteidigt,
and losing most games.
oder Verlieren eines Spiels.
has a great many variations
in vielen Variationen.
is really the thing that matters,
ist das Entscheidende.
to be really, really good.
and two defensive players,
Pick-and-Roll-Tanz vor.
the pick-and-roll dance.
annehmen oder ablehnen.
can either take, or he can reject.
abrollen oder sich absetzen.
can either go over or under.
oder unter dem Block vorbei.
or play up to touch, or play soft
Ballführer aggressiv oder soft verteigen.
either switch or blitz
entweder "switchen" oder "blitzen".
most of these things when I started
das meiste davon nicht.
according to those arrows.
mit den Pfeilen bewegen könnten.
but it turns out movement is very messy.
Aber Bewegung ist unordentlich.
these variations identified
Und das Erkennen dieser Variationen
a professional coach to believe in you.
ein Trainer an einen glaubt.
with the right spatiotemporal features
richtigen spatiotemporalen Merkmalen
to identify these variations.
Maschine diese Variationen erkennt.
almost every single contender
on a machine that understands
einer Maschine, die bewegte Punkte
that has changed strategies
Strategien verändert,
very important games,
wichtigen Spielen verhalfen.
coaches who've been in the league
30 Jahren in der Liga sind,
advice from a machine.
it's much more than the pick-and-roll.
with simple things
mit einfachen Dingen,
much of what it does,
das meiste davon nicht.
to be smarter than me,
klüger zu sein als ich.
can a machine know more than a coach?
Maschine mehr wissen als ein Trainer?
to take good shots.
Würfe von den Spielern.
it's a good shot.
ist es ein guter Wurf.
by defenders, that's generally a bad shot.
Verteidigern, ist das ein schlechter Wurf.
or how bad "bad" was quantitatively.
wie gut oder schlecht etwas wirklich war.
using spatiotemporal features,
What's the angle to the basket?
Was ist der Winkel zum Korb?
What are their distances?
In welcher Entfernung?
at how the player's moving
wir die Bewegung der Spieler
and we can build a model that predicts
und berechnen die Wahrscheinlichkeit,
would go in under these circumstances?
Umständen reingeht.
and turn it into two things:
and the quality of the shooter.
und die des Werfers.
because what's TED without a bubble chart?
ja zu jedem TED-Vortrag gehört.
and the color is the position.
Spielers und die Farbe die Position.
we have the shot probability.
Wurfwahrscheinlichkeit.
bad at the bottom.
47 percent of their shots,
takes shots that an average NBA player
Würfe macht, die ein durchschnittlicher
is that there are lots of 47s out there.
giving 100 million dollars to
100 Millionen Dollar investieren möchte,
mit guten Würfen ist.
how we look at players,
unsere Sicht auf Spieler
a couple of years ago, in the NBA finals.
ein sehr spannendes Spiel.
there was 20 seconds left.
noch 20 Sekunden zu spielen.
came up and he took a three to tie.
named Ray Allen.
They won the championship.
und die Meisterschaft.
games in basketball.
spannendsten Basektballspiele.
the shot probability for every player
eines jeden Spielers
a rebound at every second
Moment auf ungeahnte Weise beleuchten.
that we never could before.
I can't show you that video.
das Video nicht zeigen.
about 3 weeks ago.
Spiel vor 3 Wochen.
that led to the insights.
und erhielten folgende Einblicke.
This is Chinatown in Los Angeles,
Los Angeles, der Park,
the Ray Allen moment
that's associated with it.
of the professional players, it's us,
announcer, it's me.
chance of happening in the NBA
in der NBA passiert, lag bei 9 %.
and a great many other things.
it took us to make that happen.
die Szene wiederholen mussten.
Knackpunkt an diesem Video --
of every NBA game -- it's not that.
Sekunde jedes NBA-Spiels.
a professional team to track movement.
sein muss, um Bewegung aufzuzeichnen.
player to get insights about movement.
um Einblicke in Bewegung zu erhalten.
sports because we're moving everywhere.
Wir bewegen uns überall.
Was werden wir lernen?
pick-and-rolls,
the moment and let me know
den Moment identifizieren,
ihren ersten Schritt macht.
any second now.
Moment passieren könnte.
our buildings, better plan our cities.
besser zu nutzen oder Städte zu planen.
of the science of moving dots,
Wissenschaft der bewegten Punkte
we will move forward.
wir uns klüger, bewegen wir uns vorwärts.
ABOUT THE SPEAKER
Rajiv Maheswaran - ResearcherUsing advanced data analysis tools, Rajiv Maheswaran and Second Spectrum help make basketball teams smarter.
Why you should listen
Sports fans can get obsessed with stats about player performance and game-day physics. But basketball, a fluid and fast-moving game, has been tough to understand through numbers. Rajiv Maheswaran is working to change that, by offering pro basketball teams insight into game data to make better decisions. Maheswaran is the CEO and co-founder of Second Spectrum, a startup transforming sports through technology. He is also a Research Assistant Professor at the University of Southern California's Computer Science Department and a Project Leader at the Information Sciences Institute at the USC Viterbi School of Engineering, where he co-directs the Computational Behavior Group.
His research spans various aspects of multi-agent systems and distributed artificial intelligence using decision-theoretic and game-theoretic frameworks and solutions. His current interests focus on data analytics, visualization and real-time interaction to understand behavior in spatiotemporal domains. Like, say, the spatiotemporal domain around a basketball hoop.
Rajiv Maheswaran | Speaker | TED.com