Rajiv Maheswaran: The math behind basketball's wildest moves
راجيف ماهيسواران: الرياضيات خلف حركات كرة السلة العجيبة
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.
by the science of moving dots.
كما نحن نتسوق و نسافر
in our offices, as we shop and travel
and around the world.
ان نتفهم كل هذه التحركات
if we could understand all this movement?
and insight in it.
و معاني و النظر في ذلك
at capturing information about ourselves.
التقاط معلومات عن انفسنا
sensors or videos, or apps,
او الفيديو, او التطبيقات
with incredibly fine detail.
بتفاصيل دقيقة لا تصدق
where we have the best data about movement
ان افضل بيانات حول الحركة
or football or the other football,
او كرة القدم او كرة القدم الاخرى
and our players to track their movements
لاعبينا حتى نتتبع حركاتهم
is turning our athletes into --
and like most raw data,
و كمعظم البينات الاولية،
and not that interesting.
basketball coaches want to know.
مدربين كرة السلة يريدون معرفته.
because they'd have to watch every second
بحاجة الى مشاهدة كل ثانية
the game with the eye of a coach.
المباراة كما يراها المدرب
shots and rebounds.
و لقطات و الترددات.
slightly more complicated.
الاكثر تعقيدا بقليل.
and pick-and-rolls, and isolations.
انتقاء و لفات, و العزلة
Most casual players probably do.
معظم اللاعبين العاديين ربما يعرفون.
the machine understands complex events
الآلة تفهم احداث معقدة
with the eyes of a coach.
something like a pick-and-roll,
مثل "بيك-اند-رول",
it would be terrible.
سيكون فظيعا.
in basketball between four players,
الرقصة في كرة السلة بين اربعة لاعبين,
without the ball
الموقع الهجومي دون الكرة
guarding the guy with the ball,
الذي يحرس الشخص مع الكرة،
and ta-da, it's a pick-and-roll.
و تا-دا, انها "بيك-اند-رول".
of a terrible algorithm.
he's called the screener --
-- هو يسمي ب"سكرينير"--
but he doesn't stop close enough,
لا يقترب الى حد كافي,
and he does stop
it's probably not a pick-and-roll.
انها لربما ليست " بيك-اند-رول".
they could all be pick-and-rolls.
ان كلهن " بيك-اند-رول".
the distances, the locations,
و المسافات و المواقع
we can go beyond our own ability
يمكننا تجاوز قدراتنا
Well, it's by example.
"Good morning, machine.
" صباح الخير يا آلة.
and here are some things that are not.
و ها هنا بعض الاشياء التي ليست.
features that enable it to separate.
التي تمكنها من الفصل.
to teach it the difference
use color or shape?"
what are those things?
ما هي هذه الاشياء؟
the world of moving dots?
عالم النقاط المتحركة؟
with relative and absolute location,
مع موقع النسبي و المطلق،
of moving dots, or as we like to call it,
او كما نحب ان نسميه،
in academic vernacular.
في الاكادمية العامية.
you have to make it sound hard --
الامر يبدو صعب--
it's not that they want to know
انهم لا يريدون معرفة
how it happened.
So here's a little insight.
the most important play.
and knowing how to defend it,
و معرفة كيفية حمايتها,
and losing most games.
في معظم المباريات.
has a great many variations
is really the thing that matters,
to be really, really good.
ذلك يكون جيد جدا,
and two defensive players,
و لاعبين اثنين مدافعين,
the pick-and-roll dance.
can either take, or he can reject.
يأخذ او يمكنه يرفض
ال"رول" او "بوب".
can either go over or under.
اما ان يذهب من فوق او من تحت.
or play up to touch, or play soft
حتى اللمس او اللعب بليونه
either switch or blitz
most of these things when I started
according to those arrows.
تحركوا وفقا لهذه الاسهم.
but it turns out movement is very messy.
و لكن اتضح ان الحركات فوضوية جدا.
these variations identified
a professional coach to believe in you.
with the right spatiotemporal features
الزمانية و المكانية الصحيحة
to identify these variations.
في تحديد هذه المتغيرات.
almost every single contender
on a machine that understands
على الآلة التي تدرك
that has changed strategies
التي غيرت استراتيجيات
very important games,
coaches who've been in the league
مدربين الذين كانوا في الدوري
advice from a machine.
ﻹخذ نصائح من آلة.
it's much more than the pick-and-roll.
من "بيك اند رول".
with simple things
much of what it does,
to be smarter than me,
can a machine know more than a coach?
ان تعرف اكثر من المدرب؟
to take good shots.
لاتخاذ ضربات جيدة.
it's a good shot.
by defenders, that's generally a bad shot.
تلك بشكل عام ضربة سيئة.
or how bad "bad" was quantitatively.
او كيف السيئ كان "سيئا" كميا.
using spatiotemporal features,
ميزات الزمانية و المكانية،
What's the angle to the basket?
أي زاوية الى السلة؟
What are their distances?
at how the player's moving
كيف يتحرك اللاعبون
and we can build a model that predicts
و يمكننا ان نبنى نموذج الذي يتوقع
would go in under these circumstances?
ستسير وفق هذه الظروف؟
and turn it into two things:
و تحويله الى شيئين:
and the quality of the shooter.
because what's TED without a bubble chart?
لأنه ما هو TED من دون تخطيط فقاعي؟
and the color is the position.
we have the shot probability.
bad at the bottom.
السيئين في الاسفل.
اذا كان هناك لاعب
47 percent of their shots,
takes shots that an average NBA player
الذي يتخذ رميات ان لاعب الNBA العادي
is that there are lots of 47s out there.
بهذا المستوى .
giving 100 million dollars to
how we look at players,
ننظر الى اللاعبين،
a couple of years ago, in the NBA finals.
في نهائيات الNBA.
there was 20 seconds left.
و تبقى من 20 ثانية.
came up and he took a three to tie.
و اتخذ ثلاث للتعادل.
named Ray Allen.
They won the championship.
games in basketball.
جدا في كرة السلة.
the shot probability for every player
محاولات لكل لاعب
a rebound at every second
that we never could before.
لم نستطع بها من قبل.
I can't show you that video.
about 3 weeks ago.
حوالي ثلاثة اسابيع.
that led to the insights.
This is Chinatown in Los Angeles,
في لوس انجلوس،
the Ray Allen moment
that's associated with it.
of the professional players, it's us,
بدلا من لاعبين محترفين،
announcer, it's me.
انا المعلّق.
chance of happening in the NBA
حوالي 9 بالمئة في NBA
and a great many other things.
it took us to make that happen.
of every NBA game -- it's not that.
لعبة NBA -- ليس ذلك.
a professional team to track movement.
فريقا محترفا لتتابع الحركات.
player to get insights about movement.
للحصول على رؤيات الحركات.
sports because we're moving everywhere.
عن الرياضة لانه نحن نتحرك في كل مكان.
pick-and-rolls,
the moment and let me know
any second now.
our buildings, better plan our cities.
لمبانينا، تخطيط افضل لمدننا.
of the science of moving dots,
we will move forward.
سوف نتحرك للامام.
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