Recent Posts on Twitter

profile

Jaime Fitzgerald

jaimefitzgerald

Sociable

Each NBA Shot is a Decision: What's the Decision Model?

 These days, the sports world is full of people paying close attention to statistics....

  1. General Managers use unconventional statistics to find hidden value in players....
  2. Fans discuss stats when debating what a player is worth, how much playing time he should get and who he should be playing against.
  3. Journalists use statistics to write stories - at best they'll use numbers to create a new narrative; at worst they'll use numbers to conveniently support an intuitive--but false--"story hook"
David Biderman, who writes "The Count," a Sports and Numbers blog for the WSJ, seems to have fallen into the latter category with his recent piece on what the Miam's Heat's passing means for their scoring chances.  His convenient but suspicious "story hook" is that the more the Heat pass, the lower their shooting percentage gets.  Here at Team Fitzgerald, we are deeply skeptical.
 
Biderman's contention is that because the Heat's shooting % is higher when they pass less, they should start acting more selfishly, pass less, and voila...they can boost their shooting percentage.  If it were that simple....

But wait!...correlation does not equal causation.  Do players get higher percentage shots when they pass less?  Or do they pass less when they have a high percentage shot, for example on a fast break?  

Let's take an important step back:  each pass or shot is a decision.  Professional players make it thousands of times, and coaches monitor how well they make that decision.  We praise players who take higher percentage shots when they have them, and blame them when they give up a good shot with a pass they didn't need to make.  On the other hand, if a player does not have a high percentage thought, we praise the decision to pass...nobody admires a "forced shot" with low probability of hitting the net.

If a player has a sweet fast break, or an open path to basket, the rational choice is to shoot (or dunk) bc they already have their high percentage shot.  If the defense is able to get back, and the player can't easily take it himself, the rational choice is to slow it down and look for a set play that may involve more than 1.  These are the decisions a player makes based on the set of facts in front of them in the moment.  By not eliminating fast break plays, controlling for shot location, etc...Biderman is left with a data sample skewed enough to result in incorrect conclusions.  Separating them would be a better way to see the impact of passing on a given basketball play. 


So mathematically, let's run a quick scenario.  During each game hundreds of "pass vs. shoot" decisions will be made.  Rational team players shoot when the odds are in their favor (high percentage shot) and pass when they don't have a good shooting opportunity, seeking a better one.  Meanwhile, the shot clock ticks...and as time passes, the calculus changes.  Given the risk of running out of time on the shot clock, extra passes become less attractive and shots on the basket become more imperative, and players rationally are more willing to take lower-percentage, lower-quality shots.  Shooting percentages can be expected to fall, because players are forced to shoot without having found an easy layup or fast-break dunk. 

The article was based on data from only two games by one team.  We wish we could find a similar but  potentially more valuable data-set:  shooting percentages by teammate, segmented by the number of seconds on the shot clock.  This would allow us to test our hypotheses, which is that game patterns, and the related passage of time, cause patterns in shooting percentage, and that passing less is correlated with, but does not cause, higher shooting percentages.
 
We feel that Biderman's piece misses is what's at the heart of why statistics in sports are so captivating to so many of us. They allow us to reveal truths that were previously hidden.  Biderman's piece doesn't make an attempt to dig deeper into why the Heat score less when they pass more.  There is no hidden truth in this article, only a potential straw man. 

What do you think?

Alex Roberts & Jaime Fitzgerald

Speaking of Turning Data to Dollars™: Previewing two Conference Presentations this Spring

 This week, I was delighted to receive invitations to speak this coming spring at both the AIIM International Conference & Expo and at Enterprise Data World.  I’m excited about these opportunities to share real cases in the area of my passion: turning data into tangible results.

 
In late March, I’ll be presenting on “Knowledge Management for Analytic Professionals and Teams” at the AIIM International Conference & Expo @info360 in Washington, D.C.  This annual event brings together thousands of technologists and business users to learn about and discuss the fast-changing landscape of information management -- and I’m excited to be a part of it!
 
Two weeks later, in the first week of April, I’ll be speaking for the second year in a row at Enterprise Data World in Chicago, Illinois, on the role data plays in customer experience.  The title of my talk will be “Data-Driven Customer Satisfaction: The Link between Data Management, Balanced Scorecard, and Client Service Best Practices.”  I’ll be sharing the stage with a long-time client, Zohar Swaine, Managing Director of Institutional Strategy & Product at TD AMERITRADE Institutional.
 
And just in case you’re interested,  the slides from my presentation @ the 2010 Enterprise Data World conference, “Tales from a Master Data Management Transformation Roadtrip,” given with Art Garanich, Director of Information Technology at Bridgestone Firestone's Consumer Credit company, which provides private label credit cards for more than 5,000 retailers as part of the Bridgestone Firestone Corporation.
 
Besides sharing lessons learned while helping clients, these conferences are always a great opportunity to meet up with old friends and to learn from other thought leaders who are spearheading innovation in the data management space.  I hope to see you there!

 

What Motivates Analytic Professionals?

 There are many reasons why we love or hate our jobs.



Professional analysts: what motivates YOU the most? If you could design the perfect job in analytics, what is the key?  

For example, are you motivated by the nature of the work, the types of problems you solve, the financial incentives, career path upside, or something else?


Update/October 26 2010:

Hi fellow professionals -- thanks so much for all your comments on this thread.  We hosted this discussion both HERE and on the Fitzgerald Analytics page on AnalyticBridge. Both the number, and the quality, of the posts exceeded my expectations.  In about 1 week we are going to "close the window" on comments for this round, so we can plan and pursue next steps to research the topic further...and will be reaching out to all of you to invite your participation in that next round of discovery.

In the meantime if you have not commented YET, please do this week!  I'm super grateful, and will do whatever I can to make it worth your while as we pursue this further.
Cheers, 
Jaime Fitzgerald
Founder and President
Fitzgerald Analytics

Semantic Web Technologies and Open Data Innovation

 With my colleagues Kevin Cabral and Justin Goldbach, I attended and greatly enjoyed the latest event of the  Lotico New York Semantic Web Meetup which was about open data innovation, especially with a focus on Data.gov and obviously, the role of Semantic Web Technologies in making open data more accessible, more useful, and more valuable to citizens and businesses.


The speakers were incredible:

  1. Gale A. Brewer, New York City Council Member, who is clearly passionate and experienced in the benefits of applying technology and data innovation to benefit citizens, improve public services, and enable non-profits to help more people
  2.  Jim Hendler, Tetherless World Senior Constellation Professor, Rensselaer Polytechnic Institute (RPI) -- a pioneer in the field and a fascinating speaker.
  3. Deborah L. McGuinness, Tetherless World Senior Constellation Professor RPI -- gave a fascinating talk.  My favorite line:  "the government would like nothing better than to see private companies use public data to make money, create jobs...create value with this."
  4. Peter Fox, Professor and Tetherless World Research Constellation Chair RPI. Wow.  A great presentation also, about "The eScience revolution: Semantic Web platform for massive scientific collaboration"
The implications of Semantic Web Technology, together with Open Data on the Web, including government data, creates incredible promise for innovation, transparency, and value creation.  The wheels are spinning.

Helping in Haiti, Informed Giving, and Maximizing the Impact of Donations

The recent tragedy in Haiti--and the outpouring of well-deserved sympathy and generosity it stimulated--has also highlighted a crucial trend in the non-profit world: the role of information in improving productivity, efficiency, and results.

This is good, as better information about how donations are used should help inspire MORE people to give what they can, confident that they know their hard-earned dollars will be deployed efficiently. But the tools need to be used to get these benefits...

According to the Chronicle of Philanthropy, private-sector donations to relief efforts in Haiti were more than $600 Million in the first 3 weeks since the earthquake...it looks like private giving could end up exceeding a billion dollars. And while this is a large and historic surge of giving, Haiti's challenges far exceed the resources available to fully address them.

When results matter--as they surely do at a time like this in Haiti--the tools we use to maximize the positive impact of our resources becomes incredibly important. Efficiency matters. My hope is that whenever possible, donors to charities not only give resources (e.g. dollars), but give them in ways that maximize the results achieved from these resources.

To give as efficiently as possible, we need good information on our options, and the implications of HOW we give. The good news is that today's donors have access to excellent information if they choose to use it....information that can multiply the impact of donor dollars by channeling resources to the most efficient charitable organizations. Several key resources here include:

At Fitzgerald Analytics, we have made a lasting commitment to serving the social sector as well as the private sector with our skills and our passions. We made this commitment because we genuinely believe that good decisions are even more important in the public and social sectors than they are in the private sector. With better decisions--enabled by information--our vision is to maximize the positive results achieved from existing resources, which are always finite.

Using the resources listed above, my team and I at Fitzgerald Analytics chose to make a contribution this month to Partners in Health. We were attracted to their clear business model, their efficiency, and their impressive rating @ Charity Navigator.

Final notes and next steps:

The topic is far more complex than any single blog post: it is important to acknowledge the inherent complexity and challenge of measuring the effectiveness of charitable programs. The social sector is incredibly diverse, and so are the drivers of results, the types of results being pursued, and the tools for measuring outcomes. We look forward to exploring a variety of more detailed topics in future posts, and also to learning more from our allies in the non-profit sector regarding the challenges they face, the nuances they encounter, and the emerging non-profit models they see getting the most traction towards the results donors so generously seek to enable.

Please share your insights: Please do get in touch so that we can "thought-partner" and collaborate further on this essential but complex topic.

Posted By:
Jaime Fitzgerald

 

Showing 1 - 5 of 15 results.
Items per Page
Page of 3

Friday, December 10, 2010

Each NBA Shot is a Decision: What's the Decision Model?

 
These days, the sports world is full of people paying close attention to statistics....
  1. General Managers use unconventional statistics to find hidden value in players....
  2. Fans discuss stats when debating what a player is worth, how much playing time he should get and who he should be playing against.
  3. Journalists use statistics to write stories - at best they'll use numbers to create a new narrative; at worst they'll use numbers to conveniently support an intuitive--but false--"story hook"
David Biderman, who writes "The Count," a Sports and Numbers blog for the WSJ, seems to have fallen into the latter category with his recent piece on what the Miam's Heat's passing means for their scoring chances.  His convenient but suspicious "story hook" is that the more the Heat pass, the lower their shooting percentage gets.  Here at Team Fitzgerald, we are deeply skeptical.
 
Biderman's contention is that because the Heat's shooting % is higher when they pass less, they should start acting more selfishly, pass less, and voila...they can boost their shooting percentage.  If it were that simple....

But wait!...correlation does not equal causation.  Do players get higher percentage shots when they pass less?  Or do they pass less when they have a high percentage shot, for example on a fast break?  

Let's take an important step back:  each pass or shot is a decision.  Professional players make it thousands of times, and coaches monitor how well they make that decision.  We praise players who take higher percentage shots when they have them, and blame them when they give up a good shot with a pass they didn't need to make.  On the other hand, if a player does not have a high percentage thought, we praise the decision to pass...nobody admires a "forced shot" with low probability of hitting the net.

If a player has a sweet fast break, or an open path to basket, the rational choice is to shoot (or dunk) bc they already have their high percentage shot.  If the defense is able to get back, and the player can't easily take it himself, the rational choice is to slow it down and look for a set play that may involve more than 1.  These are the decisions a player makes based on the set of facts in front of them in the moment.  By not eliminating fast break plays, controlling for shot location, etc...Biderman is left with a data sample skewed enough to result in incorrect conclusions.  Separating them would be a better way to see the impact of passing on a given basketball play. 


So mathematically, let's run a quick scenario.  During each game hundreds of "pass vs. shoot" decisions will be made.  Rational team players shoot when the odds are in their favor (high percentage shot) and pass when they don't have a good shooting opportunity, seeking a better one.  Meanwhile, the shot clock ticks...and as time passes, the calculus changes.  Given the risk of running out of time on the shot clock, extra passes become less attractive and shots on the basket become more imperative, and players rationally are more willing to take lower-percentage, lower-quality shots.  Shooting percentages can be expected to fall, because players are forced to shoot without having found an easy layup or fast-break dunk. 

The article was based on data from only two games by one team.  We wish we could find a similar but  potentially more valuable data-set:  shooting percentages by teammate, segmented by the number of seconds on the shot clock.  This would allow us to test our hypotheses, which is that game patterns, and the related passage of time, cause patterns in shooting percentage, and that passing less is correlated with, but does not cause, higher shooting percentages.
 
We feel that Biderman's piece misses is what's at the heart of why statistics in sports are so captivating to so many of us. They allow us to reveal truths that were previously hidden.  Biderman's piece doesn't make an attempt to dig deeper into why the Heat score less when they pass more.  There is no hidden truth in this article, only a potential straw man. 

What do you think?
 

Alex Roberts & Jaime Fitzgerald 

 

Past Fitzgerald Analytics Blog Entries:

Next
Each NBA Shot is a Decision: What's the Decision Model?

 These days, the sports world is full of people paying close attention to statistics....

  1. General Managers use unconventional statistics to find hidden value in players....
  2. Fans discuss stats when debating what a player is worth, how much playing time he should get and who he should be playing against.
  3. Journalists use statistics to write stories - at best they'll use numbers to create a new narrative; at worst they'll use numbers to conveniently support an intuitive--but false--"story hook"
David Biderman, who writes "The Count," a Sports and Numbers blog for the WSJ, seems to have fallen into the latter category with his recent piece on what the Miam's Heat's passing means for their scoring chances.  His convenient but suspicious "story hook" is that the more the Heat pass, the lower their shooting percentage gets.  Here at Team Fitzgerald, we are deeply skeptical.
 
Biderman's contention is that because the Heat's shooting % is higher when they pass less, they should start acting more selfishly, pass less, and voila...they can boost their shooting percentage.  If it were that simple....

But wait!...correlation does not equal causation.  Do players get higher percentage shots when they pass less?  Or do they pass less when they have a high percentage shot, for example on a fast break?  

Let's take an important step back:  each pass or shot is a decision.  Professional players make it thousands of times, and coaches monitor how well they make that decision.  We praise players who take higher percentage shots when they have them, and blame them when they give up a good shot with a pass they didn't need to make.  On the other hand, if a player does not have a high percentage thought, we praise the decision to pass...nobody admires a "forced shot" with low probability of hitting the net.

If a player has a sweet fast break, or an open path to basket, the rational choice is to shoot (or dunk) bc they already have their high percentage shot.  If the defense is able to get back, and the player can't easily take it himself, the rational choice is to slow it down and look for a set play that may involve more than 1.  These are the decisions a player makes based on the set of facts in front of them in the moment.  By not eliminating fast break plays, controlling for shot location, etc...Biderman is left with a data sample skewed enough to result in incorrect conclusions.  Separating them would be a better way to see the impact of passing on a given basketball play. 


So mathematically, let's run a quick scenario.  During each game hundreds of "pass vs. shoot" decisions will be made.  Rational team players shoot when the odds are in their favor (high percentage shot) and pass when they don't have a good shooting opportunity, seeking a better one.  Meanwhile, the shot clock ticks...and as time passes, the calculus changes.  Given the risk of running out of time on the shot clock, extra passes become less attractive and shots on the basket become more imperative, and players rationally are more willing to take lower-percentage, lower-quality shots.  Shooting percentages can be expected to fall, because players are forced to shoot without having found an easy layup or fast-break dunk. 

The article was based on data from only two games by one team.  We wish we could find a similar but  potentially more valuable data-set:  shooting percentages by teammate, segmented by the number of seconds on the shot clock.  This would allow us to test our hypotheses, which is that game patterns, and the related passage of time, cause patterns in shooting percentage, and that passing less is correlated with, but does not cause, higher shooting percentages.
 
We feel that Biderman's piece misses is what's at the heart of why statistics in sports are so captivating to so many of us. They allow us to reveal truths that were previously hidden.  Biderman's piece doesn't make an attempt to dig deeper into why the Heat score less when they pass more.  There is no hidden truth in this article, only a potential straw man. 

What do you think?

Alex Roberts & Jaime Fitzgerald

Next

  • Comments
Trackback URL:

drakensang online скачать клиент торрент
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 12:26 AM.

Top Top
brilliance audio
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 12:46 AM.

Top Top
listen to audio books free online
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 12:51 AM.

Top Top
audiobook download sites
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 12:57 AM.

Top Top
kindle app for ipod
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 1:19 AM.

Top Top
android ebooks free download
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 1:35 AM.

Top Top
audio books for children
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 4:11 AM.

Top Top
audiobook free
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 4:54 AM.

Top Top
fake ray ban sunglasses
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 5:40 AM.

Top Top
download free audio books library
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 9:19 AM.

Top Top
weight loss retreat
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 9:46 AM.

Top Top
wholesale handbags uk
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 9:54 AM.

Top Top
Ralph lauren outlet Online
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 9:58 AM.

Top Top
eating to lose weight
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 9:59 AM.

Top Top
hand steam cleaner
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 11:46 AM.

Top Top
north of las vegas
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 12:22 PM.

Top Top
furnace brands Waukesha
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 12:50 PM.

Top Top
carpet cleaning machines for rent
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/12/14 1:02 PM.

Top Top
pure forskolin contains no stimulants.
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/17/14 6:34 PM.

Top Top
pure forskolin extract
[...] Fitzgerald Analytics - Each NBA Shot is a Decision: What's the Decision Model? - Blog [...] Read More

Posted on 7/17/14 6:34 PM.

Top Top