Forecasting NBA Basketball Playoff Outcomes Using the Weighted Likelihood
by Feifang Hu & James V Zidek ~ in our From the Archives Press & Video Clippings


Forecasting NBA Basketball Playoff Outcomes Using the Weighted Likelihood
by Feifang Hu & James V Zidek ~ in our From the Archives Press & Video Clippings

From the Archives Press Clippings:
Forecasting NBA Basketball Playoff Outcomes Using the Weighted Likelihood
By Feifang Hu & James V Zidek

Basketball Statistician's Friday Press Clippings
From Natural Sciences and Engineering Research Council of Canada
By Feifang Hu & James V Zidek
Download source here:
Forecasting NBA Basketball Playoff Outcomes Using the Weighted Likelihood
by Feifang Hu & James V Zidek

"... Predicting the outcome of a future game between two sports teams poses a challenging problem of interest to statistical scientists as well as the general public. To be effective such prediction must exploit special contextual features of the game. In this paper, we confront such features and address the need to:

(i) use all relevant sample information;
(ii) reflect the home game advantage.

To do so we use the weighted likelihood. Finally we demonstrate the value of the method by showing how it could have been used to predict the 96/97 NBA playoff results. Our weighted likelihood-based method proves to be quite accurate. ..."

"... This paper demonstrates the use of the weighted likelihood (WL) to predict the winner of National Basketball Association (NBA) final games and applies the method in each of the games played in the 1996-97 finals between the two top ranked teams, the Chicago Bulls and the Utah Jazz. However, as we try to indicate, the WL has much wider applicability inside as well outside the domain of sports.way dunker created by a teammate’s steal, offensive rebound, and or perfect outlet pass? ..."

"... The genesis of our work lies in two statistical problems encountered in sports:
(i) the prediction of the outcome of a future game between two specified sports teams;
(ii) the assessment of the accuracy of this prediction.

Since typically these two teams will not have met more than just a few times in the given season, little direct information will be available to the forecaster. The consequent small sample size will make naive predictions inaccurate and the associated prediction intervals excessively large. ..."

1996-97 Chicago Bulls Championship: #1 ~ Randy Brown: #30 ~ Jud Buechler; #35 ~ Jason Caffey; #18 ~ Bison Dele; #9 ~ Ron Harper; #23 ~ Michael Jordan; #25 ~ Steve Kerr; #7 ~ Toni Kukoc; #13 ~ Luc Longley; #00 ~ Robert Parish; #33 ~ Scottie Pippen; #91 ~ Dennis Rodman; #8 ~ Dickey Simpkins; #6 ~ Matt Steigenga, #34 	Bill Wennington In this photograph from Operation Sports presented by eBA Stats Basketball Statistics Analysis .
Photograph: Operation Sports

"... Usually in sports, the outcome of any one game derives from the combined efforts of two teams that have seldom played each other before. Yet these games yield the only direct sample information available about the relative strength of these two teams.

At the same time, each of these teams will have played many games against other teams thereby generating relevant (although not direct) sample information. The predictive probability of a win in the next game between these two teams, should combine both kinds of information ..."

"... Based on the probabilities and the confidence intervals of Table 1, we can find the probabilities with which the Bulls (and Jazz) will win the Final by Game 4, 5, 6 or 7.

Also we can calculate the total win probabilities for the Bulls against the Jazz based on its home and away win probabilities given by each of the three estimation methods.
Confidence intervals for these win probabilities may be obtained as well.

In Table 2 where the results are reported, and in the tables that follow, that interval is obtained for any pair of teams say A and B from the 95% asymptotic intervals for A’s home- and A’s away-win-against-B probabilities.. ..."

"... The method in this paper provides guidelines for the development of a prediction strategy. Its implementation, more specifically the construction of weights entails the incorporation of any special features that may obtain when the game is played.

For example, one might need to incorporate the knowledge that certain key players cannot play in that game."

To be more precise, we fitted a logistic model using the software R with the response variable being 1 or 0 according as the outcome of any game during the season was a visitor or home victory. We used dummy variables to represent visitor and home teams in each game throughout the season.

Thus for example, BullsV=1 and Super-sonics=1, all other dummies being 0, would mean those two teams were playing for that particular game, the visitors being the Bulls. For each of the factors, ‘visitor’ and ‘home’ we represented by the dummies in this way, we arbitrarily chose the 76er’s as the baseline team.

Thus, in effect, the fitted intercept, suitably transformed, provides an estimate of the likelihood of a ‘1’ in the purely hypothetical situation where the 76er’s played themselves at home as the visitors.

The coefficients for the remaining dummies represent the deviations from the 76er’s performance for each of the other teams depending on whether they were playing at home or awa ..." Read more ...

Read more articles in Natural Sciences and Engineering Research Council of Canada

Get your Warriors 2018 Western Conference Champs at NBA Store!

To read today's 'Press Clippings' in Spanish,
click this week above on the 'Español' tab or afterwards click here:
“Análisis Espacial de Lanzamientos en Baloncesto; el Caso de L.A. Lakers
by López, F.A., Martínez, J.A. y Ruiz, M.

From the Archives Video Clippings:
Va' pensiero Riccardo Muti speaking about Italian culture, Opera di Roma

Va' pensiero Riccardo Muti speaking about Italian culture, Opera di Roma

Muti said : yes, viva Italia... But, as we are in house, quite at home, let us speak together.
"Va pensiero", in old days, was a political symbol. I'm not a politician but I can say that if our culture goes on being slain, our Italia will be "si bella e perduta". So, as the choir has beautifully sung it, we should join them to sing another time "Va, pensiero !"

'From the Archives Press and Video Clippings'
is a Series of Notes edited the regularly!

To Read ALL the Clippings in this Series, click here: Categories > Contents Online

Your comments are Welcomed ! ... by means of eBA InterMail English or just here !

Post Your Comment

tags:  ~ ~ ~ ~ ~ ~ ~
visit:  eBA Portal  ~  Forums  ~  Encyclopedia  ~  Clinics ~ eBA System Book ~ eBA Facebook ~ eBA Twitter ~ eBA Google+ ~ YouTube BA ~ eBA Store


No Pingbacks for this post yet...

Custom Search

View blog top tags

November 2019
Sun Mon Tue Wed Thu Fri Sat
 << <   > >>
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
Recommend Our Blog ! Thanks !
Recomienda Nuestro Blog ! Gracias !

Basketball Coaching Videos & Statistics Blog

BASKETBALL Coaching, Statistics, Videos & Couses BLOG.


FEBRUARY 2019 Selected Pages !

About the eBA Stats Group Social Networks, Basketball Plays & Statistics Formulas!

About the Basketball Efficiency
in Different Possession Start Cases

About Basketball Assists
& Potential Assists

About the Right Way
to Evaluate the Basketball Defense

About Two Assists
in the SAME Basketball Play

About the Basketball
Team Defense Analysis

Open Letter About LinkedIn

About Basketball Stats,
Methods and Players

The Hand IS Part of the Ball
in Basketball

Underrated or Overrated
Individual Stats

Game Pace & TO Ratios
of Individual Players

Tracking Each Shot and the Game Situation with the Best Indicators of the Winner

Basketball Game Analyst
or Scientific Analyst ?

About Basketball Scouters & Statisticians

Modern Basketball Assist-Turnover
Ratio Levels

The Statistical Decision Limit
in Basketball

Basketball Rebounds
as Statistical Factor

Statistics Never Tell
the WHOLE Story

Basketball Starters and Substitutes

About the Basketball
Plus / Minus Data Analysis

Statistics Box Score Balance Errors

Basketball Assists
& Potential Assists

Basketball Team Defense

Basketball TO
& the "Game Pace" Factor

Páginas Seleccionadas
de FEBRERO 2019 !

Sobre las Redes Sociales del Grupo eBA Stats, Jugadas de Baloncesto y Fórmulas Estadísticas!

Sobre la Eficiencia en Baloncesto
Según el Comienzo de Posesión

Sobre las Asistencias del Baloncesto
y las Asistencias Potenciales

Sobre la Forma Correcta de Evaluar
la Defensa en Baloncesto

Sobre Dos Asistencias
en Una MISMA Jugada en Baloncesto

Análisis de la Defensa de Equipo
en Baloncesto

Carta Abierta Sobre LinkedIn

Sobre Estadísticas, Métodos
y Jugadores

La Mano ES Parte del Balón
en Baloncesto

'Ritmo de Juego'
& la 'Relación de BP'

Cada Lanzamiento Conforme a la Situación y los Mejores Indicadores de una Victoria

Analista de Partido de Baloncesto
ó Analista Científico ?

Sobre Escouters y Estadígrafos
del Baloncesto

Niveles de Relación Asistencias-Balones Perdidos en el Baloncesto

El Límite Estadístico Determinante
en Baloncesto

Concepto del Rebote
como Factor Estadístico en Baloncesto

Las Estadísticas No cuentan TODO !

Titulares y Suplentes en Baloncesto

Sobre el Análisis de los
Datos del Plus / Minus en Baloncesto

Errores en el Balance de las Estadísticas

Asistencias y Asistencias Potenciales

Análisis de la Defensa de Equipo

BP y el Factor "Ritmo de Juego"

Hey There ! LIKE US on Facebook !
Thank You !

Hola Gente ! Regálanos un LIKE en Facebook ! Muchas Gracias !


XML Feeds

What is this?

Bookmark and Share