没有合适的资源?快使用搜索试试~ 我知道了~
2024年美赛35篇特等奖O奖论文-C-2418251.pdf
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 132 浏览量
2024-05-06
11:41:44
上传
评论
收藏 3.68MB PDF 举报
温馨提示
![preview](https://dl-preview.csdnimg.cn/89271471/0001-a00e0b61e3ee07e5681036ea49f2f72a_thumbnail.jpeg)
![preview-icon](https://csdnimg.cn/release/downloadcmsfe/public/img/scale.ab9e0183.png)
试读
24页
大学生,数学建模,美国大学生数学建模竞赛,MCM/ICM,2024年美赛特等奖O奖论文
资源推荐
资源详情
资源评论
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![pdf](https://img-home.csdnimg.cn/images/20210720083512.png)
![pdf](https://img-home.csdnimg.cn/images/20210720083512.png)
![pdf](https://img-home.csdnimg.cn/images/20210720083512.png)
![](https://csdnimg.cn/release/download_crawler_static/89271471/bg1.jpg)
Problem Chosen
C
2024
MCM/ICM
Summary Sheet
Team Control Number
2418251
Momentum: Powerful Magic for Winning
Summary
In the 2023 Wimbledon Finals, Carlos Alcaraz dethroned Novak Djokovic in a captivating
display of youthful exuberance over seasoned skill. The match’s dynamic ebb and flow highlighted
the intangible factor of momentum. To capture this phenomenon, our team crafted mathematical
models that leveraged Wimbledon data to quantify player momentum, offering fresh insights into
what drives tennis triumphs.
For Model 1, we started by cleaning the data to remove outliers. We then extracted the match
data for Novak Djokovic from the Finals. By feeding his data into a Recurrent Neural Network
(RNN), we calculated his winning probability for each game when serving and receiving. The re-
sults showed a significantly higher winning probability when serving than receiving, which aligns
with experience. Additionally, we created visualization charts to intuitively display the changes in
winning probability throughout the match.
For Model 2, we first defined an algorithm for potential points. Potential point is a state quan-
tity related to five features: service points, break points, continuous scoring, unforced errors, and
distance run. We used Principal Component Analysis (PCA) on five features to extract three prin-
cipal component vectors as input features and built a momentum model based on Support Vector
Machines (MSVM). This model was used to obtain the single-game momentum and accumulated
set momentum for two players. We conducted a correlation analysis between the calculated single-
game momentum and the single-game winning probability obtained from Model 1, yielding corre-
lation values of 0.704 and 0.729. This indicates that a player’s success is not random but related to
momentum.
For Model 3, we built an ARIMA time series forecasting model. Using the momentum data
from model 2, we obtained predictions for the accmulated momentum of the two players. The
results suggestted that Carlos Alcaraz would have higher momentum, making it more likely for
him to win. Following this, we offered advice on how to capitalize on momentum shifts during
a match, focusing on seizing opportunities for continuous scoring, handling service games, and
converting break points.
Next, we tested and verified our model using data from another match in the Wimbledon Cham-
pionships (match 1502). The model’s calculations of accumulated momentum closely matches the
match flow. We also provided general advice for modifying the model for different types of com-
petition. After conducting a sensitivity analysis, the results shows that the momentum calculation
model is highly sensitive to various input features.
Finally, we wrote a memo for tennis coaches. The memo summarizes our model, explains the
role of momentum, and offers recommendations.
Keywords: Momentum; RNN; PCA; The MSVM Model; ARIMA
![](https://csdnimg.cn/release/download_crawler_static/89271471/bg2.jpg)
Contents
1 Introduction 2
1.1 Problem Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Restatement of the Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Ourwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Assumptions and Justification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Notations 4
3 Establishment and Solution of Single Game Winning Probability Model 4
3.1 Single Game Winning Probability Model Based on Recurrent Neural Network (RNN) 4
3.1.1 Serving Player Winning Probability Model . . . . . . . . . . . . . . . . . 5
3.1.2 Receiving Player Winning Probability Model . . . . . . . . . . . . . . . . 7
3.2 Visualization of the Match Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4 Establishment and Solution of the Momentum Model 8
4.1 Feature Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.2 Potential Point Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.3 Principal Component Analysis (PCA) Dimensionality Reduction . . . . . . . . . . 10
4.4 Momentum Model Based on Support Vector Machines (SVM) . . . . . . . . . . . 12
4.5 Statistical Test of Factors Affecting Momentum . . . . . . . . . . . . . . . . . . . 14
4.5.1 The Impact of Momentum on the Probability of Winning a Match . . . . . 14
4.5.2 The Impact of other Factors on Momentum . . . . . . . . . . . . . . . . . 15
5 Predictive Model for Shifts in Momentum 15
5.1 Accumulated Momentum Prediction Model Based on ARIMA . . . . . . . . . . . 16
5.2 Advice Based on Shifts in Momentum . . . . . . . . . . . . . . . . . . . . . . . . 17
6 Verification and Application of the Model 18
6.1 Verification within the Same Type of Competition . . . . . . . . . . . . . . . . . . 18
6.2 Application to other Types of Competitions . . . . . . . . . . . . . . . . . . . . . 19
7 Sensitivity Analysis 19
8 Strengths and Weaknesses 20
8.1 Strengths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
8.2 Weaknesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Appendices 23
1
![](https://csdnimg.cn/release/download_crawler_static/89271471/bg3.jpg)
Team # 2418251 Page 2 of 23
1 Introduction
Billy Shirley once said:“Watch Out for Momentum : There may not even be such
a thing in sports, but you can be sure you’ll be told whenever it shifts.”
1.1 Problem Background
Figure 1: Novak Djokovic & Carlos Alcaraz
[1]
In the 2023 Wimbledon Gentlemen’s cham-
pionship, 20-year-old Carlos Alcaraz emerged
victorious against 36-year-old Novak Djokovic,
handing Djokovic his first Wimbledon defeat
since 2013 and halting his remarkable run in
Grand Slam tournaments. The final was filled
with dramatic turns of performance. Djokovic
started strong by claiming the first set 6-1, only
for Alcaraz to make a comeback, securing the
second set through a tie-breaker and dominating
the third set 6-1. Although Djokovic managed
to take the fourth set 6-3, Alcaraz ultimately
clinched the win with a 6-4 score in the decid-
ing fifth set. The game showcased notable shifts
in momentum, underscoring the difficulty in quantifying how various events within the match af-
fect a player’s or team’s performance. Yet the omnipresence of momentum is so pervasive that it’s
acknowledged not solely by commentators, mentors, and competitors but also sensed by a laid-
back spectator lounging on their sofa. So for coaches, comprehending precisely what momentum
is can be vital, as this understanding will aid them in directing the game.
1.2 Restatement of the Problem
Considering the background information and restricted conditions identified in the problem
statement, we need to solve the following problems:
• Problem 1:Develop a tennis match flow model to assess player performance, considering
serve advantage and visualize the match dynamics.
• Problem 2:Assess the tennis coach’s skepticism about the role of “momentum” versus ran-
domness in match outcomes using a performance model/metric.
• Problem 3:Develop a predictive model for tennis match swings and advise on strategy con-
sidering past “momentum” differentials.
• Problem 4:Test and validate the tennis swing prediction model on other matches, identify
performance factors, and assess generalizability to different contexts.
• Problem 5:Include a page memo summarizing results with advice for coaches on the role of
“momentum”, and how to prepare players to respond to events that impact the flow of play
during a tennis match.
![](https://csdnimg.cn/release/download_crawler_static/89271471/bg4.jpg)
Team # 2418251 Page 3 of 23
1.3 Ourwork
Figure 2: Our Work
1.4 Assumptions and Justification
• Assumption 1:Assume that after each set, the athlete’s accumulated momentum is updated.
Justification:After each set, the athlete gets a chance to rest, stabilize, and the coach will
provide corresponding strategies, thus it’s appropriate to update the athlete’s momentum.
• Assumption 2:Ignore the impact of tiebreakers on momentum trends.
Justification:In tiebreakers, the situation changes dramatically with alternating service rights,
and the athletes’ mental states are highly tense, so the influence of momentum is minimal.
Therefore, the impact of tiebreakers on momentum trends can be neglected.
• Assumption 3:The momentum in each game is discrete.
Justification:The model considers momentum to be discrete within each game, as there
is no more continuous data available to accurately depict it. Therefore, the calculation of
momentum is based on discrete intervals within the games.
• Assumption 4:Ignore the impact of environmental factors and athletes’ psychological states.
Justification:The environment of the court, wind speed, and the athletes’ psychological
states can all have an impact on the progression of the match. However, since no specific
data is provided in the question and psychological states cannot be quantitatively described,
the influence of these factors is therefore neglected.
剩余23页未读,继续阅读
资源评论
![avatar-default](https://csdnimg.cn/release/downloadcmsfe/public/img/lazyLogo2.1882d7f4.png)
![avatar](https://profile-avatar.csdnimg.cn/fcd62adb0120465d9af280215b0ff722_snowtshan.jpg!1)
阿拉伯梳子
- 粉丝: 1662
- 资源: 5735
上传资源 快速赚钱
我的内容管理 展开
我的资源 快来上传第一个资源
我的收益
登录查看自己的收益我的积分 登录查看自己的积分
我的C币 登录后查看C币余额
我的收藏
我的下载
下载帮助
![voice](https://csdnimg.cn/release/downloadcmsfe/public/img/voice.245cc511.png)
![center-task](https://csdnimg.cn/release/downloadcmsfe/public/img/center-task.c2eda91a.png)
最新资源
- 【后端开发框架】教程&案例&相关项目
- setup asdfasfd
- 信号与系统、数字信号处理、通信原理等课程内容及相关实验项目
- 简鹿视频格式转换器 1.0 离线包.zip
- 《javascript网页编程》项目考试要求.doc
- 基于python实现的机器学习算法
- yolov8实战第十天-pyqt5-yolov8实现数据结构化、yolov8目标跟踪、过线检测计数系统(参考论文完整部署代码)
- C# opencvsharp对Mat数据进行序列化或者反序列化以及格式化输出演示源码.7z
- 主机性能指标采集器,感兴趣的运维小伙伴们可以下载试试
- 计算机网络和现代通信组网是信息科学和电子工程领域的重要分支,涉及大量的理论知识和实践技能 以下是一些学习资源和项目思路,可以帮助
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
![feedback](https://img-home.csdnimg.cn/images/20220527035711.png)
![feedback](https://img-home.csdnimg.cn/images/20220527035711.png)
![feedback-tip](https://img-home.csdnimg.cn/images/20220527035111.png)
安全验证
文档复制为VIP权益,开通VIP直接复制
![dialog-icon](https://csdnimg.cn/release/downloadcmsfe/public/img/green-success.6a4acb44.png)