- 學位論文
應用相關連動賠率之美國職籃博弈預測系統
A Related Link Odds Prediction System of National Basketball Association Betting
摘要
本論文提出了一種新的預測美國職業籃球比賽投注輸贏的新方法，我們只分析變動的賠率數據，而不考慮所選之球隊強或弱或近期的戰績與發展，此系統還能改善一般複雜的分析。 本論文分為四個部分，首先我們在不同時間收集數據，並將讓分值和各種賠率輸入系統中。再來我們根據分類規則中不同的讓分值和不同時間點所形成的賠率趨勢對輸入的數據進行分類。第三部分，我們通過自己開發的座標系統來分析各種趨勢種類，此步驟將會比較輸入的賠率和標準值的賠率以找出差異。之後我們還會找出一些特別的賠率值。最後我們將會通過此系統獲得所期望的預測結果。 本論文的研究貢獻如下: 1. 我們的數據只採用讓分值和賠率，因此在預測過程中我們不會因為個人情緒而導致預測錯誤。 2. 本系統的分析方法比一般常見的系統簡單易懂，因此可以讓玩家更容易理解上手。 3. 使用此系統可以獲得穩定且不差的勝率和穩定高的利潤。
並列摘要
In this thesis, we will propose a new method to predict the wins or loses of the basketball games betting in NBA. We only analyze the data of mutative odds without considering if the chosen team is stronger than the opposite team- or the recent developments of the two teams. It can also improve the generally complex analysis. There are four parts in our proposed system. At first, we collect data at different time and input the point spreads and various odds into the system. Second, we classify the entered data by it’s different point spread and the trend of odds in each time in classification rules. Third, we analyze the trends classifications by the coordinate systems developed by the authors. It will compare the odds which are entered and the standard values to find out the discrepancy. After that, we recognize a few special data. Finally, the value of forecast is obtained by the proposed system. The contributions in this thesis are as follows: 1. We only adopt the data of point spreads and odds. Therefore, we will not make mistakes by our sentiment during predicting. 2. The analysis method of our system is simpler than common systems. Thus the players can understand more easily. 3. Using our system, the winning rate is steady and the profit is continuously stable and high.