Six-game prediction method: based on a certain recent trend of the team

Six-game prediction method: based on a certain recent trend of the team

  • 5758
  • 原文

The previous article was "Static Prediction Method (2): Goal Rate Prediction Method", and today is "Static Prediction Method (3): Six-game Prediction Method".


Six-game Prediction Method


The Six-game Prediction Method is a simple prediction method provided by the British media to lottery players. The basic condition for the prediction of this method is that a certain recent trend of a team will maintain continuity. For example, if a team has won 6 consecutive games at home, use W as the winning symbol, then the recent score is WWWWWW; if the away team has lost 6 consecutive games, use L as the losing symbol, Then the recent performance of the visiting team is LLLLLL; the prediction result is that the home team wins. For another example, assuming that the two participating teams have played 4 draws in the last 6 games, and X is used as a tie symbol, then the recent game performance is XWXXLX, and the two teams will have a draw in the upcoming game. .

微信图片_20221227142943.jpg

This method is based on the continuity of the team's state performance. Although it has some truth, it is extremely one-sided. For example, in the previous example, the situation where both teams happen to be WWWWWW vs. LLLLLL is almost a once-in-a-century situation. Furthermore, it is difficult to find a definition criterion for this method in actual operation, so its application is greatly limited. The rationality of the six-match prediction method is that the method is based on the performance of the team in a stage to predict the base point, so it can better reflect the probability of the team's victory or defeat under the decision of the competitive state. To this end, some gaming professionals have improved the six-field prediction method, making it widely used.


1. How to use the Six-game Prediction Method


The improved method is to set the difference value according to the strength or state difference of the two participating pairs, and then predict the result of the game according to the difference value. The specific calculation method is: in the results of the last 6 games of a team, every win (that is, 1 W) is counted as 3 points, every tie (that is, 1 X) is counted as 1 point, and every loss 1 field (i.e. 1 L) counts as 0 points. If a team has won all the previous 6 games, that is, according to the six-game prediction method, the team will score 18 points; if it is LLLLLL, it will be counted as 0 points; if it is WLXXWW, it will be counted as 11 points.


2. Prediction definition


The Six-game Prediction Method is also restricted by home and away factors and team strength. Using the improved 6-match prediction method, nearly 600 matches of the Premier League (microblog topic) and Serie A (microblog topic) league games were predicted, and the following prediction definitions were obtained.


Definition 1: When the point difference between the two teams in the six games is 6 or more, the team with the higher points in the six games wins;


Definition 2: When the point difference between the two teams in the six games is 5, if the home team has the highest points in the six games, the home team wins; if the home team has the lowest points in the six games, the home team wins or draws;


Definition 3: When the point difference between the two teams in the six games is 2-4, the team with the higher points in the six games wins.


Definition 4: When the point difference between the two teams in six games is 1 or 0, the home team wins or draws.


3. The accuracy rate of the Six-game Prediction Method


1. Overall accuracy


A total of 321 Premier League games were predicted from the seventh round, of which 182 games were correctly predicted, with an accuracy rate of 56.7%. A total of 252 games were predicted from the seventh round of Serie A, of which 136 games were predicted correctly, with an accuracy rate of 54.0%.


2. Specific accuracy


In the Premier League, the accuracy rate is 42.7% when the point difference in six games is 6 or more; the accuracy rate is 70.6% when the point difference in six games is 5; The field integral difference is 1 or 0, and the accuracy rate is 77.5%. When the score difference of six games in Serie A is 6 or more, the accuracy rate is 48.1%; when the score difference of six games is 5, the accuracy rate is 57.1%; When the field integral difference is 0 or 1, the accuracy rate is 78.0%.


In the Premier League, the points that are significantly lower than the overall accuracy rate are when the six-match point difference is 2, 7, 8, or more than 10;


The number of Premier League games with six points of 2, 7, 8 and 10 or more accounted for 27.7% of the total number of games; the number of Serie A games with six points of 2, 3, and 8 accounted for 30.9% of the total number of games. That is to say, for the Premier League, 72.3% of games can be predicted more accurately. For Serie A, 69.1% of games can be predicted more accurately. Therefore, the six-match forecasting method is applied to Premier League matches, and 7 out of 10 matches in each round can be predicted more accurately. However, when applied to Serie A games, out of 9 games in each round, 6 games can be predicted more accurately.


The above three static prediction methods all appear in the form of mathematical models. They predict the outcome of the game from the game itself, and can also be said to be a process of imitating bookmakers to draw odds conclusions. Although the prediction methods we know are not necessarily the ultimate basis for the odds issued by the bookmaker, from a theoretical point of view, the odds model of the bookmaker is also based on these three objective data models. Therefore, it is found that they The commonality between them can effectively predict the result of the game, and its auxiliary role should not be underestimated. Another point is that some highly savvy players will use these three mathematical models to combine with handicap or odds analysis methods to form their own analysis models.