WO2023053882A1

WO2023053882A1 - Betting ticket purchasing support system, betting ticket purchasing support method, and betting ticket purchasing support program

Info

Publication number: WO2023053882A1
Application number: PCT/JP2022/033618
Authority: WO
Inventors: 大下永; 俊資石井
Original assignee: 大下永; 俊資石井
Priority date: 2021-09-29
Filing date: 2022-09-07
Publication date: 2023-04-06
Also published as: JP2023049274A

Abstract

Provided is a betting ticket purchasing support system that presents and encourages the purchasing of a betting ticket that can satisfy the wishes of a purchaser by inputting various indicators disclosed in racers to improve the accuracy of predicting wins and losses. The system is provided with: a basic race data acquisition unit that acquires basic race data including statistical information, odds information, and race result information on past racers in a race to be bet on; a reference position prediction data generation unit that estimates and generates, as reference predicted position data, positions among a plurality of racers participating in the race; a pre-race position prediction data generation unit that estimates and generates, as pre-race predicted position data, positions on the bases odds information on each racer; a pre-race predicted odds generation unit that estimates and generates pre-race odds as pre-race predicted odds data; and a probability prediction unit that predicts a winning probability of a betting ticket set for a racer at the time when the betting ticket is purchased to generate probability prediction result data.

Description

BOTTING TICKET PURCHASE SUPPORT SYSTEM, BOTTING TICKET PURCHASE SUPPORT METHOD, AND BOTTING TICKET PURCHASE SUPPORT PROGRAM

The present invention relates to a betting ticket purchase support system, a betting ticket purchase support method, and a betting ticket purchase support program, and more particularly to a betting ticket purchase support system, method, and program for finding advantageous conditions for purchasing betting tickets in a competition. .

For example, in public competitions such as horse racing, bicycle racing, boat racing, etc., the ratio of payouts to bets is called odds. For example, individual odds are displayed on a betting ticket such as a betting ticket before a competition (race) is held. Purchasers of betting tickets for competitions are interested in high payback amounts for bets, high odds (magnification), and the like.

In this case, the organizers of the competition will release a large number of indicators, such as the odds of the betting tickets, the popularity of the racehorses and other competitions in the case of horse racing, and the number of purchasers of the betting tickets. In addition, various indicators fluctuate even before the competition starts. Therefore, it is not easy for the purchaser to predict the development of the competition before the competition starts and to purchase the most suitable betting ticket that meets his or her wishes.

Therefore, devices and systems have been proposed for the purchaser of the voting ticket to select and support the purchase of the voting ticket (see

Patent Documents

1 and 2, etc.). The devices, systems, etc., are designed to perform a certain function as a guideline for the purchase of voting tickets and the like by purchasers.

However, the types of indicators (data) that can be handled are limited in the technologies represented by patent documents. Therefore, the accuracy of the presentation of the voting ticket derived from the index, data, etc. related to the competition was not sufficient.

Japanese Patent Application Laid-Open No. 2004-199479 JP 2019-120977 A

The present invention has been made in view of the above points, and by incorporating a plurality of various indicators disclosed in the competition, the accuracy of predicting the outcome of the competition in the competition is improved, and the betting ticket that can satisfy the wishes of the purchaser. To provide a voting ticket purchase support system, method and program for presenting and prompting purchase.

That is, the voting ticket purchase support system of the embodiment includes statistics information of the competition, odds information for the competition, competition of the competition of the competition in the competition executed by the competition targeted for voting, A competition basic data acquisition unit that acquires competition basic data including result information, and a reference ranking prediction that estimates the mutual ranking of a plurality of competitions participating in the competition based on the competition basic data and generates reference prediction ranking data. Pre-competition generating pre-competition prediction ranking data by estimating the pre-competition ranking of the competition based on the reference prediction ranking data and the odds information for each competition in the competition to be conducted by the competition, and a data generation unit A ranking prediction data generation unit, a pre-competition prediction odds generation unit for estimating pre-competition odds of a competition in a competition to be held by the competition based on the competition basic data and generating pre-competition prediction odds data, and pre-competition prediction ranking. Based on data and pre-competition prediction odds data, predict the probability of winning the voting ticket set for the competition in the competition that the competition will be held in the future, and a probability prediction unit that generates probability prediction result data.

Furthermore, a betting ticket purchase unit is provided for generating betting ticket data to be purchased for purchasing a betting ticket whose purchase is determined based on the probability prediction result data from the organizer of the competition held by the sporting event to be voted for. may be

Furthermore, the competition basic data acquisition unit may acquire the competition basic data from the organizer of the competition held by the competition subject to vote.

Furthermore, the game basic data acquisition unit may include a statistical data generation unit that generates statistical data for each game from the game statistical information, the odds information for the game, and the game result information for the game.

Furthermore, the game basic data acquisition unit may include a feature amount generation unit that generates a game item feature amount for each game item in the game that the game item will be implementing in the future.

Further, the reference ranking prediction data generation unit, when estimating the ranking among a plurality of competitions participating in the competition based on the competition basic data, selects the most popular competition in consideration of the odds information for the competition. may be prioritized to generate the predicted ranking data, or the most popular competitor may be excluded to generate the predicted ranking data.

Furthermore, the probability prediction unit may generate probability prediction result data including the winning feature amount related to the betting ticket of the game that was successful in the past competition.

Furthermore, generation of reference prediction ranking data in the reference prediction ranking data generation unit, generation of pre-competition prediction ranking data in the pre-competition prediction ranking data generation unit, generation of pre-competition prediction odds data in the pre-competition prediction odds generation unit, and probability prediction Machine learning may be used in generating the probability prediction result data in the part.

According to the betting ticket purchase support system of the present invention, the statistical information of the competition, the odds information for the competition, and the competition result of the competition in the competition executed by the competition targeted for voting in the past A competition basic data acquisition unit that acquires competition basic data including information, and reference ranking prediction data that estimates the mutual ranking of multiple sports participating in a competition based on the competition basic data and generates reference prediction ranking data. a generating unit, and a pre-competition ranking that estimates the pre-competition ranking of the competition based on the reference prediction ranking data and the odds information for each competition in the competition that the competition will implement from now on, and generates pre-competition prediction ranking data; A predicted data generation unit, a pre-competition predicted odds generation unit that estimates pre-competition odds of a sport in a competition to be conducted by the sport based on the game basic data and generates pre-competition predicted odds data, and pre-competition predicted ranking data. And based on the pre-competition prediction odds data, predict the probability of winning the betting ticket set for the competition in the competition that the competition will be held in the future. and a probability prediction unit that generates prediction result data, and by taking in a plurality of various indicators disclosed in the competition, the accuracy of predicting the victory or defeat of the competition in the competition is improved, and the vote that can satisfy the purchaser's wishes. The ticket can be presented and the purchase can be urged. Similar effects can also be obtained in the voting ticket purchase support method and program.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic diagram which shows the structure of the voting ticket purchase assistance system of embodiment. It is a block diagram which shows the structure of the function part in the computer of a voting ticket purchase assistance system. It is a schematic diagram which shows the flow of a process of a voting ticket purchase assistance system. It is a flowchart which shows the flow of a process of a game basics data acquisition part. FIG. 5 is a flow chart showing the flow of processing of a reference rank prediction data generation unit; It is a flowchart which shows the flow of a process of a pre-game ranking prediction data production|generation part. It is a flowchart which shows the flow of a process of the prediction odds production|generation part before a competition. It is a flowchart which shows the flow of a process of a probability prediction part. It is a block diagram which shows the functional structure of a voting ticket purchase part. It is a flow chart explaining a voting ticket purchase support method of an embodiment. It is the 1st graph which shows the calculation result of the voting ticket purchase assistance system of embodiment. It is a 2nd graph which shows the calculation result of the voting ticket purchase assistance system of embodiment. It is the 3rd graph which shows the calculation result of the voting ticket purchase assistance system of embodiment.

The voting ticket purchase support system of the embodiment is a system that assists the purchaser in determining which sports and which voting content should be purchased in the competition executed by the sports to be voted for. is. This system is a system that enables the purchase of a desired betting ticket from the organizer of the competition. According to this voting ticket purchase support system, since various types of information accumulated in the past on the sports to be voted for are extensively included, it is possible to select the content of the competition, the individual information of the sports, and even the selection of the voting ticket. Even purchasers who are unfamiliar with voting can lower the entry barrier to purchasing voting tickets.

The competition here is a competition in which multiple competitions compete for ranking, such as public competitions. Specifically, "horse racing" such as central horse racing sponsored by the Japan Racing Association, local horse racing sponsored by the National Association of Horse Racing, and "boat racing" (motorboat racing) sponsored by local governments, local governments "Keirin" or "Auto race" sponsored by is listed. Furthermore, to the extent permitted by the legal system, sports lotteries such as soccer are also included. Of course, it is not limited to the listed types of competition.

In addition, "competition" means the subject who competes. When the type of competition is "horse racing", the competition items are horses and jockeys. Similarly, when the competition is "boat racing", the competition is a boat racer, when "bicycle", the competition is a bicycle racer, and when "auto racing", the competition is an auto racer. Furthermore, in the case of a sports lottery such as soccer, it is a team. In the description of the betting ticket purchase support system, method, and program of the embodiments, the “competition” is “horse racing,” the “event” is “horse,” and the “betting ticket” is “betting ticket.” do. "Odds" refers to the ratio of payouts to bets for publicly managed games. Naturally, it is permissible to change to the above-mentioned various competitions and competitions.

FIG. 1 is a schematic diagram showing the configuration of the voting ticket purchase support system 1 of the embodiment. In the betting ticket purchase support system 1, a competition (horse racing) venue 5 (a so-called racecourse, racecourse), a computer 3 (server) of the competition (horse racing) organizer 2, and a betting ticket purchase support system operator. 6 computers 10 (servers) are connected by an Internet line 4 . Therefore, various information such as the location (racecourse) of the competition to be held, the content (which race), the competition (horse) to participate in, the results of the competition, etc. are communicated between the organizer 2 and the venue 5, and these information Among them, the business operator 6 of the voting ticket purchase support system acquires the information disclosed by the organizer 2 from the computer 3 of the organizer 2 through the Internet line 4 .

The computer 3 of the organizer 2 and the computer 10 of the voting ticket purchase support system operator 6 are electronic computers (computation resources) such as known mainframes, workstations, and cloud computing systems. Note that the computer 10 also includes electronic computers (computation resources) such as personal computers, smart phones, and tablet terminals.

FIG. 2 is a block diagram showing the configuration of functional units in the computer 10 of the business operator 6 of the voting ticket purchase support system. Specifically, the computer 10 includes a CPU 11, a ROM 12, a RAM 13, a storage unit 14, an I/O 15 (input/output interface), and the like. Of course, the computer 3 of the host 2 also has the same configuration.

Each functional unit in the CPU 11 of the computer 10 is shown in the block diagram of FIG. Each functional unit includes a competition basic data acquisition unit 110, a reference ranking prediction data generation unit 120, a pre-competition ranking prediction data generation unit 130, a pre-competition prediction odds generation unit 140, a probability prediction unit 150, a betting ticket purchase unit 160, and the like. . The operation and execution of the computer 10 are implemented by software such as a betting ticket purchase support program loaded in the main memory.

When the functional units of the computer 10 in FIG. 2 are implemented by software, the computer 10 is implemented by executing instructions of a program, which is software that implements each function. A "non-temporary tangible medium" such as a CD, a DVD, a semiconductor memory, a programmable logic circuit, or the like can be used as a recording medium for storing this program. Also, this program may be supplied to the computer 10 of the operator 6 of the voting ticket purchase support system via any transmission medium (communication network, broadcast wave, etc.) capable of transmitting the program.

The storage unit 14 of the computer 10 is a known storage device such as an HDD or SSD. The storage unit 14 may be an external server (not shown). The storage unit 14 stores various data, information, a voting ticket purchase support program, various data necessary for executing the program, and the like. Also, each functional unit that executes various calculations, calculations, etc. is a computing element such as the CPU 11 . In addition, an input device (not shown) such as a keyboard, mouse, etc., a display unit (display device such as a display), an output device for outputting data, etc. may also be suitably connected to the I/O 15 of the computer 10. .

FIG. 3 is a schematic diagram showing the flow of processing in each functional unit in the voting ticket purchase support system 1 of the embodiment. First, the game basic data acquisition unit 110 (data generation) acquires and generates the game basic data necessary for subsequent processing. Next, reference prediction ranking data that serves as a reference is generated in the reference ranking prediction data generation unit 120 (fundamental prediction). Subsequently, the pre-competition prediction ranking data of the relevant competition is generated in the pre-competition ranking prediction data generation unit 130 (technical prediction). In addition, the pre-competition predicted odds generation unit 140 (final odds estimation) generates competition measured odds data. Then, the probability prediction unit 150 (betting ticket probability estimation) generates betting ticket purchase data. Based on a series of calculation results, specific betting ticket purchase information is generated in the betting ticket purchase unit 160 (purchase betting ticket determination).

After that, the planned purchase betting ticket data (information on the betting ticket purchase plan) is accumulated in the purchase planned betting ticket database 513, and the purchase of the desired betting ticket (betting ticket) is executed from the organizer of the competition via the Internet line 4 in the process batch 514. be. Henceforth, it demonstrates separately.

The game basic data acquisition unit 110 collects statistical information of the game (horse) of the game (horse) in the game (horse) executed by the game (horse) to be voted. , odds information for the game (horse), and game basic data including game result information for the game (horse). Specifically, the weight of the horse that is the competition as a record of past races, the win or loss in the race, the odds record of the horse, the information on the refund of the betting ticket, and other information on the racecourse where the race is held (running length), etc. Furthermore, the condition of the racecourse (racecourse) that serves as the venue, the character of the competition (horse), the pedigree, etc. may be added to the competition basic data.

FIG. 4 is a flowchart showing the flow of processing by the game basic data acquisition unit 110 (data generation). The data acquisition unit 111 can periodically acquire information from the organizer of the competition to be voted on, for example, "JRA-VAN" or "JRDB" of the Japan Racing Association. As information to be acquired, the types of races and horses to be entered are acquired at any time. The odds and the index of the horse's popularity are obtained 10 minutes before, 5 minutes before, and at the time of the deadline for purchase of voting tickets in the competition (race), respectively.

The acquired competition basic data is registered and updated in various databases (DB) such as the race information runner database 501 and the odds popularity database 502.

The information acquired by the data acquisition unit 111 is transferred to the statistical data generation unit 112 (index statistical data generation). The statistical data generation unit 112 generates statistical data for each game from the game statistical information, the odds information for the game, and the game result information for the game.

The statistical data generation unit 112 (index statistical data generation) in FIG. 4 generates various evaluation indices that are considered to influence the development of the race (horse racing). For example, it is the running power (running speed), weight, body length, and race development in past races (leading type, catch-up type, etc.) of the horse that is the competition. These are calculated as averages, deviation values, etc. from past race results. In addition, time information such as race time, finish time, front 3rd floor time, rear 3rd floor time, rear 4th floor time, etc., from past races is also acquired, and averages, deviations, etc., are calculated as time evaluation indexes from these. .

Various statistical data generated by the statistical data generation unit 112 are registered and updated in various databases (DB) of the statistical index database 503 and the statistical database 504. Statistical data registered and updated in these databases (DB) is basic competition data.

Various statistical data generated by the statistical data generation unit 112 are transferred to the feature amount generation unit 113 (feature amount data generation). The feature quantity generation unit 113 generates a game object feature quantity for each game object in a game (horse race) to be held by the game object (horse).

In the feature amount generation unit 113 (feature amount data generation) in FIG. 4, various The statistical data of is adjusted to data that emphasizes the feature amount. As the statistical data to be supplied to the reference ranking prediction data generation unit 120 (fundamental prediction), the feature amount (game feature amount) is prepared from the statistical data that does not change after the horse weight is announced. In addition, as the statistical data to be supplied to the pre-competition prediction data generation unit 130 (technical prediction), feature amounts (competition feature amounts) centered on statistical data that fluctuate until the voting deadline of the betting ticket (betting ticket) of the race is prepared. When averaging statistical data, for example, taking the winning rate as an example, the ranking of the last 10 races may be weighted higher.

Various feature amounts (game feature amounts) generated by the feature amount generation unit 113 are registered and updated in various databases (DB) of the fundamental prediction feature amount database 505 and the technical prediction feature amount database 506. The feature amount (game item feature amount) registered and updated in these databases (DB) is game basic data.

A reference ranking prediction data generation unit 120 (fundamental prediction) calculates a plurality of competition objects (horses) participating in a competition (race) based on the competition basic data generated by the competition basic data acquisition unit 110 (data generation unit). A mutual ranking is estimated and generated as reference prediction ranking data. In particular, the reference rank prediction data generator 120 is a basic ranking prediction that does not depend on popularity, odds, etc., and gradient boosting rank learning is used. The reference ranking prediction data generator 120 estimates the mutual ranking of a plurality of competition objects (horses) participating in the competition based on the competition basic data.

FIG. 5 is a flowchart showing the processing flow of the reference rank prediction data generation unit 120 (fundamental prediction). The game basic data generated by the game basic data acquisition unit 110 (data generation unit) is stored in the race information runner database 501, the statistical index database 503, the statistical database 504, and the fundamental prediction feature amount database 505, as shown in FIG. It is In the reference ranking prediction data generation unit 120 of the embodiment, central horse racing, local horse racing (Monbetsu, Kanazawa, Nagoya, Saga, Kochi), Hyogo horse racing (Sonoda, Himeji), Nankan horse racing (Funabashi, Kawasaki, Urawa, Oi), Iwate It is constructed as statistical data for each racecourse of horse racing (Morioka, Mizusawa) and Obihiro horse racing. Examples described later show the results of calculations and the like relating to central horse racing.

The basic competition data stored in each database is incorporated into two types of models and calculated according to the recovery rate. The first is the collection rate-oriented model section 122 and the second is the hit rate-oriented model section 123 . Machine learning, which will be described later, is applied to the computation of both models. Both models are calculated from various data stored in the race information runner database 501 , the statistical index database 503 , the statistical database 504 , and the fundamental prediction feature amount database 505 . In both models, as the same data, only the recovery rate may be changed to a predetermined value (multiplied by a coefficient), or the type of original data may be changed for calculation.

In the recovery rate-oriented model unit 122, arithmetic processing is performed to increase the rank of lower-popular voting tickets (betting tickets) so that the rate of recovery from betting is higher than the amount invested (high return). The collection rate-oriented model section 122 is a so-called long shot model. Specifically, based on the competition basic data, the odds for the most popular competition (horse) are low, so we intentionally exclude the most popular competition (horse) and choose the second most popular, or even the least popular. However, for example, the inclination is adjusted in consideration of inclusion of a sporting object (horse) with an increased winning record in the most recent race.

The hit rate-oriented model unit 123 emphasizes the winning rate of betting (low return), and performs arithmetic processing to increase the rank of the betting ticket (betting ticket) that is the most likely to win in the race. The hit-rate-emphasis model unit 123 is a model aiming to hit the order of arrival with certainty. Specifically, although the odds for the most popular race (horse) will be low based on the basic competition data, the emphasis will be placed on the finish order result, which will increase the hit rate while maintaining the collection rate. The slope is adjusted so that the percentage of votes for (horse) is preferentially increased.

In the collection rate-oriented model unit 122 or the hit rate-oriented model unit 123, adjustments are made so as not to include feature amounts that are likely to overlearn or feature amounts that fluctuate after the horse weight is announced. In the fundamentals statistical data generation unit 124, from the results of the arithmetic execution in the collection rate-oriented model unit 122 or the hit rate-oriented model unit 123, the pre-competition ranking prediction data generation unit 130 (technical prediction). Ranking estimates between sports are generated to generate reference predicted ranking data (fundamental statistical data).

Specifically, the reference prediction ranking data is a prediction of the order of finish of the horses to be held in the race to be held, taking into consideration the result of the execution of calculations by the collection-rate-oriented model unit 122 or the hit-rate-oriented model unit 123 . At this stage, the ranking of the competition is predicted based on various information and data obtained roughly before the start of the race. The reference prediction ranking data is registered and updated in the database (DB) of the fundamental prediction result database 507 .

The pre-competition ranking prediction data generation unit 130 (technical prediction) estimates the pre-competition ranking of the competition of the competition based on the odds information for each competition in the competition that the competition will implement from now on and the reference prediction ranking data. Generated as pre-prediction ranking data. In the pre-competition ranking prediction data generation unit 130 (technical prediction), the feature amounts that fluctuate until the competition (race) deadline, such as the popularity of the competition (horse), the odds of the betting ticket (betting ticket), etc., and the above-mentioned fundamental As a result of the race prediction, statistical data is used to predict the race. In other words, it is added for the purpose of last-minute final adjustment in order to improve the accuracy when selecting a betting ticket (betting ticket).

FIG. 6 is a flow chart showing the processing flow of the pre-competition ranking prediction data generation unit 130 (technical prediction). In the pre-competition prediction data generation unit 130 (technical prediction), the above-mentioned race information runner database 501, fundamental prediction result database 507, technical prediction feature amount database 506, odds popularity database 502, and statistical database 504 are stored. The technical learning model unit 132 performs ranking prediction calculations based on the various data received.

The technical learning model unit 132 obtains the 1st, 2nd, and 3rd finish rates (which horses will finish what) for the events (each runner) of the scheduled race. Therefore, as a result of the fundamentals prediction, in addition to statistical data, odds five minutes before the voting deadline are used as feature amounts. The past odds and popularity data used as feature amounts during learning in the technical learning model unit 132 are data up to 5 minutes before the start deadline. This is because by limiting the data to be used at intervals of time, learning is performed so that new self calculation results are not swayed around by leaking the self calculation results. For the model division and machine learning method of the technical learning model unit 132, the calculation method in the case of the aforementioned reference rank prediction data generation unit 120 (fundamental prediction) is applied.

Based on the prediction result of the order of finish calculated by the technical learning model unit 132, the technical statistical data generation unit 133 performs pre-competition prediction of the prediction result for use in predicting the betting ticket to be purchased (recommended betting ticket). Ranking data (technical statistics data) is generated. Pre-competition predicted ranking data is registered and updated in the database (DB) of the technical prediction result database 508 .

The pre-competition predicted odds generation unit 140 (final odds estimation) estimates the pre-competition odds of the event in the event that the event will be held based on the game basic data, and generates pre-competition prediction odds data. Note that in the embodiment, in addition to the competition basic data, reference predicted ranking data and pre-competition predicted ranking data are included.

FIG. 7 is a flow chart showing the processing flow of the pre-game predicted odds generator 140 (final odds estimation). The pre-competition prediction odds generation unit 140 (final odds estimation) is based on various data stored in the aforementioned race information runner database 501, technical prediction result database 508, fundamental prediction result database 507, and odds popularity database 502. Then, final odds are calculated in the final closing odds estimating unit 142 .

　The odds 10 minutes before the start of the competition (race), the odds 5 minutes before the start of the competition, the results of the above predictions, and the final winning odds and double winning odds of the competition (each runner in the relevant race) are estimated from the statistical data. For example, if 10 horses are scheduled to run in a certain race, ie, the 1st horse to the 10th horse, the odds for the 1st horse to the 10th horse are disclosed. For example, if the 4th horse is predicted to finish 1st, the 7th horse is predicted to finish 2nd, and the 3rd horse is predicted to finish 3rd, the odds are 1.4 for 4th, 3.6 for 7th, and 3.6 for 3rd. Horses: A number such as 6.1 is generated as pre-competition predicted odds data. Of course, the various estimated odds also take into account the odds according to the method of purchasing the betting ticket, such as single wins and multiple wins.

Based on the pre-competition odds calculated by the final deadline odds estimation unit 142, the purchase statistical data generation unit 143 generates purchase statistical data on the types of betting tickets (betting tickets) specifically recommended for purchase in the race. is generated. Purchase statistical data is registered and updated in the database (DB) of the final odds prediction result database 509 . In purchase statistics data, the estimated final odds of each horse in the race can be retrieved.

A probability prediction unit 150 predicts the winning probability of a betting ticket when purchasing a betting ticket set for a sporting event in a sporting event to be held based on pre-competition predicted ranking data and pre-competition predicted odds data. Then, the amount of money to be refunded due to the purchase amount of the betting ticket is predicted from the hit probability, and the probability prediction result data for the game to be held by the game is generated. In addition, the probability prediction unit 150 generates probability prediction result data including the winning feature amount related to the betting ticket of the game that was hit in the past competition.

FIG. 8 is a flowchart showing the processing flow of the probability prediction unit 150 (betting ticket probability prediction). The probability prediction unit 150 (betting ticket probability prediction) is based on various data stored in the above-mentioned race information runner database 501, technical prediction result database 508, fundamentals prediction result database 507, and final odds prediction result database 509. , a prediction calculation is performed in the hit rate prediction unit 152 (betting ticket hit rate prediction) regarding the winning of the betting ticket (betting ticket) in the race.

In the hit rate prediction unit 152 (betting ticket hit rate prediction), hits are collated from predictions and race results in races held in the past, and learning is repeated between predictions and hit betting tickets. A relationship found between a past prediction result and an actual winning betting ticket is calculated as a winning feature amount. Therefore, based on the accumulation of past data, a prediction calculation is performed to determine whether or not each combination betting ticket of a race to be held from now on will hit, and the prediction of the winning probability of each individual betting ticket (betting ticket) is calculated.

Based on the hit probability prediction data calculated by the hit rate prediction unit 152, the recovery rate calculation unit 153 (statistical data + recovery rate calculation) calculates the target of the betting ticket when the betting ticket is purchased. A probabilistic prediction result data is generated regarding the amount of money (recovery rate) that can be obtained within. Specifically, from the product of the winning probability of a betting ticket (betting ticket) calculated from the winning probability prediction data and the dividend of the betting ticket (betting ticket) at the time of prediction, is calculated. The recovery rate referred to here is the quotient obtained by dividing the dividend amount by the purchase amount.

The probability prediction result data is registered and updated in the database (DB) of the probability prediction result database 510. When the probability prediction result data is generated, the information of the payout data 511 (via the database DB) provided by the organizer of the competition to be voted may be added and generated. In the probability prediction result data, it is possible to retrieve the winning probability, recovery rate, and dividend amount of each betting ticket of each horse in the race. Finally, a matching betting ticket (betting ticket) is retrieved within the range of the set target hit probability and target recovery probability.

A voting ticket purchase unit 160 (purchase voting ticket determination) purchases a voting ticket, the purchase of which has been decided based on the probability prediction result data, from the organizer of the competition in which the sport to be voted for is to be purchased. Generate data.

FIG. 9 is a flowchart showing the flow of processing of the voting ticket purchase unit 160 (determination of purchasing voting ticket). Based on various data stored in the race information runner database 501, the user information/setting database 512, and the probability prediction result database 510, the purchased betting ticket determination unit 162 (purchased betting ticket determination logic) votes in the race. Calculations for specifically specifying the types and combinations of tickets (betting tickets) are executed to generate data on betting tickets to be purchased. The user information/setting database 512 stores information on how to purchase a voting ticket desired by a user (user) of the voting ticket purchase support system 1. Furthermore, information such as respective ratios when both are combined is accumulated as set values.

The purchased betting ticket determination unit 162 (purchased betting ticket determination logic) selects a betting ticket so that the synthetic odds, synthetic hit rate, synthetic recovery rate, etc., are close to the target values set by the user according to the user's set values. (Betting ticket) combination is determined and generated as purchase planned betting ticket data. The voting ticket data to be purchased is registered and updated in the database (DB) of the voting ticket database 513 to be purchased. In the scheduled purchase betting ticket data, it is possible to search for the type of betting ticket (betting ticket) determined and selected for each user by the purchase betting ticket determination unit 162 (purchase betting ticket determination logic) as the expected purchase betting ticket in the race.

　Machine learning is used for the calculations and calculations in each functional part detailed so far. Specifically, the generation of reference prediction ranking data in the reference prediction ranking data generation unit 120 (fundamental prediction) in FIG. Generation, generation of pre-game predicted odds data in the pre-game predicted odds generation unit 140 (final odds estimation) in FIG. 7, generation of probability prediction result data in the probability prediction unit 150 (betting ticket probability estimation) in FIG. 8, and betting ticket When the purchasing unit 160 (determining a betting ticket to purchase) generates betting ticket data to be purchased, machine learning is executed based on various data accumulated in each database. In addition, machine learning is also executed when generating statistical data in the statistical data generation unit 112 of the game basic data acquisition unit 110 (data generation unit) in FIG. be done.

For machine learning, methods such as Support Vector Machine (SVM), model tree, decision tree, neural network, multiple linear regression, local weighted regression, probability search method, etc. are used. In the embodiment, a quartic higher-order equation is calculated as an approximation by the method of multiple linear regression, as in the examples described later.

From now on, both the voting ticket purchase support method and the voting ticket purchase support program in the voting ticket purchase support system 1 of the embodiment will be described using the flowchart of FIG. The voting ticket purchase support method is executed by the CPU 11 of the computer 10 based on the voting ticket purchase support program. The betting ticket purchase support program provides the computer 10 of FIG. Execute various functions of the purchase function. Each of these functions is executed in the order shown. Since each function overlaps the description of the voting ticket purchase support system and the flow charts of FIGS. 4 to 9, details thereof are omitted.

From the flow chart of FIG. 10, the processing of the CPU 11 of the computer 10 includes a competition basic data acquisition step (S110), a standard ranking prediction data generation step (S120), a pre-competition ranking prediction data generation step (S130), and a pre-competition prediction odds generation step. (S140), a probability prediction step (S150), and a voting ticket purchase step (S160). Of course, various steps required for the operation of the CPU 11 processing itself are naturally included.

The basic competition data acquisition function is used to obtain information about competitions that have been performed by the competitions that are the target of voting, including statistical information of the competitions, odds information for the competitions, and competition result information of the competitions. It is acquired as basic data (S110; competition basic data acquisition step). The reference ranking prediction data generation function generates reference prediction ranking data by estimating mutual rankings of a plurality of sports participating in the competition based on the competition basic data (S120; reference ranking prediction data generation step).

The pre-competition ranking prediction data generation function estimates the pre-competition ranking of the competition based on the reference prediction ranking data and the odds information for each competition in the competition that the competition will implement from now on, and generates it as pre-competition prediction ranking data. (S130; pre-competition ranking prediction data generation step). The pre-game predicted odds generation function estimates pre-game odds for a game to be held by the game based on the game basic data, and generates pre-game predicted odds data (S140; pre-game predicted odds generating step).

The probability prediction function predicts the probability of winning a betting ticket when purchasing a betting ticket set for the competition in the event that the competition will be held based on the pre-competition prediction ranking data and pre-competition prediction odds data. , generate the probability prediction result data in the competition that the sporting object will carry out from now on (S150; probability prediction step). The betting ticket purchase function generates betting ticket data to be purchased for purchasing a betting ticket determined to be purchased based on the probability prediction result data from the organizer of the competition held by the sporting event to be voted for (S160; voting ticket purchase step).

The computer program of the present invention described above may be recorded on a processor-readable recording medium. A logic circuit or the like can be used.

The computer program can be implemented using, for example, script languages such as ActionScript and JavaScript (registered trademark), object-oriented programming languages such as Objective-C and Java (registered trademark), markup languages such as HTML5, and the like.

The inventor obtained information on horse races and runners at the Japan Racing Association at the present time from 2017 to 2021, and conducted a simulation of the betting ticket purchase support system of the present invention. The results are shown in Tables 1 to 15 and FIGS. 11, 12 and 13.

In Tables 1 to 5, the axis is 1 horse, the minimum opponent is 4 horses, and the maximum opponent is 6 horses. And I decided to buy evenly.

In Tables 6 to 10, the axis is 1 horse, the minimum opponent is 4 horses, and the maximum opponent is 6 horses.
Then, all races were targeted, and based on the deviation value of the stallion calculated from the prediction result value and the specified tilt ratio, the tilt distribution purchase was made.
The purchase rate of the slope allocation was obtained from the following formula. The purchase price was set to 100 yen per point×purchase rate. Note that the inclination ratio was set to 3.0.
(1) When the overall deviation value is greater than or equal to the threshold Purchase rate = 1.0 + {(overall deviation value - threshold) / slope ratio}
(2) When the overall deviation value is less than the threshold Purchase rate = 1.0 - {(overall deviation value - threshold) / slope ratio}

In Tables 11 to 15, the axis is 1 horse, the minimum opponent is 4 horses, and the maximum opponent is 6 horses.
Then, the inclination distribution purchase was made based on the deviation value of the stallion calculated from the prediction result value and the specified inclination ratio for the race.
The purchase rate of the slope allocation was obtained from the following formula. The purchase price was set to 100 yen per point×purchase rate. Note that the inclination ratio was set to 3.0.
(1) When the overall deviation value is greater than or equal to the threshold Purchase rate = 1.0 + {(overall deviation value - threshold) / slope ratio}
(2) When the overall deviation value is less than the threshold Purchase rate = 1.0 - {(overall deviation value - threshold) / slope ratio}

From the results in Tables 1 to 15, it was found that the return rate for the investment amount was higher when betting on the predicted rank order than when betting on the popularity of the horses (competitions).

Next, we verified the prediction accuracy. In the graph of FIG. 11, the vertical axis (y-axis) is the probability of winning first place, and the horizontal axis (x-axis) is the hit prediction result value+collection type prediction result value.
Then, the approximate expression "y=ax^4+bx^3+cx^2+dx+e" is set.
Accuracy was determined by the least squares method of the coefficient of determination "R ² ".
The data range is information on horse races and runners of the current Japan Racing Association from 2013 to 2021.
In the graph of FIG. 12, the vertical axis represents the probability of finishing second, and in the graph of FIG. 13, the vertical axis represents the probability of finishing third.

From each graph and its coefficient of determination ^R2 , a very high correlation was confirmed as a result of simulation.

1 voting ticket purchase support system 2 competition organizer 3 computer (server)
4 Internet line 5 Venue 6 Provider of voting ticket purchase support system 10 Computer 11 CPU
12 ROMs
13 RAM
14 storage unit 15 I/O
110 competition basic data acquisition unit 120 reference ranking prediction data generation unit 130 pre-competition ranking prediction data generation unit 140 pre-competition prediction odds generation unit 150 probability prediction unit 160 betting ticket purchase unit

Claims

Competition basics acquired as competition basic data, including statistical information of competitions, odds information for competitions, and competition result information of competitions of competitions in the competitions conducted by the competitions targeted for voting in the past a data acquisition unit;
a reference prediction ranking data generation unit that estimates mutual ranking among a plurality of sports participating in a competition based on the game basic data and generates reference prediction ranking data;
Generating pre-competition predicted ranking data for generating pre-competition predicted ranking data by estimating a pre-competition ranking of a sport based on the reference predicted ranking data and odds information for each sport in a competition to be conducted by the sport. Department and
a pre-competition predicted odds generation unit that estimates pre-competition odds of a competition in a competition to be held by the competition based on the competition basic data and generates pre-competition prediction odds data;
Based on the pre-competition predicted ranking data and the pre-competition predicted odds data, when purchasing a betting ticket set for a competition in a competition that the competition will be held in the future, predict the probability of winning the said betting ticket, and a probability prediction unit that generates probability prediction result data for a competition that will be held from now on.
A voting ticket purchasing unit that generates voting ticket data to be purchased for purchasing a voting ticket, the purchase of which is determined based on the probability prediction result data, from a sponsor of a competition held by a sporting event to be voted for. 1. The voting ticket purchase support system according to 1.
　The betting ticket purchase support system according to claim 1, wherein the game basic data acquisition unit acquires the game basic data from an organizer of a competition held by a game to be voted for.
2. The game basic data acquisition unit includes a statistical data generation unit that generates statistical data for each game from the statistical information of the game, the odds information for the game, and the game result information of the game. The voting ticket purchase support system described.
　The betting ticket purchase support system according to claim 1, wherein the game basic data acquisition unit includes a feature amount generation unit that generates a game feature amount for each game in a game that the game will implement from now on.
The reference ranking prediction data generation unit selects the most popular competition in consideration of odds information for the competition when estimating the ranking among the plurality of competitions participating in the competition based on the competition basic data. 2. The voting ticket purchase support system according to claim 1, wherein the predicted ranking data is generated by prioritizing or excluding the most popular competitors to generate the predicted ranking data.
2. The betting ticket purchase support system according to claim 1, wherein said probability prediction unit generates said probability prediction result data including a winning feature amount related to a winning ball in a competition held in the past. .
Generation of the reference prediction ranking data in the reference prediction ranking data generation section, generation of the pre-competition prediction ranking data in the pre-competition prediction ranking data generation section, and generation of the pre-competition prediction odds data in the pre-competition prediction odds generation section 2. The voting ticket purchase support system according to claim 1, wherein machine learning is used in generating said probability prediction result data in said probability prediction unit.
the computer
Competition basics acquired as competition basic data, including statistical information of competitions, odds information for competitions, and competition result information of competitions of competitions in the competitions conducted by the competitions targeted for voting in the past a data acquisition step;
a reference predicted ranking data generating step of estimating mutual rankings among a plurality of sports participating in a competition based on the game basic data and generating reference predicted ranking data;
Generating pre-competition prediction ranking data for estimating pre-competition ranking of sports in a competition based on the reference prediction ranking data and odds information for each competition in a competition to be performed by the competition, and generating pre-competition prediction ranking data. a step;
A pre-competition predicted odds generation step of estimating pre-competition odds of a competition in a competition to be held by the competition based on the competition basic data and generating pre-competition prediction odds data;
Based on the pre-competition predicted ranking data and the pre-competition predicted odds data, when purchasing a betting ticket set for a competition in a competition that the competition will be held in the future, predict the probability of winning the said betting ticket, and a probability prediction step of generating probability prediction result data for a competition to be conducted by a betting ticket purchase support method.
the computer
Competition basics acquired as competition basic data, including statistical information of competitions, odds information for competitions, and competition result information of competitions of competitions in the competitions conducted by the competitions targeted for voting in the past data acquisition function;
A reference ranking prediction data generation function for estimating the mutual ranking of a plurality of sports participating in the competition based on the competition basic data and generating reference prediction ranking data;
Generating pre-competition prediction ranking data for estimating pre-competition ranking of sports in a competition based on the reference prediction ranking data and odds information for each competition in a competition to be performed by the competition, and generating pre-competition prediction ranking data. function and
A pre-competition predicted odds generation function for estimating pre-competition odds of a competition in a competition to be conducted by the competition based on the competition basic data and generating pre-competition prediction odds data;
Based on the pre-competition predicted ranking data and the pre-competition predicted odds data, when purchasing a betting ticket set for a competition in a competition that the competition will be held in the future, predict the probability of winning the said betting ticket, A betting ticket purchase support program characterized by realizing a probability prediction function for generating probability prediction result data in a competition to be held in the future.