Patent Application
20240087072
Live sporting event broadcasts are huge with millions of people often tuning in to watch their favorite teams plays. However, following everything going on in a game can be difficult for those who are not as familiar with the sport in question, and even for experts and sports fanatics all the information they may want is not readily available in an easily digestible fashion while they watch the game.
Viewers generally have to rely on what information is shared or displayed by commentators on the game, but even the commentators may not have access to up-to-date live probabilities and statistics as the game goes on. Many viewers may wish to have a greater depth of live, constantly updating information available to them in an easily digestible fashion.
A method, system, and apparatus for generating probabilities, for example for displaying on a video feed, which may be generated or adjusted using machine learning and/or artificial intelligence. One embodiment includes a method for generating and adjusting probabilities, including receiving statistical information of a live event in real time, storing the results of an action in the live event in a historic action database, filtering data in the historic action database related to situational data that matches upcoming action in the live event, performing correlations on similar historical data related to the situational data that matches upcoming action in the live event, determining a difference between correlated data of the similar historical data and the data that matches the upcoming action in the live event, comparing the difference to a recommendations database, and adjusting probabilities based on the recommendations database.
In further embodiments the method, system, and apparatus may use the generated probabilities in order to create and display probability graphics on a display device. The graphics may be generated and/or updated using artificial intelligence or machine learning. Display information may be updated in real-time as changes are made in the game (e.g. a player injury, substitutions, changing weather conditions, etc.). In some embodiments the changes may be detected by a detection system on the field, attached to players or equipment, etc. In an exemplary embodiment the detection system may be one or more microchips that may be embedded in the pads of each player to automatically detect player movements or physiological data. In other embodiments the detection system may be a high powered camera systems may be use that are able to detect, for example, player facial expressions, spin on the ball, speed of players or the ball, etc. In yet other embodiments the detection system may be one or more sensors arranged on or around the field. More than one detection system may be used to supplement detection accuracy or speed, or the breadth of data collected.
Advantages of embodiments of the present invention will be apparent from the following detailed description of the exemplary embodiments. The following detailed description should be considered in conjunction with the accompanying figures in which:
Aspects of the invention are disclosed in the following description and related drawings directed to specific embodiments of the invention. Alternate embodiments may be devised without departing from the spirit or the scope of the invention. Additionally, well-known elements of exemplary embodiments of the invention will not be described in detail or will be omitted so as not to obscure the relevant details of the invention. Further, to facilitate an understanding of the description discussion of several terms used herein follows.
As used herein, the word “exemplary” means “serving as an example, instance or illustration.” The embodiments described herein are not limiting, but rather are exemplary only. It should be understood that the described embodiments are not necessarily to be construed as preferred or advantageous over other embodiments. Moreover, the terms “embodiments of the invention”, “embodiments” or “invention” do not require that all embodiments of the invention include the discussed feature, advantage or mode of operation.
With respect to the embodiments, a summary of terminology used herein is provided.
An action refers to a specific play or specific movement in a sporting event. For example, an action may determine which players were involved during a sporting event. In some embodiments, an action may be a throw, shot, pass, swing, kick, hit, performed by a participant in a sporting event. In some embodiments, an action may be a strategic decision made by a participant in the sporting event such as a player, coach, management, etc. In some embodiments, an action may be a penalty, foul, or type of infraction occurring in a sporting event. In some embodiments, an action may include the participants of the sporting event. In some embodiments, an action may include beginning events of sporting event, for example opening tips, coin flips, opening pitch, national anthem singers, etc. In some embodiments, a sporting event may be football, hockey, basketball, baseball, golf, tennis, soccer, cricket, rugby, MMA, boxing, swimming, skiing, snowboarding, horse racing, car racing, boat racing, cycling, wrestling, Olympic sport, eSports, etc. Actions can be integrated into the embodiments in a variety of manners.
Digital gaming refers to any type of electronic environment that can be controlled or manipulated by a human user for entertainment purposes. A system that enables a human and a computer to interact according to set of both implicit and explicit rules, in an electronically powered domain for the purpose of recreation or instruction. “eSports” refers to a form of sports competition using video games, or a multiplayer video game played competitively for spectators, typically by professional gamers. Digital gaming and “eSports” can be integrated into the embodiments in a variety of manners.
The term event refers to a form of play, sport, contest, or game, especially one played according to rules and decided by skill, strength, or luck. In some embodiments, an event may be football, hockey, basketball, baseball, golf, tennis, soccer, cricket, rugby, MMA, boxing, swimming, skiing, snowboarding, horse racing, car racing, boat racing, cycling, wrestling, Olympic sport, etc. Event can be integrated into the embodiments in a variety of manners.
Managed service user interface service is a service that can help customers (1) manage third parties, (2) develop the web, (3) do data analytics, (4) connect thru application program interfaces and (4) track and report on player behaviors. A managed service user interface can be integrated into the embodiments in a variety of manners.
Managed service risk management services are a service that assists customers with (1) very important person management, (2) business intelligence, and (3) reporting. These managed service risk management services can be integrated into the embodiments in a variety of manners.
Managed service compliance service is a service that helps customers manage (1) integrity monitoring, (2) play safety, (3) responsible gambling and (4) customer service assistance. These managed service compliance services can be integrated into the embodiments in a variety of manners.
Managed service and technology platform are services that helps customers with (1) web hosting, (2) IT support and (3) player account platform support. These managed service and technology platform services can be integrated into the embodiments in a variety of manners.
Managed service and marketing support services are services that help customers (1) acquire and retain clients and users, (2) provide for bonusing options and (3) develop press release content generation. These managed service and marketing support services can be integrated into the embodiments in a variety of manners.
“Fantasy sports connector” are software connectors between method steps or system elements in the embodiments that can integrate fantasy sports. Fantasy sports allow a competition in which participants select imaginary teams from among the players in a league and score points according to the actual performance of their players. For example, if a player in a fantasy sport is playing at a given real time sports, probabilities could be changed in the real time sports for that player.
Software as a service (or SaaS) is a method of software delivery and licensing in which software is accessed online via a subscription, rather than bought and installed on individual computers. Software as a service can be integrated into the embodiments in a variety of manners.
Automatic content recognition (ACR) is an identification technology to recognize content played on a media device or present in a media file. Devices containing ACR support enable users to quickly obtain additional information about the content they see without any user-based input or search efforts. To start the recognition, a short media clip (audio, video, or both) is selected. This clip could be selected from within a media file or recorded by a device. Through algorithms such as fingerprinting, information from the actual perceptual content is taken and compared to a database of reference fingerprints, each reference fingerprint corresponding to a known recorded work. A database may contain metadata about the work and associated information, including complementary media. If the fingerprint of the media clip is matched, the identification software returns the corresponding metadata to the client application. For example, during an in-play sports game a “fumble” could be recognized and at the time stamp of the event, metadata such as “fumble” could be displayed. Automatic content recognition (ACR) can be integrated into the embodiments in a variety of manners.
Augmented reality means a technology that superimposes a computer-generated image on a user's view of the real world, thus providing a composite view. In an example, a real time view of the game can be seen and a “guess” which is a computer-generated data point is placed above the player that is guessed on. Augmented reality can be integrated into the embodiments in a variety of manners.
Some embodiments of this disclosure, illustrating all its features, will now be discussed in detail. It can be understood that the embodiments are intended to be open ended in that an item or items used in the embodiments is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items.
As used herein Robotic Process Automation (RPA) may be the use of software with some aspects of AI or machine learning capabilities to handle high-volume, repeatable tasks that previously required humans to perform. An example of this may be software for a “web crawler” that can eliminate hundreds of hours of searching each time the same searches need to be conducted on specific web searches. Robotic process automation may be integrated with one or more of the embodiments.
As used herein a database may include, but is not limited to, publicly available data sets such as dictionaries, databases behind a paywall, internal databases of, for example, non-disclosure agreements, invention disclosures, competitive analysis, statistical information of, for example, a sports game, etc. It may be understood that some databases may include an Application Specific Interface (API) which may enable a program to extract data from a search.
As used herein, a scheme may refer to any game plan, play, or strategy used before, during, or after a sporting game or event.
It can be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise. Although any systems and methods similar or equivalent to those described herein can be used in the practice or testing of embodiments, only some exemplary systems and methods are now described.
Further, a live action input module 104 may receive data about each individual action in a game or match and store the data in a live event database 106. In some embodiments, an action may be a specific play or specific event in a sporting event. In some embodiments, an action may be a throw, shot, pass, swing, kick, hit, performed by a participant in a sporting event. In some embodiments, an action may be a strategic decision or scheme made by a participant in the sporting event such as a player, coach, management, etc. In some embodiments, an action may be a penalty, foul, or type of infraction occurring in a sporting event. In some embodiments, an action may include the participants of the sporting event. In some embodiments, an action may include beginning events of sporting event, for example opening tips, coin flips, opening pitch, national anthem length, and the like. In some embodiments, a sporting event may be football, hockey, basketball, baseball, golf, tennis, soccer, cricket, rugby, MMA, boxing, swimming, skiing, snowboarding, horse racing, car racing, boat racing, cycling, wrestling, Olympic sport, and the like. The live event database 106 may be provided, which may store data collected by the live event module 104 such as the results of the action that has just occurred as well as the situational data for the next upcoming action.
In some embodiments the system for artificial intelligence based live game probability calculation may include a detection system such as a plurality of sensors 108. The sensors may include, for example, motion sensors, temperature sensors, humidity sensors, cameras such as an RGB-D Camera which is a digital camera providing color (RGB) and depth information for every pixel in an image, microphones, radiofrequency receiver, a thermal imager, a radar device, a lidar device, an ultrasound device, a speaker, wearable devices etc. Also, the plurality of sensors 108 may include tracking devices, such as RFID tags, GPS chips or other such devices embedded on uniforms, in equipment, in the field of play, in the boundaries of the field of play, or other markers on the field of play. Imaging devices, including cameras, may also be used as tracking devices such as player tracking that provides statistical information through real-time X, Y positioning of players and X, Y, Z positioning of the ball. Further, the imaging devices or cameras may be programmed to automatically track certain players in certain circumstances. For example, a “star” or popular player may be automatically tracked by one or more cameras. In other situations, a player with a ball (such as in football or basketball) may be tracked, or a player who is anticipated to be in a play or action (such as a batter in baseball), may be automatically tracked. In still other embodiments, and incorporating other historical and/or sensor data below, large language models or modelling (LLM), machine learning (ML) or artificial intelligence (AI) can be used to track one or more players who are likely to be involved in a play or an action. For example, in a football game, a wide receiver who is determined by LLM/ML/AI to be open or uncovered, who is determined to be likely to be open or uncovered, or who is likely to be targeted with a pass or handoff may be automatically tracked. In another embodiment, for example in a basketball game where the offensive team is losing late in a game, one or more players who are the best three-point shooters on a team may be automatically tracked. In other embodiments related to basketball, even if a player does not have a high three-point shooting percentage, LLM/ML/AI could determine that the player is “clutch” or has a higher rate of game winning, game tying, important, etc. shots made, and that player could be automatically tracked.
In some embodiments, the sensor data is collected from the live event 102 and sent to a server or a cloud 110 where it is stored in a historical action database 130. In some embodiments, the sensor data may be collected from a third party source and stored on the server or cloud 110. Further, the availability of sensor data may be displayed to a user and/or any sensor data itself may be displayed to a user. Further, the availability or use of sensor data may be activated or deactivated by a user.
A cloud 110 or communication network may be a wired and/or a wireless network. The communication network, if wireless, may be implemented using communication techniques such as visible light communication (VLC), worldwide interoperability for microwave access (WiMAX), long term evolution (LTE), wireless local area network (WLAN), infrared (IR) communication, public switched telephone network (PSTN), radio waves, and other communication techniques, as desired. The communication network may allow ubiquitous access to shared pools of configurable system resources and higher-level services that can be rapidly provisioned with minimal management effort, often over internet and relies on sharing of resources to achieve coherence and economies of scale, like a public utility, while third-party clouds enable organizations to focus on their core businesses instead of expending resources on computer infrastructure and maintenance. A live event data API 112 delivers data from the live event 102 to a data network 122. A display device API 114 delivers data between the data network 122 and a display device 116.
The display device 116 may be a computing device, laptop, smartphone, tablet, computer, smart speaker, television screen, teleprompter, or I/O device. I/O devices may be present in the computing device. Input devices may include keyboards, mice, trackpads, trackballs, touchpads, touch mice, multi-touch touchpads and touch mice, microphones, multi-array microphones, drawing tablets, cameras, single-lens reflex camera (SLR), digital SLR (DSLR), CMOS sensors, accelerometers, infrared optical sensors, pressure sensors, magnetometer sensors, angular rate sensors, depth sensors, proximity sensors, ambient light sensors, gyroscopic sensors, or other sensors. Output devices may include video displays, graphical displays, speakers, headphones, inkjet printers, laser printers, and 3D printers. Devices may include a combination of multiple input or output devices, including, e.g., Microsoft KINECT®, Nintendo Wiimote® for the WIT, Nintendo® WII U GAMEPAD®, or Apple IPHONE®. Some devices allow gesture recognition inputs through combining some of the inputs and outputs. Some devices may provide for facial recognition which may be utilized as an input for different purposes including authentication and other commands. Some devices may provide for voice recognition and inputs, including, e.g., Microsoft KINECT®, SIRI® for IPHONE by Apple, Google Now or Google Voice Search, and the like.
Additional devices may have both input and output capabilities, including, e.g., haptic feedback devices, touchscreen displays, or multi-touch displays. Touchscreen, multi-touch displays, touchpads, touch mice, or other touch sensing devices may use different technologies to sense touch, including, e.g., capacitive, surface capacitive, projected capacitive touch (PCT), in-cell capacitive, resistive, infrared, waveguide, dispersive signal touch (DST), in-cell optical, surface acoustic wave (SAW), bending wave touch (BWT), or force-based sensing technologies. Some multi-touch devices may allow two or more contact points with the surface, allowing advanced functionality including, e.g., pinch, spread, rotate, scroll, or other gestures. Some touchscreen devices, including, e.g., Microsoft PIXELSENSE® or Multi-Touch Collaboration Wall, may have larger surfaces, such as on a table-top or on a wall, and may also interact with other electronic devices. Some I/O devices, display devices or group of devices may be augmented reality devices. The I/O devices may be controlled by an I/O controller. The I/O controller may control one or more I/O devices, such as, e.g., a keyboard and a pointing device, e.g., a mouse or optical pen. Furthermore, an I/O device may also provide storage and/or an installation medium for the computing device. In still other embodiments, the computing device may provide USB connections (not shown) to receive handheld USB storage devices. In further embodiments, an I/O device may be a bridge between the system bus and an external communication bus, e.g. a USB bus, a SCSI bus, a FireWire bus, an Ethernet bus, a Gigabit Ethernet bus, a Fibre Channel bus, or a Thunderbolt bus. The display device 116 can leverage the sensors in for purposes such as automatic content recognition, augmented reality or the synchronization of screens between the user device interface and other displays. The display device 116 may display the probabilities for the next action of the live game 102. It may also be appreciated that the display device 116 could be part of a live video display, live television display or any other live video feed of the live game 102 on which probabilities are being provided. In some embodiments there may be a data GUI 118, or guided user interface or graphical user interface, which may display the data options and probabilities for various actions in the live event 102. The interface(s) may either accept inputs from users or provide outputs to the users or may perform both the actions. In one case, a user can interact with the interface(s) using one or more user-interactive objects and devices. The user-interactive objects and devices may include user input buttons, switches, knobs, levers, keys, trackballs, touchpads, cameras, microphones, motion sensors, heat sensors, inertial sensors, touch sensors, virtual reality, augmented reality, eye tracking, or a combination of the above. Further, the interface(s) may either be implemented as a command line interface (CLI), a graphical user interface, a voice interface, or a web-based user-interface. The data network 122 may provide an artificial intelligence based software module that compares data from the live event to data in the historic action database 130 in order to calculate probability of the next action in the live game 102. The data network 122 may be located on the server or cloud 110 which may perform real time analysis on the type of play and the result of a play or action. The server, or cloud 110, may also be synchronized with game situational data, such as the time of the game, the score, location on the field, weather conditions, and the like, which may affect the choice of play utilized. For example, in other exemplary embodiments, the server may not receive data gathered from sensors and may, instead, receive data from an alternative data feed, such as Sports Radar. Data may be received from a remote terminal or mobile device that may be located at the live event 102. The data related to the live event 102 may be entered into the remote terminal or mobile device by one or more administrators of the live game probability calculation system, employee, technician, or other third-party persons. For example, a person may be located at the live event 102 and may provide the position of the players on the field, such as the offensive formation in an American football game, through an input into a remote terminal or mobile device. This human collected data may be used independently, or in conjunction with sensor data, in the generation of probabilities. This data may be provided following the completion of any play and the data from this feed may be compared with a variety of team data and league data based on a variety of elements, including down, possession, score, time, team, and so forth, as described in various exemplary embodiments herein. The system for artificial intelligence based live game probability calculation may further include a base module 124 which receives the live event database 106 from the live event module 104, which contains historical and situational data on the live event 102 currently occurring. The base module 124 stores the historical data in the historic action database 130 and sends the situational data to a probabilities module 126 and initiates the probabilities module 126. The probabilities module 126 uses the situational data from the live event 102 to filter the historic action database 130 on previous actions with some the same situational data and performs correlations on the similar actions in order to determine the difference in the correlations and compare the difference in correlations to the recommendation database 132 in order to adjust the probabilities within the probabilities database 134 accordingly. A data module 128 compares the probabilities database 134 to the adjustment database 136 in order to determine if there is a match in the probabilities IDs. Then, if there is a match, then the probabilities are adjusted accordingly, by the data module 128. A historic action database 130 stores all the historic actions of an event. A recommendation database 132 is used to determine the appropriate adjustment in the probabilities by using the difference in the correlated data from the data module 126. The probabilities database 134 contains the current probabilities. An adjustment database stores the probabilities ID and the appropriate adjustment, for example increase by 5% or decrease by 5%. Any adjusted probabilities can then be displayed via the display device 116, or interface, on a live video feed, such as on a television feed, or any other video live stream. Further, adjusted probabilities may be shown in real time, for example showing a 66% chance of a running play and a 34% chance of a passing play in a football game when the quarterback is positioned under the center and changing to a 75% chance of a passing play and 25% chance of a running play when the quarterback then moves back into a shotgun formation.
In other embodiments, it may be appreciated that the previous formula may be varied depending on a variety of reasons, for example adjusting probability based on further factors, adjusting probability based on changing conditions or additional variables. Additionally, in other exemplary embodiments, one or more alternative equations may be utilized in the probability module 126. One such equation could be Zobserved=(z1—z2)/(square root of [(1/N1−3)+(1/N2−3)], where z1 is the correlation coefficient of the first dataset, z2 is the correlation coefficient of the second dataset, N1 is the sample size of the first dataset, and N2 is the sample size of the second dataset, and the resulting Zobserved to compare the two correlation coefficient based on statistical significance as opposed to the difference of the two correlation coefficients. Another equation used may be Z=b1−b2/Sb1−b2 to compare the slopes of the datasets or may introduce any of a variety of additional variables, such as b1 is the slope of the first dataset, b2 is the slope for the second dataset, Sb1 is the standard error for the slope of the first dataset and Sb2 is the standard error for the slope of the second dataset. The results of calculations made by such equations may then be compared to the recommendation database 132 and the probability module 126 may then extract a probability adjustment from the recommendation database 132. The extracted probability adjustment is then stored in the adjustment database 136 and the data module 128 is initiated by the probability module 126, as in the above.
In other embodiments, the probability module 126 may adjust probabilities based on one or more current states of the live event 102. In one state, the probability module 126 may use data from the historic action database 130 to create probabilities for one or more potential outcomes of the current state of the live event 102. For example, the data network 122 may determine there is a 75% chance that Aaron Judge gets a hit in his current at-bat, based on his batting average in previous at-bats against the current pitcher. In a second state, the probability module 126 may utilize data from the live event data API 112 to adjust the probability. For example, spin rate of pitches may be tracked in real-time. The average spin rate on four-seam fastballs thrown by Clayton Kershaw in the current game may be 2732. That may be 10% higher than his average spin rate has been on that pitch this season. This may be used to adjust the probability of Aaron Judge getting a hit in the current at-bat from the 75% that would be used based on the two players' historical interactions to 80%.
It can be noted that the probability module 126 can be made available for access, reconfiguration, modification, or control for customers or used for managed service user interface service, managed service risk management services, managed service compliance service, managed service pricing and trading service, managed service and technology platform, managed service and marketing support services, payment processing services, business applications, engaging promotions, business applications, state based integration, game configurator, “fantasy sports connector”, software as a service, synchronization of screens, automatic content recognition (ACR), joining social media, Augmented reality, digital gaming, or “eSports”.
Functioning of the data module 128 will now be explained with reference to
Functioning of the historic sensor database 120 will now be explained with reference to
Functioning of the adjustment database 136 will now be explained with reference to
Functioning of another exemplary embodiment of the probabilities database 134 will now be explained with reference to
The probabilities database 134 may contains a list of all current probabilities available to the users of the server. The database may contain probability data such as the probability ID, a description of action, and the probability. The database may contain event data related to the probability such as the team, the quarter or time period for the upcoming play, the down, and the distance to gain.
In one or more exemplary embodiments large language models (LLM), AI, and/or ML may be used to generate probabilities based on a plurality of factors.
Referring to exemplary
In an exemplary embodiment the clock base module 1602 may control the one or more other clock modules to identify factors in a sports game related to the clock that may impact one or more probabilities of the one or more plays. The probabilities may be adjusted at any point between when the prediction market opens and closes or may be adjusted as pre-market or post-market probabilities.
It may be understood that probabilities offered may be changed at any point, and may change dynamically, in response to changes in circumstances before, during, or after the one or more plays. In some embodiments specific sub-modules may be triggered in response to specific events, for example a coaching adjustment module 1604 may be triggered at the completion of a play, as that may be the time when the context of the game including time on the clock, score, field position, etc are all known factors for determining probability for a next play. In another embodiment the occurrence of substitutions may trigger a personnel adjustment module 1606. In yet another embodiment a position adjustment module 1608 may be triggered when one or more players line up at the beginning of the next play. One or more of these, or other, circumstances may also trigger a wider market adjustment module 1610, which may determine the impact a next play specific probability may have on markets outside the next play, for example on future plays, overall game markets, and/or plays in other games. Finally, the probabilities determined may be stored in a clock probabilities database 1612. It may be understood that the modules may be called in any order and/or that multiple modules may be run simultaneously.
Now explaining the modules in more detail, in an exemplary embodiment the coaching adjustment module 1604 may use AI to predict how a coach or coaching staff tends to drive probabilities for different outcomes in one or more prediction markets. For example, an initial or general probability may be determined, such as an average NFL team passes on 60% of offensive plays. Additional play context may then be used to further tailor the probability, such as probability adjustments associated with the positions on the field, with similar score and time remaining, etc. As an example, it may be determined that 52% of 1st and 10 plays are passes, but 68% of 2nd and 10 plays are passes.
The coaching adjustment module 1604 may then determine additional adjustments based on one or more coaches tendencies, for example the coaching adjustment module 1604 may determine that the Cleveland browns only passed 40% of their 1st and 10 plays between 2018 and 2021, while the Kansas City Chiefs passed on 61% of 1st and 10 plays during the same period. The coaching adjustment module 1604 may then take these trends into account in order to adjust the probability of the current play.
In some embodiments the time remaining on the game clock, or the time remaining relative to a specific portion of a game (e.g. time until end of quarter) may impact the tendency of the coach or team and therefore affect the related probabilities. For example, a team may implement a “four-minute offense” strategy if, for example, they are leading a game and there are give minutes left in the fourth quarter. The “four-minute offense” strategy may allow the team to burn remaining time off the clock by avoiding incomplete passes and going out of bounds while still achieving first downs. Therefore, it may be determined that teams leading with less than five minutes in a fourth quarter may pass significantly less than in other situations, for example less than 20% of the time. In other situations, other specific strategies or schemes may tend to skew probabilities, other strategies may include, for example, “two-minute offense”.
In some embodiments specific coaching staffs may have different tendencies in approaching specific aspects of the game, for example a first coach may be twice as likely to pass while leading with less than five minutes left in the fourth quarter than the average coach because his offensive scheme relies on a lot of bubble screens in that situation.
In an exemplary embodiment machine learning may be applied to historical play data and may identify coaching patterns and tendencies to apply to probabilities in given situations. For example, while coach A may be 10% more likely than average to run on a play when trailing close and late, that coach may be 25% more likely to run when they have two or more timeouts. It may be understood that in many situations each coaching staff may have multiple members, for example the head coach, offensive coordinator, defensive coordinator, etc. Some head coaches may insist on making most of the decisions and play calls themselves, while other head coaches may delegate play-calling to their coordinators and coaches. In an embodiment LLM, AI, and/or ML may identify which coaches are the decision-makers whose tendencies should be tracked by comparing trends across different staff compositions. The probabilities for each play event may then be calculated and stored, for example in the clock probabilities database 1612. Further, as noted above, AI and ML, as well as large language modeling (LLM) may be utilized with cameras to automatically track certain players, such as wide receivers who are likely to be open or targeted or “clutch” basketball players. In still further embodiments using automatic camera tracking and LLM, AI and/or ML, it could be determined that a player is open to receive a pass, and thus a camera could automatically track that player, or it could be determined that a player on a basketball court has made an aggressive move or been the beneficiary of a pick play to get open, which could lead to a camera automatically tracking that player.
Now explaining the personnel adjustment module 1606 in more detail. The personnel adjustment module 1606 may extract initial probabilities from, for example, the clock probabilities database 1612. The personnel adjustment module 1606 may use LLM, AI, and/or ML to predict how the personnel on the field, and the attributes of those personnel, drive the probabilities of a play. For example, different defensive backs may have different abilities to match up in coverage with different types of receivers. The better the defensive backs on the field, the lower the chances of a successful passing play. This may, for example, increase the likelihood that a coach or quarterback changes the play to a run. Likewise, the formation or a number of a specific position adjust the probabilities of certain actions, for example a five defensive back formation vs. a six defensive back formation. The strengths and weaknesses of each personnel decision may further vary based on the context of the game, such as field position, score, and the number of timeouts remaining. For example, a defense with six of the best defensive backs in the league may deploy in dime defense. Normally this may be a good defense to run against, but a team trailing late in the game with no timeouts remaining may still have to pass to avoid being tackled in bounds and having the clock run out.
The combination of coaching tendencies, personnel, and time on the clock may also be identified. The identity of the players in a specific position, for example the identity of the defensive backs in a particular formation, may have an additional impact on the probabilities of various outcomes. For example, a cornerback shorter than 5′10″ will have less success covering receivers over 6′5″, leading to an identifiable matchup advantage. The context of the game, specifically coaching tendencies, position on the field, and time remaining, may significantly impact the likelihood of this type of matchup being exploited by the offense. A coach may be more likely to motion pre-snap to get the matchup between the tall receiver and the short cornerback. Similarly, certain quarterbacks may be more likely than others to check out of a run play to exploit the height advantage of the receiver. The updated probabilities may then be sent to a database, for example the clock probabilities database 1612.
Now explaining the position adjustment module 1608 in more detail. The position adjustment module 1608 may extract initial probabilities from, for example, the clock probabilities database 1612.
The position adjustment module 1608 may utilize the positioning of personnel on the field, and attributes of that personnel, in order to adjust the probabilities of one or more plays in a prediction market. For example, in an exemplary football game the number of players in the box may significantly impact the probabilities of certain outcomes. The position adjustment module 1608 may use one or more sensors to detect the number of players in the box, and who those players are. For example, sensors may identify there are eight defenders in the box and identify that the eighth defender is the strong safety creeping toward the line of scrimmage. The sensors may also determine there is a strong safety who may be a player who is an excellent pass rusher. Their ability as a pass rusher may be determined by their efficiency percentage in creating pressure on the quarterback when they blitz. So, the safety's presence in the box and pressure efficiency may make them a threat to a passing play, even against eight men in the box. Machine learning may determine the checks a given quarterback is likely to make at the line of scrimmage against specific defenses. A quarterback may be more likely to check to a slant pass to the slot wide receiver into the area that is about to be vacated by the blitzing strong safety, and the system may increase the probabilities of a pass when detecting the safety position. It may be understood that LLM, AI, or ML may be used to make the probability or odds adjustments. The adjusted probabilities may be stored in, for example, the clock probabilities database 1612.
In some embodiments real-time assessment of player position may be obtained by, for example, applying object recognition to video or sensors attached to the players. The real-time information may allow for automatic determination of one or more player's positions and may allow for the calculation of the above probability adjustments. For example, in an exemplary football game sensors in the shoulder pads of the safety and receivers could tell the system how close the safety is to the line of scrimmage and that the receiver is lined up behind the line of scrimmage, making a blitz and backward pass more likely.
Now explaining the wider market adjustment module 1610 in more detail. The market adjustment module 1610 may extract initial probabilities from, for example, the clock probabilities database 1612. The initial probabilities extracted may be from a different live game or prediction market than the currently plays being utilized by the other modules.
The market adjustment module 1610 may determine how the probabilities adjustments made by other modules, for example the coaching adjustment module 1604, the personnel adjustment module 1606, and/or the position adjustment module, either individually or together, impact the probabilities of other guessing markets including those for plays other than the next play, larger game wagers or predictions, and other games. For example, moneyline and spread guesses on the game's outcome, over/under guesses on the game's total score, and certain prop bets, such as will Player A will score a touchdown in the current game.
In an exemplary embodiment, when an adjustment is made to one or more outcomes of a next play, the adjusted expected outcome may be compared to the markets outside of the next play to determine if the adjusted play outcome impacts the identified wider market. For example, an increase in the probabilities of a run may be made based on the safety's and receiver's positioning, the player profiles and tendencies in the context of the game clock, and the number of available timeouts. The identified wider market may be, “Will Player A score a touchdown in the current game?” If Player A is lined up as the outside receiver, this increased likelihood of a run play to him increases the number of opportunities he will have in the current game to score a touchdown. The adjustment based on the alignment of the defense and the game clock may result in a decreased payout offered on the wider market to reflect the greater likelihood that Player A will score a touchdown in the current game. The adjustments made to the wider market may then be stored in a database, for example the clock probabilities database 1612. In some embodiments the updates may further include at least the reasons for changing the probabilities.
Now explaining the clock probabilities database 1612 in may detail. The clock probabilities database 1612 may store all that data that is created in, for example, the execution of the clock base module 1602, the coaching adjustment module 1604, the personnel adjustment module 1606, the position adjustment module 1608, etc. The module may include information on, for example but not limited to, the players, plays, and probabilities for a plurality of plays, games, and/or sports.
In other embodiments a plurality of other factors may be considered in determining probability adjustments. For example, instead of clock rate the modules may instead focus on the position or positioning of one or more quarterbacks on the field. Referring to
The modules in the exemplary system for generating or optimizing probabilities based on one or more quarterbacks 1700 may operate and interact in the same fashion as the corresponding modules in the exemplary system for generating or optimizing probabilities based on the clock 1700. Instead of looking at the clock, the system may instead receive information on the position or positioning of the quarterbacks and identify how these factors may impact the probabilities of one or more plays outcomes. For example, in an exemplary embodiment how the quarterback lines up, such as under center or in the shotgun, may impact the probabilities of the play's outcome. The alignment may have a significant impact on the run-pass play selection of the offense. For example, The run rate, the percentage of plays that were a run, of plays with the quarterback under center may be 69%, compared to just 31% from the shotgun alignment. In an exemplary embodiment these averages may serve as the baseline probabilities for a run/pass market on a given play.
Referring now to
In an exemplary embodiment the no-huddle base module 1804 may optimize or generate probabilities or odds by directing one or more sub-modules to identify factors related to how the offense does or does not utilize a huddle in between plays that may impact the probabilities of the play's outcome. For example, in an embodiment just over 10% of plays may be run without a huddle, and 57% of plays run with a huddle may be passing plays, while 68% of no-huddle plays may be passes. These averages may serve as the baseline probabilities for a run/pass market on a given play.
In an exemplary embodiment the no-huddle base module 1804 may use one or more sub modules to, for example, determine how the coaching staff's tendencies, personnel and play positions drive the probabilities of different outcomes in the guessing markets. The no-huddle base module 1804 may further determine the impact of one or more sub-modules on wider markets, such as plays after the next play and/or plays in other games and may store generated or updated odds in one or more databases.
In an exemplary embodiment the box base module 1806 may optimize or generate probabilities or odds by directing one or more sub-modules to identify factors related to how defense is lined up in the box, which players are in the box, and how that may impact the probabilities of the play's outcome. The box count may be used to define the number of defenders inside the tackle box, and/or within three yards of the line of scrimmage. In an embodiment, a higher box count may correlate to better success rate of passing plays, while a lower box count may correlate to better success with rushing plays. The offenses historical success against the box count may serve as a baseline probabilities for one or more markets on a given play.
In an exemplary embodiment the box base module 1806 may use one or more sub modules to, for example, determine how the coaching staff's tendencies, personnel and play positions drive the probabilities of different outcomes in the guessing markets. The no-huddle base module 1804 may further determine the impact of one or more sub-modules on wider markets, such as plays after the next play and/or plays in other games, and may store generated or updated odds in one or more databases.
In an exemplary embodiment the field position base module 1808 may optimize or generate probabilities or odds by directing one or more sub-modules to identify factors such as where the drive started on the field and whether any explosive drives happened, and how those factors may impact one or more probabilities or odds of a play's outcome. The starting field position on a drive may greatly impact the probabilities of the offense scoring on a given drive. For example, the probabilities of a team scoring may be much lower on a drive that starts on their 10-yard line than on a drive that starts on the opponent's 20-yard line. A play may be considered successful when certain thresholds are reached, for example in a football game a play may be considered successful if at least 40% of the yards to go are covered on first down, 60% of the yards to go on second down, and 100% of the yards to go on third or fourth down. Field position may impact the likelihood of a score on a given drive and the success rate of a given play. For example, the success rate on rushing plays from inside a team's own 20-yard line may be significantly higher than on the opponent's side of the field. The offense's historical success from a given starting field position may serve as the baseline probabilities for a first down market on a given play and the scoring probabilities for a given drive. For each play on the drive, the modules may identify the current field position, starting field position, and if there has been at least one explosive play on the current drive, and may calculate baseline probabilities for a first down market on a given play and the scoring probabilities for a given drive.
In an exemplary embodiment the field position base module 1806 may use one or more sub modules to, for example, determine how the coaching staff's tendencies, personnel and play positions drive the probabilities of different outcomes in the guessing markets. The no-huddle base module 1804 may further determine the impact of one or more sub-modules on wider markets, such as plays after the next play and/or plays in other games and may store generated or updated odds in one or more databases.
In an exemplary embodiment the player rating module 1810 may optimize or generate probabilities or odds by directing one or more sub-modules to identify factors such as, for example, the player ratings of one or more players, the players alignment, and how these may impact the probabilities of a plays outcome. Player skill ratings may be quantified in a variety of fashions, for example for an exemplary football team there may be, for example, a pro football focus's player grade and a football outsiders' defensive-adjusted value over average score. A pass rusher's effectiveness may be quantified based on the percentage of snaps they generate pressure on the opposing quarterback. An offensive lineman's performance may be similarly quantified based on the percentage of time they allow pressure. A defender's ability to cover the pass may be based on the percentage of plays in which they are both playing man-to-man coverage and allow separation above a threshold, for example, one yard of separation between themselves and the receiver they are covering. Zone coverage effectiveness may be quantified based on the break rate or the percentage of times the defender got a good break on a pass. A good break may be, for example, defined as covering at least one-third of the distance to the ball between when the ball is thrown and the ball reaches the receiver. Missed tackles, passes defended, forced fumbles, etc., may be other defensive metrics that could be considered in a player rating. Players in other games, e.g. baseball, soccer, tennis, etc. may have their own rating systems.
In some embodiments, the player ratings may be calculated based on human observations of the plays, identifying the players involved and whether they had a positive or negative impact on the play. Alternatively, and/or in addition, AI may examine sensor data, including video, to grade players on each play. Automated systems could grade effort on a play by, for example, comparing the player's top/average speed on a given play to their top/average recorded speeds.
In some embodiments individual player ratings may be combined to rate position groups, for example a defensive line, linebackers, and/or defensive backs. For example, a linebacker group may have one rating for their ability to stop the run based on a combination of the yards before first contact, missed tackle rate, yards after first contact, explosive play rate, etc. They may be rated for their ability to stop the pass based on a combination of break rate on zone coverage plays, average separation on man coverage plays, average yards after the catch, pressure rate, passer rating when targeted, etc. Position group ratings may be used to adjust the probabilities of next play markets and wider markets within the game.
In an exemplary embodiment the player rating base module 1810 may use one or more sub modules to, for example, determine how the coaching staff's tendencies, personnel and play positions drive the probabilities of different outcomes in the guessing markets. The no-huddle base module 1804 may further determine the impact of one or more sub-modules on wider markets, such as plays after the next play and/or plays in other games, and may store generated or updated odds in one or more databases.
In an exemplary embodiment the missed tackle base module 1812 may optimize or generate probabilities or odds by directing one or more sub-modules to identify factors related to the tackling effectiveness of the teams and players and how that may impact the probabilities of the play's outcome. Tacking effectiveness may be measured in a variety of ways. For example, the missed tackle rate may be (missed tackles/(missed tackles+tackles) of the individual defenders or the defense as a whole. Additional measures may be incorporated to get a more complete understanding of the tackling effectiveness of the defense. For example, one or more of the average yards before first contact on running plays, the average yards after the catch on passing plays, and the average yards after first contact on running plays may be incorporated into a tackling effectiveness rating. The tackling effectiveness rating across all game situations may serve as the baseline probabilities for calculating the probabilities of an explosive play on a given play.
In an exemplary embodiment the context of the game, such as field position, time on the clock, score, etc., may be used to refine the dataset of historical tackling efficiency. Offensive team and player effectiveness may also be incorporated. For example, the missed tackles forced by the offense's potential ball carriers or the average yards after the catch of their potential receivers.
In an exemplary embodiment the missed tackle base module 1812 may use one or more sub modules to, for example, determine how the coaching staff's tendencies, personnel and play positions drive the probabilities of different outcomes in the guessing markets. The no-huddle base module 1804 may further determine the impact of one or more sub-modules on wider markets, such as plays after the next play and/or plays in other games, and may store generated or updated odds in one or more databases.
It may be understood that the above modules may be used alone or in any combination, and in some embodiments additional modules may be used. For example, an exemplary embodiment for a baseball game may use a clock base module, a pitcher module, a player rating module, etc. It may be contemplated that the modules may be implemented in any of a variety of team or individual sports, including but not limited to football, hockey, basketball, baseball, golf, tennis, soccer, cricket, rugby, MMA, boxing, swimming, skiing, snowboarding, horse racing, car racing, boat racing, cycling, wrestling, Olympic sport, etc. It may be further understood that AI may be integrated with any of the modules.
Referring generally to
In some embodiments the graphic display may be on a television program and the same for every viewer. In other embodiments the graphic display may be part of an internet stream and may be watched on, for example but not limited to, a smart TV, a computer, a mobile phone, a tablet, etc. In some embodiments the display may be customized to the particular viewer, based on, for example, previously inputted user preferences, other viewer data such as watch history, and/or integration with, for example, a fantasy sports team. In some embodiments the user may be able to adjust preferences during the game, or may be able to manually select a player, team, position, etc. to follow more closely.
Exemplary
Exemplary
Exemplary
Exemplary
Exemplary
Referring generally to exemplary
Exemplary
Exemplary
Exemplary
Exemplary
Exemplary
Exemplary
Exemplary
In another exemplary embodiment an “in play” game software application may be provided. The “in play” game software may be executed on a user device, for example a computer, phone, tablet, etc., and may be linked to the internet or a cloud that runs software on a prediction network. Further, in some embodiments, the “in play” software may be utilized to generate simulations, simulate plays, simulate players, simulate games, parts of games, time intervals of games, possessions, drives, innings, and the like of games.
In some embodiments, visual simulation can be a method of creating realistic or abstract visual experiences that mimic real-world scenarios or concepts. There are various ways to describe (or enhance multiple embodiments) in visual simulation, depending on its specific context and purpose. Some exemplary or alternative ways to describe visual simulation include:
Further, it can be appreciated that these terms can be used interchangeably or in combination, depending on the specific context and purpose of the visual simulation. The choice of terminology often reflects the emphasis placed on different aspects of the simulation, whether it is immersion, graphics quality, interactivity, or the application domain.
Referring back to exemplary
Then, utilizing other information, such as data from recommendations database 132 and probabilities database 134, various odds related to the occurrence and outcome of a play may be simulated using instructions executed on a processor. It can be appreciated that such simulations may take into account further variables, such as changing the location or conditions associated with a game. In some embodiments, additional variables may be entered into the simulation which can provide or affect outcomes in a further manner—for example, injuries, weather conditions, substitutions, sun positioning, location, fatigue, and/or any other variable or combination of variables could be inserted into the simulation, which may affect an outcome of a simulated play, drive, action, or the like. Also, it may be appreciated that variables may be introduced before any simulated play, drive, or action, during any play, drive, or action, or at any other time which may affect an outcome of the simulation. For example, a home team's drive in football against an away team could be simulated with the home and away roles reversed, which could affect the availability or outcome of certain plays in a football game. In other sports, such as baseball, the simulation may utilize other stadiums which can affect a player's ability to hit a homerun, for example. Further, following the result of any real or simulated play, a graphic showing the results of the play may be provided on the display. Additionally, following simulation of a play, drive, or some type of action, graphical data may be displayed, audio data may be outputted, or some combination of graphically displayed data and outputted audio may be used to describe the play, drive, or action, provide result information of the play, drive, or action (or a related wager), or otherwise output information via appropriate display and speaker elements. In further embodiments, simulated audio may be provided before, during, or after any play, action, and/or drive, which may serve as color commentary. In additional embodiments, the simulated color commentary may provide additional insights into likely upcoming plays or actions, insights into results of plays or actions, incentives to listen to the commentary and/or engage in a wagering game associated with the commentary, or any other relevant information.
In still other embodiments, some embodiments may utilize a completed historical game or action, such as a drive that actually occurred in a played game of football, and have the outcome of each play and of each drive stored in a memory. Then, using information from the databases described above, odds for various scenarios can be generated to reflect potential outcomes of each drive. The odds for plays or actions may be presented in various formats, and various simulated outcomes of different plays could be utilized to create different predictions for the outcome of a drive, such as predicting a punt or a score at the end of a drive. Further, in some embodiments, the generation of probabilities or odds may be verified due to the use of actual, completed historical sports actions or events, so that the end result is known ahead of time. In some other embodiments, where actual historic outcomes of sporting events or actions are used, such historical events could be replayed to a viewer and include overlaid odds or predictions that could be utilized for virtual or simulated sports betting or other entertainment purposes.
As may be appreciated, in any simulation embodiment provided herein, any combination of plays, actions, drive(s), or the like can be simulated in an on-going basis or may be simulated in full at the beginning of an interactive game, such as a wagering game, as soon as the game begins and appropriate rules for the game are set or determined.
In still other embodiments, a random number generator may be utilized to select plays and/or determine results in any play, drive, and/or action. In addition to the verification methodologies discussed above, a random number generator may be utilized to ensure true simulations are provided without predetermined outcomes. In one embodiment, utilizing a historical action database, such as historical action database 130 of
Further, in some embodiments, the play selected by the random number generator could include the outcome of the play. The outcome of the play selected by the random number generator could then be used as an input for the next action or play in the simulation or scenario, or could be interpreted as an action, event, or outcome that ends the simulation. For example, in a simulation of an American football game, if the outcome of the selected play in the simulation results in a touchdown (or other score) or a turnover, that could be utilized as a trigger to end the simulation and, in some other embodiments, resolve any outstanding wagers. In other embodiments, for example in a basketball game, if the selected play in the simulation results in a team scoring a basket to take a three point lead with only two second left in a simulated game, the simulation may then advance to a next play or action in the simulated basketball game, where there may be a high likelihood of the trailing team shooting a long distance three point shot to try to tie the game before it ends. Further, in still other embodiments, the random number generator may select a series of plays to use in a simulation until a result is achieved. In an example related to American football, plays could be effectively simulated until there is a score, a turnover, or the end of a half of game. In an example related to basketball, actions, plays, and possessions may be simulated for a predetermined amount of time, until there is a score, or until the end of a quarter, half, or game. In an example related to baseball, actions and simulations could be utilized for single at-bats, partial or whole innings, groups of innings, or entire games. It may be appreciated that other sports may have simulations that involve various different intervals, possessions, time periods, game scenarios, game lengths, and the like. Additionally, for any sport, scenarios may be simulated that are not necessarily based on an entire game. For example, a two minute drill or red zone possession may be simulated and have results provided or displayed for an American football game. A walk-off scenario with a runner on base and two outs in the bottom of the ninth inning in a baseball game could be simulated. Extra time in a soccer game could be simulated. These are merely examples, however, and the embodiments are not limited thereto.
The foregoing description and accompanying figures illustrate the principles, preferred embodiments and modes of operation of the invention. However, the invention should not be construed as being limited to the particular embodiments discussed above. Additional variations of the embodiments discussed above will be appreciated by those skilled in the art.
Therefore, the above-described embodiments should be regarded as illustrative rather than restrictive. Accordingly, it should be appreciated that variations to those embodiments can be made by those skilled in the art without departing from the scope of the invention as defined by the following claims.
This application is a Continuation-in-Part of International Patent Application No. PCT/US2023/065568, filed on Apr. 10, 2023, and which claims priority from U.S. Provisional Application No. 63/477,892, filed on Dec. 30, 2022, U.S. Provisional Application No. 63/335,368, filed on Apr. 27, 2022, U.S. Provisional Application No. 63/334,275, filed on Apr. 25, 2022, U.S. Provisional Application No. 63/332,744, filed on Apr. 20, 2022, U.S. Provisional Application No. 63/332,330, filed on Apr. 19, 2022, U.S. Provisional Application No. 63/332,333, filed on Apr. 19, 2022, U.S. Provisional Application No. 63/331,951, filed on Apr. 18, 2022, U.S. Provisional Application No. 63/331,338, filed on Apr. 15, 2022, U.S. Provisional No. 63/331,339, filed on Apr. 15, 2022, and U.S. Provisional Application No. 63/328,858, filed on Apr. 8, 2022, the entire contents of which are incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| 63477892 | Dec 2022 | US | |
| 63335368 | Apr 2022 | US | |
| 63334275 | Apr 2022 | US | |
| 63332744 | Apr 2022 | US | |
| 63332330 | Apr 2022 | US | |
| 63332333 | Apr 2022 | US | |
| 63331951 | Apr 2022 | US | |
| 63331338 | Apr 2022 | US | |
| 63331339 | Apr 2022 | US | |
| 63328858 | Apr 2022 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/US2023/065568 | Apr 2023 | US |
| Child | 18486214 | US |
