Fixture Specific Models for Bet Simulations and Pricing of Real Time Events
Devices, systems, and process for generating fixture specific models for use in adjusting pricing of real-time betting lines are described. A system may include a front-end system including an event-activity-fixture (EAF) server that adjusts betting lines based on results of simulations generated by an EAF simulation server (EAFSS). The EAFSS generates the simulations using fixture specific models that have been generated based upon adaptations of generic models, where the adaptations occurring using historic and real-time EAF data and fixture specific modeling data. The fixture specific models adapted from one or more generic models are leveled and stored in a database for use by the EAFSS on a real-time basis as the EAF occurs. A server for adapting the generic models instantiate one or more computer engines including a model adaptation engine, a generic modeling engine, an EAF data engine, and an FSM leveling engine.
1 . A server comprising:
a data store storing non-transitory first computer instructions for instantiating a model adaptation engine (“MAE”);
a processor configured to execute the first computer instructions and instantiate the MAE;
wherein the MAE, when instantiated by the processor, instructs the server to perform MAE operations (“MAEO”) including:
searching a database for fixture specific model (“FSM”) data (“FSMD”) pertinent to a given event-activity-fixture (“EAF”);
obtaining generic model data (“GMD”) for a given event-activity pairing; and
modifying the GMD with the FSMD to generate FSM adapted data (“FSMAD”) for the given EAF; and
an FSMS bus coupling the processor with the data store.
2 . The server of claim 1 ,
wherein the data store further stores non-transitory second computer instructions for instantiating a generic modeling engine (“GME”);
wherein the processor is further configured to execute the second computer instructions and instantiate the GME; and
wherein the GME, when instantiated by the processor, instructs the server to perform GME operations (“GMEO”) including:
obtaining the GMD from a generic model database (“GMDB”).
3 . The server of claim 2 ,
wherein the GMDB includes non-transitory generic models identifiable based on at least one of an event identifier and an activity identifier.
4 . The server of claim 3 ,
wherein at least one of the generic models has been refined using a supervised artificial intelligence and machine learning (“AI/ML”);
wherein the AI/ML has been initially trained based on past instances of an event-activity pairing;
wherein the AI/ML has been refined based on additional event-activity pairings; and
wherein the GME further performs the GMEO of obtaining the GMD from the GMDB by utilizing the AI/ML to identify, on a Cloud storage device, data pertinent to one or more event-activity pairings that are substantially similar to the given event-activity pairing.
5 . The server of claim 4 ,
wherein the GMEO further include:
receiving event-activity-fixture data (“EAFD”) from at least one of a historical EAF database (“HEAFDB”) and an EAF database (“EAFDB”); and
further obtaining the generic model based on at least one of HEAF data (“HEAFD”) received from the HEAFDB and RTEAF data (“RTEAFD”) received from the EAFDB.
6 . The server of claim 5 ,
wherein the data store further stores non-transitory third computer instructions for instantiating an EAF data engine (“EAFDE”);
wherein the processor is further configured to execute the third computer instructions and instantiate the EAFDE;
wherein the EAFDE, when instantiated by the processor, instructs the server to perform EAFDE operations (“EAFDEO”) including:
requesting the HEAFD from the HEAF database;
receiving the RTEAFD from a real-time data server (“RTDS”); and
providing at least one of the HEAFD and the RTEAFD to the MAE.
7 . The server of claim 6 ,
wherein the data store further stores non-transitory fourth computer instructions for instantiating an FSM leveling engine (“FSMLE”);
wherein the processor is further configured to execute the fourth computer instructions and instantiate the FSMLE;
wherein the FSMLE, when instantiated by the processor, instructs the server to perform FSMLE operations (“FSMLEO”) including:
receiving the FSMAD from the MAE;
leveling the FSMAD received from the MAE in a relational database; and
outputting leveled FSM data (“FSMLD”) for storage in an FSM level database (“FSMLDB”).
8 . The server of claim 7 ,
wherein the FSMLD stored in the FSMLDB is utilized by a simulation server during a real time EAF to simulate one or more outcomes of the real-time EAF; and
wherein the one or more outcomes are utilized by an EAF pricing server to determine one or more betting lines for the event.
9 . A process for generating a fixture specific model (“FSM”) comprising:
searching a database for FSM data (“FSMD”) pertinent to a given event-activity-fixture (“EAF”);
obtaining generic model data (“GMD”) for a given event-activity pairing; and
modifying the GMD with the FSMD to generate FSM adapted data (“FSMAD”) for the given EAF.
10 . The process of claim 9 ,
wherein the GMD is obtained from a generic model database; and
wherein the generic model database includes non-transitory generic models identifiable based on at least one of an event identifier and an activity identifier.
11 . The process of claim 10 ,
wherein at least one of the generic models has been refined using a supervised artificial intelligence and machine learning (“AI/ML”);
wherein the AI/ML has been initially trained based on past instances of an event-activity pairing;
wherein the AI/ML has been refined based on additional event-activity pairings; and
wherein the process further comprises:
utilizing the AI/ML to identify, on a Cloud storage device, data pertinent to one or more event-activity pairings that are substantially similar to the given event-activity pairing.
12 . The process of claim 11 , further comprising:
receiving event-activity-fixture data (“EAFD”) from at least one of a historical EAF database (“HEAFDB”) and a real-time EAF database (“EAFDB”); and
further obtaining the generic model based on HEAF data (“HEAFD”) received from at least one of the HEAFDB and RTEAF data (“RTEAFD”) received from the EAFDB.
13 . The process of claim 12 , further comprising:
requesting the HEAFD from the HEAFDB; and
receiving the RTEAFD from a real-time data server (“RTDS”).
14 . The process of claim 13 , further comprising:
leveling the FSMAD in a relational database; and
outputting leveled FSM data (“FSMLD”) for storage in an FSM level database (“FSMLDB”).
15 . The process of claim 14 ,
wherein the FSMLD stored in the FSMLDB is utilized by a simulation server during a real time EAF to simulate one or more outcomes of the real-time EAF; and
wherein the one or more outcomes are utilized by an EAF pricing server to determine one or more betting lines for the event.
16 . A computer readable medium non-transiently storing non-transitory first computer
instructions which when executed by a processor, in a server, instantiates a model adaptation engine (“MAE”) that instructs the server to perform MAE operations (“MAEO”) including:
searching a database for fixture specific model (“FSM”) data (“FSMD”) pertinent to a given event-activity-fixture (“EAF”);
obtaining generic model data (“GMD”) for a given event-activity pairing; and
modifying the GMD with the FSMD to generate FSM adapted data (“FSMAD”) for the given EAF.
17 . The computer readable medium of claim 16 ,
wherein the computer readable medium further stores non-transitory second computer instructions which, when executed by the processor, instantiates a generic modeling engine (“GME”) which instructs the server to perform GME operations (“GMEO”) including:
obtaining the GMD from a generic model database (“GMDB”); and
wherein the GMDB includes non-transitory generic models identifiable based on at least one of an event identifier and an activity identifier.
18 . The computer readable medium of claim 17 ,
wherein at least one of the generic models has been refined using a supervised artificial intelligence and machine learning (“AI/ML”);
wherein the AI/ML has been initially trained based on past instances of an event-activity pairing;
wherein the AI/ML has been refined based on additional event-activity pairings; and
wherein the GME further performs the GMEO of obtaining the GMD from the GMDB by utilizing the AI/ML to identify, on a Cloud storage device, data pertinent to one or more event-activity pairings that are substantially similar to the given event-activity pairing.
19 . The computer readable medium of claim 17 ,
wherein the computer readable medium further stores non-transitory third computer instructions which, when executed by the processor, instantiates an EAF data engine (“EAFDE”) which instructs the server to perform EAFDE operations (“EAFDEO”) including:
providing event-activity-fixture data (“EAFD”) to the MAE; and
wherein the EAFD is obtained from at least one of a historical EAF database “HEAFDB”) and an EAF database (“EAFDB”).
20 . The computer readable medium of claim 19 ,
wherein the computer readable medium further stores non-transitory fourth computer instructions which, when executed by the processor, instantiates an FSM leveling engine (“FSMLE”) which instructs the server to perform FSMLE operations (“FSMLEO”) including:
receiving the FSMAD from the MAE;
leveling the FSMAD received from the MAE in a relational database; and
outputting leveled FSM data (“FSMLD”) for storage in an FSM level database (“FSMLDB”);
wherein the FSMLD is utilized by a simulation server during a real time EAF to simulate one or more outcomes of the real-time EAF; and
wherein the one or more outcomes are utilized by an EAF pricing server to determine one or more betting lines for the event.