| Author: |
[Systems & Strategies] Topic: MLB Prediction Formula |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#26 Posted: 6/21/2012 11:23:06 AM v2.0 Plays
YTD: 9-4 69% +4.54u
rockies 5.2 phillies 10.9
dodgers 5.6 athletics 3.9
rays 4.0 nationals 5.5
twins 3.5 pirates 7.1
|
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#27 Posted: 6/21/2012 11:35:36 AM v3.0
YTD: 0-0 -% +0.00u
No plays today. Not enough data on pitchers. |
|
quote |
|
gbpackman |
RSI  View Space | Blog | Friends | Playbook | |

Veteran
Joined: Oct 2011
Posts: 2436
Location: Mauritius |
#28 Posted: 6/21/2012 2:53:34 PM Do you factor in juice? |
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#29 Posted: 6/21/2012 3:06:45 PM Yea I used the closing lines on Covers. If the juice is really high though i usually use RL -1 which I record as -150 with the ML being -180 or greater.
This is all still preliminary and I wouldn't bet on anything yet. Degen and I are still doing a lot of adjusting and calibrating but I see it being up and running by Monday possibly.
I'd rather post all this prelimenary stuff and let everyone see what we are doing rather then set it all up and then start posting plays with no background information. |
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#30 Posted: 6/21/2012 9:59:40 PM Rough day today with the end of series games. Start fresh series tomorrow. |
|
quote |
|
|
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#31 Posted: 6/22/2012 4:28:16 PM Degen and I have been working out different subsystems with different adjustments to each of them
version 1.0.1 is what I have been posting each day. version 1.0.2 uses career and season ERA for pitchers. version 1.0.3 takes hot and cold pitchers into consideration version 1.0 are plays that all the above systems agree on version 1.1 is the same as 1.0 but uses a 1.0 run margin filter version 1.2 are v1.0 plays with a 2.0 margin, we play RL on these.
All of these subsystems are repeated with v2. v2 takes home and away stats into consideration
v3 are plays that every single subsystem agree on. These will be higher unit plays.
|
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#32 Posted: 6/22/2012 4:28:31 PM Today's games
v1.0.1 dodgers athletics mariners orioles phillies tigers jays red sox reds yankees indians rangers white sox cards dbacks
|
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#33 Posted: 6/22/2012 4:29:31 PM v1.0.2dodgers athletics mariners orioles phillies tigers jays red sox reds yankees indians rangers white sox cards dbacks |
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#34 Posted: 6/22/2012 4:30:23 PM v1.0.3dodgers athletics mariners nationals phillies pirates jays reds yankees indians rangers white sox royals dbacks |
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#35 Posted: 6/22/2012 4:32:02 PM v1.0dodgers athletics mariners phillies jays reds yankees indians rangers white sox dbacks |
|
quote |
|
Stake24 |
View Space | Friends | Playbook | |

Prospect
Joined: Apr 2007
Posts: 323
Location: Nevada |
#36 Posted: 6/23/2012 10:09:21 PM nice work Marko and Degen keep it up..... I recommend using jeff sagarin's pitcher data as he takes it to another level.... take a look

|
|
quote |
|
DegenGamble |
View Space | Friends | Playbook | |

Rookie
Joined: Oct 2011
Posts: 908
Location: New Jersey |
#37 Posted: 6/25/2012 10:02:33 AM QUOTE Originally Posted by Stake24:
nice work Marko and Degen keep it up..... I recommend using jeff sagarin's pitcher data as he takes it to another level.... take a look

Thanks Stake - I have seen a bunch of systems that incorporate Jeff Sagarin's pitcher data - i.e. If the pitcher is on the list - scratch the play.
So I am defintely going to take a look and potentially incorporate it. |
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#38 Posted: 6/25/2012 1:11:00 PM Great link Stake!
Alright, so Degen and I have worked out through the kinks and with the new system, it showed a considerable loss on friday which seemed to be a bad day overall but it's done really well on Saturday and Sunday. Playing the ML and RL those two days you'd be up 23 units.
It will continue being a work in progress with small adjustments and the addition of filters but it seems pretty solid at this point.
I know Degen has played the weekend plays and has made a profit. I'm going to start placing money on the picks today. I play ML and RL on each play but play whatever makes you comfortable if you decide to tail.
I'll start the record fresh from today, so YTD: 0-0 +0units
Today's Plays:
DODGERS YANKEES PHILLIES STL METS ASTROS RANGERS RAYS WHITE SOX NATIONALS
There might be changes to the picks if pitchers change and whatnot so check back here later tonight, theyre all night games. I'll post the plays and any changes on my twitter as well.
@MarkoV12345
|
|
quote |
|
Stake24 |
View Space | Friends | Playbook | |

Prospect
Joined: Apr 2007
Posts: 323
Location: Nevada |
#39 Posted: 6/25/2012 1:15:10 PM Well you wouldnt scratch the play if they are on the list. I like his numbers because they incorporate more stats into the ERA, and the rickey and mob give you a more accurate assessment of power. I'm not really using any systems to bet I use his numbers compared to normal ERA to see who is over valued and under valued. But I like what I'm seeing with this prediction model.
On another note I was thinking the other day about maybe categorizing pitchers as knuckleballer, power thrower, 4 pitch, 5 pitch and so on and keeping track of how teams do against the different type of pitchers. But I wouldn't even know where to start there is just too much data for me...maybe you could automate it?
|
|
quote |
|
DegenGamble |
View Space | Friends | Playbook | |

Rookie
Joined: Oct 2011
Posts: 908
Location: New Jersey |
#40 Posted: 6/25/2012 1:29:08 PM QUOTE Originally Posted by marko123: Great link Stake!
Alright, so Degen and I have worked out through the kinks and with the new system, it showed a considerable loss on friday which seemed to be a bad day overall but it's done really well on Saturday and Sunday. Playing the ML and RL those two days you'd be up 23 units.
It will continue being a work in progress with small adjustments and the addition of filters but it seems pretty solid at this point.
I know Degen has played the weekend plays and has made a profit. I'm going to start placing money on the picks today. I play ML and RL on each play but play whatever makes you comfortable if you decide to tail.
I'll start the record fresh from today, so YTD: 0-0 +0units
Today's Plays:
DODGERS
YANKEES
PHILLIES
STL
METS
ASTROS
RANGERS
RAYS
WHITE SOX
NATIONALS
There might be changes to the picks if pitchers change and whatnot so check back here later tonight, theyre all night games. I'll post the plays and any changes on my twitter as well.
@MarkoV12345
MLB RUN ESTIMATOR
This spreadsheet is based off of Marko123 – MLB Prediction Formula thread. As Marko mentions - this is a work in progress in its early stages and still work being done by Marko to fine tune.
Features:
· Today’s matchups are updated from baseballpress.com when you click the “Update Lineup” button
· Pitching and Team stats needed to complete the Run Estimation formula are updated when you click on the “Update Pitching Stats” and “Update team Stats” Button
· NOTE: From time to time a new pitcher (say one called up from AAA) or if a pitcher goes to a different team – an error will occur when updating the pitching stats. You will need to update the Pitcher List with the proper links from baseball reference and re-run the query.
You can download from the link below:
https://docs.google.com/open?id=0B5dydT9Oe6Jbdkd1alJRRENtVnM
Also available in my spreadhseets and more thread.
|
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#41 Posted: 6/25/2012 1:34:20 PM I forgot to say that if the play is a dog I play alternate RL. That should be an option on your book. The lines are like +250 for dogs -1.5 |
|
quote |
|
DegenGamble |
View Space | Friends | Playbook | |

Rookie
Joined: Oct 2011
Posts: 908
Location: New Jersey |
#42 Posted: 6/25/2012 1:35:22 PM QUOTE Originally Posted by Stake24:
Well you wouldnt scratch the play if they are on the list. I like his numbers because they incorporate more stats into the ERA, and the rickey and mob give you a more accurate assessment of power. I'm not really using any systems to bet I use his numbers compared to normal ERA to see who is over valued and under valued. But I like what I'm seeing with this prediction model.
On another note I was thinking the other day about maybe categorizing pitchers as knuckleballer, power thrower, 4 pitch, 5 pitch and so on and keeping track of how teams do against the different type of pitchers. But I wouldn't even know where to start there is just too much data for me...maybe you could automate it?
Interesting Stake - i will have to look at Rickey and MOB in more detail.....thanks
As far as catagorizing pitchers - pretty much anything can be automated - but I'm not sure I see the value in catagorizing the pitchers since a knuckleball pitcher like RA Dickey for example may not perform the same against a specific team as say another knucleball pitcher. I see where you're going though - see if a specific team hits a specific pitcher type better or worse. But there is no way to know if that team doesn't hit that pitcher type well - OR that specific PITCHER well. I think you would be better off looking at past track record of team vs. that pitcher and if no data is available - lay off that game. |
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#43 Posted: 6/26/2012 1:51:24 PM Not a good day yesterday. Seems like this system always does horrible on the first game of a new series. This also happened on friday.
YTD: 5-15 (RL and ML plays) -11.25u
Today's Plays
Mariners Angels Phillies Jays Reds Cards Braves Mets Astros Rangers Rays White Sox Nationals
Degen and I are currently working on finding the proper filters for this system and there should be less and less plays as we move through the season, leaving only the most confident plays and getting rid of leans.
|
|
quote |
|
marko123 |
View Space | Friends | Playbook | |

Prospect
Joined: Jan 2012
Posts: 120
Location: |
#44 Posted: 6/27/2012 11:45:21 AM Yesterday went a lot better.
16-10 +7.6 units
YTD: 21-25 -3.65 units
Today's Picks:
YANKEES WHITE SOX RED SOX REDS ATHLETICS DODGERS ORIOLES PHILLIES CARDS ASTROS RANGERS
|
|
quote |