I am looking building an MLB database with the results of every game for the past 3-5 seasons.
I am adept at scrapping the information off the web - and putting in excel - or maybe I may use access. Either way, I have the ability to scrape the information off the web. My only problem I have had problems finding sites that have this past information. They either have some info, but missing something like starting pitcher - or they have the matchup and who won, but not the score.
1) Can you guys give some suggestions on websites that would have as much matchup information possible for past 3-5 seasons with mandatory data being - final score, starting pitchers, box score of game (i.e. H,W,HR, etc)
2) Hypothetically, what information would you want in the ideal MLB database
off the top of my head I have:
- matchup result - Home or Away - starting pitchers - Starting pitcher stats going into game (ERA, WHIP, etc) - Opening line - Closing line - which game in series - Box Score of game (H, W, Etc) - Team stats going into game (W/L record, OBP, BA, etc)
And once you have all the games some of the information can be calculated - for example (W/L record can be calculated if I have the gamelogs from the beginning of the current season)
Thanks for the help guys.....
Oh and just remember - I tend to share whatever I develop - so think of this as helping yourselves -
0
To remove first post, remove entire topic.
Hey guys,
I am looking building an MLB database with the results of every game for the past 3-5 seasons.
I am adept at scrapping the information off the web - and putting in excel - or maybe I may use access. Either way, I have the ability to scrape the information off the web. My only problem I have had problems finding sites that have this past information. They either have some info, but missing something like starting pitcher - or they have the matchup and who won, but not the score.
1) Can you guys give some suggestions on websites that would have as much matchup information possible for past 3-5 seasons with mandatory data being - final score, starting pitchers, box score of game (i.e. H,W,HR, etc)
2) Hypothetically, what information would you want in the ideal MLB database
off the top of my head I have:
- matchup result - Home or Away - starting pitchers - Starting pitcher stats going into game (ERA, WHIP, etc) - Opening line - Closing line - which game in series - Box Score of game (H, W, Etc) - Team stats going into game (W/L record, OBP, BA, etc)
And once you have all the games some of the information can be calculated - for example (W/L record can be calculated if I have the gamelogs from the beginning of the current season)
Thanks for the help guys.....
Oh and just remember - I tend to share whatever I develop - so think of this as helping yourselves -
Atleast 3 years I'd say, 2007 may be a bit much? What kind of trends are you looking to look at?
Not sure yet - any suggestions are welcome, but one trend I want to verify is I read online that whenever a team scored more than 15 runs - they lose straight up in their next game. Now this said it tracked 100% in the past 3 years....now you can imagine it doesn't produce that many plays a year, but ideally that's what I want to do - find trends that track in the 90-100% range.
I don't care if it only produces a handful of plays a year - because I plan to automate all the tracking - I just need to find a bunch of them.
0
Quote Originally Posted by peter1988:
Atleast 3 years I'd say, 2007 may be a bit much? What kind of trends are you looking to look at?
Not sure yet - any suggestions are welcome, but one trend I want to verify is I read online that whenever a team scored more than 15 runs - they lose straight up in their next game. Now this said it tracked 100% in the past 3 years....now you can imagine it doesn't produce that many plays a year, but ideally that's what I want to do - find trends that track in the 90-100% range.
I don't care if it only produces a handful of plays a year - because I plan to automate all the tracking - I just need to find a bunch of them.
Not sure yet - any suggestions are welcome, but one trend I want to verify is I read online that whenever a team scored more than 15 runs - they lose straight up in their next game. Now this said it tracked 100% in the past 3 years....now you can imagine it doesn't produce that many plays a year, but ideally that's what I want to do - find trends that track in the 90-100% range.
I don't care if it only produces a handful of plays a year - because I plan to automate all the tracking - I just need to find a bunch of them.
Degenerate you are a genius degenerate excel'er. I hope you make this avail for more than just yourself
Btw would you be able to do chase system using excel sheets?
0
Quote Originally Posted by DegenGamble:
Not sure yet - any suggestions are welcome, but one trend I want to verify is I read online that whenever a team scored more than 15 runs - they lose straight up in their next game. Now this said it tracked 100% in the past 3 years....now you can imagine it doesn't produce that many plays a year, but ideally that's what I want to do - find trends that track in the 90-100% range.
I don't care if it only produces a handful of plays a year - because I plan to automate all the tracking - I just need to find a bunch of them.
Degenerate you are a genius degenerate excel'er. I hope you make this avail for more than just yourself
Btw would you be able to do chase system using excel sheets?
The answer to that is YES and YES - I was thinking of automating Ciscos / AllDayGamblers chase systems. I have a couple projects in the works - so a lot on my plate. I will share whatever I do - like I have with everything - and probably put them all in one thread so people know where they are.
0
The answer to that is YES and YES - I was thinking of automating Ciscos / AllDayGamblers chase systems. I have a couple projects in the works - so a lot on my plate. I will share whatever I do - like I have with everything - and probably put them all in one thread so people know where they are.
Degen, would you kindly accept my friend request? I'd be interested in helping you out and sharing ideas that I have. I'm busy with school now but that soon will be over.
0
Degen, would you kindly accept my friend request? I'd be interested in helping you out and sharing ideas that I have. I'm busy with school now but that soon will be over.
Just wondering if you've seen or used this before https://killersports.com/mlb.py/query
Keep up the impeccable work. A wise man once told me to focus on one project at a time, don't overwhelm yourself!
Oil - checked out the site. It's great. I was able to check on the trend if a team wins by 15 or more runs in their previous game do they lose their next game.
Unfortunately - no trend...
I think the site would serve some of the purpose of an excel DB....i'll have to do some more investigation. I am not that familiar with SQDL and what it can do.
I also don't know what trends to look for
0
Quote Originally Posted by oilcountry99:
Just wondering if you've seen or used this before https://killersports.com/mlb.py/query
Keep up the impeccable work. A wise man once told me to focus on one project at a time, don't overwhelm yourself!
Oil - checked out the site. It's great. I was able to check on the trend if a team wins by 15 or more runs in their previous game do they lose their next game.
Unfortunately - no trend...
I think the site would serve some of the purpose of an excel DB....i'll have to do some more investigation. I am not that familiar with SQDL and what it can do.
Just wondering if you've seen or used this before https://killersports.com/mlb.py/query
Keep up the impeccable work. A wise man once told me to focus on one project at a time, don't overwhelm yourself!
Thanks again Oil - this is very helpful after some investigation. It's useful to a point - the one thing I think it lacks is allow you to filer even further down. I will post an example - and maybe everyone can help me figure out how we can make the exampe system have a better win rate
0
Quote Originally Posted by oilcountry99:
Just wondering if you've seen or used this before https://killersports.com/mlb.py/query
Keep up the impeccable work. A wise man once told me to focus on one project at a time, don't overwhelm yourself!
Thanks again Oil - this is very helpful after some investigation. It's useful to a point - the one thing I think it lacks is allow you to filer even further down. I will post an example - and maybe everyone can help me figure out how we can make the exampe system have a better win rate
So here is the example system i'm trying to figure out how to add more filters to bring the win % higher.
Low Scoring Underdogs off a Win
bet the underdog that is +150 or less in the current game who won their previous game, but won their previous game as an underdog by scoring three runs or less.
Here is the query if you're interested:
0<tp:line and 0<t:line and 150>=t:line and tp:runs<=3 and p:W and season = 2012
Either delete or change season parameter for other seasons or all results
I ran it through the killersports.com SQDL query engine and although it looks good over the past 3 years
SU: 341-358 (-0.1 rpg) average line: +123 / -133 on / against: +$6,093 / -$9,473 ROI: +8.7% / -10.2%
It is hit or miss season to season (see below)
2009 SU: 34-44 (-0.3 rpg) average line: +122 / -132 on / against: -$220 / -$120 ROI: -2.8% / -1.2% 2010 SU: 51-53 (-0.1 rpg) average line: +124 / -134 on / against: +$990 2011 SU: 50-40 (0.6 rpg) average line: +119 / -129 on / against: +$1,999 / -$2,499 ROI: +22.2% / -21.5% 2012 SU: 19-19 (-0.4 rpg) average line: +124 / -133 on / against: +$414 / -$574 ROI: +10.9% / -11.4%
So my thoughts are to add additional filters based on the matchup. I added the filter (no dog greater than +150) - because although you will win some of those - chances are dogs over +150 are over 150 for a reason - they will more than likely lose
So what's next? I'm thinking looking at the pitching. But not sure what to look for ERA? WHIP? or do I look at something else.
What do you guys think? What filters should I use? I wouldn't mind if it narrows the plays but produces a better win rate. Because my plan is to find as many of these systems as possible - narrow the filters to produce smaller amount of plays - but then automate them into a spreadsheet.
0
So here is the example system i'm trying to figure out how to add more filters to bring the win % higher.
Low Scoring Underdogs off a Win
bet the underdog that is +150 or less in the current game who won their previous game, but won their previous game as an underdog by scoring three runs or less.
Here is the query if you're interested:
0<tp:line and 0<t:line and 150>=t:line and tp:runs<=3 and p:W and season = 2012
Either delete or change season parameter for other seasons or all results
I ran it through the killersports.com SQDL query engine and although it looks good over the past 3 years
SU: 341-358 (-0.1 rpg) average line: +123 / -133 on / against: +$6,093 / -$9,473 ROI: +8.7% / -10.2%
It is hit or miss season to season (see below)
2009 SU: 34-44 (-0.3 rpg) average line: +122 / -132 on / against: -$220 / -$120 ROI: -2.8% / -1.2% 2010 SU: 51-53 (-0.1 rpg) average line: +124 / -134 on / against: +$990 2011 SU: 50-40 (0.6 rpg) average line: +119 / -129 on / against: +$1,999 / -$2,499 ROI: +22.2% / -21.5% 2012 SU: 19-19 (-0.4 rpg) average line: +124 / -133 on / against: +$414 / -$574 ROI: +10.9% / -11.4%
So my thoughts are to add additional filters based on the matchup. I added the filter (no dog greater than +150) - because although you will win some of those - chances are dogs over +150 are over 150 for a reason - they will more than likely lose
So what's next? I'm thinking looking at the pitching. But not sure what to look for ERA? WHIP? or do I look at something else.
What do you guys think? What filters should I use? I wouldn't mind if it narrows the plays but produces a better win rate. Because my plan is to find as many of these systems as possible - narrow the filters to produce smaller amount of plays - but then automate them into a spreadsheet.
oh and another things the query engine on killersports.com lacks is line movement - I would love to be able to see how often a DOG wins when the the DOG line moves from +150 down to whatever - so the line movement differential.
I think the angle there is sharp betters are looking to play the dog - so if the lines move the favorite down - I would theorize the sharps are betting the dog - so theoretically a system could be
bet the DOG if the line is +145 or less and the line moved at least 20 points from the opening line.
The only way I know of backtesing that is having the opening and closing lines though - not just the closing line.
0
oh and another things the query engine on killersports.com lacks is line movement - I would love to be able to see how often a DOG wins when the the DOG line moves from +150 down to whatever - so the line movement differential.
I think the angle there is sharp betters are looking to play the dog - so if the lines move the favorite down - I would theorize the sharps are betting the dog - so theoretically a system could be
bet the DOG if the line is +145 or less and the line moved at least 20 points from the opening line.
The only way I know of backtesing that is having the opening and closing lines though - not just the closing line.
Here's a link from About.com's sport gambling site. Most of the stuff on there is pretty beginner but the systems seem to be profitable in the past. Each link gives you the description and like 2 or 3 year past performance.
I've also been working on some systems and formulas, I'll send you a message with some of the stuff if that's ok with you.
0
Here's a link from About.com's sport gambling site. Most of the stuff on there is pretty beginner but the systems seem to be profitable in the past. Each link gives you the description and like 2 or 3 year past performance.
Here's a link from About.com's sport gambling site. Most of the stuff on there is pretty beginner but the systems seem to be profitable in the past. Each link gives you the description and like 2 or 3 year past performance.
I've also been working on some systems and formulas, I'll send you a message with some of the stuff if that's ok with you.
sounds good....marko. I'll take a look at the link, maybe run a few through the killersports query engine
0
Quote Originally Posted by marko123:
Here's a link from About.com's sport gambling site. Most of the stuff on there is pretty beginner but the systems seem to be profitable in the past. Each link gives you the description and like 2 or 3 year past performance.
This one looks promising - maybe with additional filters we could bring up the win %
Series Starter System: Bet against teams off a win in a previous game in which they had 10 or more hits and are now at home in the first game of a new series.
Killersports.com SQDL Query t:site=away and opo:runs<op:runs and 10<=op:hits and SG=1 and season = 2009
2009 SU: 94-113 (-0.1 rpg) average line: +126 / -140 on / against: -$15 / -$1,185 ROI: -0.1% / -3.9% 2010 SU: 110-100 (-0.1 rpg) average line: +125 / -139 on / against: +$3,621 / -$4,978 ROI: +15.9% / -16.4% 2011 SU: 100-99 (-0.1 rpg) average line: +117 / -129 on / against: +$1,590 / -$2,654 ROI: +7.3% / -9.9% 2012 SU: 36-40 (0.0 rpg) average line: +115 / -126 on / against: +$260 / -$663 ROI: +3.2% / -6.7%
2009 was a losing season - but that looks to be the year of the favorite - so all dog systems probably lost
0
This one looks promising - maybe with additional filters we could bring up the win %
Series Starter System: Bet against teams off a win in a previous game in which they had 10 or more hits and are now at home in the first game of a new series.
Killersports.com SQDL Query t:site=away and opo:runs<op:runs and 10<=op:hits and SG=1 and season = 2009
2009 SU: 94-113 (-0.1 rpg) average line: +126 / -140 on / against: -$15 / -$1,185 ROI: -0.1% / -3.9% 2010 SU: 110-100 (-0.1 rpg) average line: +125 / -139 on / against: +$3,621 / -$4,978 ROI: +15.9% / -16.4% 2011 SU: 100-99 (-0.1 rpg) average line: +117 / -129 on / against: +$1,590 / -$2,654 ROI: +7.3% / -9.9% 2012 SU: 36-40 (0.0 rpg) average line: +115 / -126 on / against: +$260 / -$663 ROI: +3.2% / -6.7%
2009 was a losing season - but that looks to be the year of the favorite - so all dog systems probably lost
Hot Forum Topics
