Rx Community Baseball System

crazy · May 19, 2009

Jackle, I knew you used the similar things from the same places I look at, ( a few years now)I love your questions you pose too over there, but you are just scratching the tip of the iceburgh. You need to get a little more creative, yet i don't doubt you are one of the few I have encountered that has that ability.

crazy · May 19, 2009

crazy · May 19, 2009

8-23 (-1.39, 25.8%) -990 870 14-17 (0.11, 45.2%) -760 570 18-12-1 (0.66, 60.0%) 515 -755
41-109 (-1.71, 27.3%) -4205 3560 67-83 (-0.21, 44.7%) -3675 2775 67-74-9 (0.46, 47.5%) -1330 -85
63-101 (-1.11, 38.4%) 115 -1130 88-76 (0.39, 53.7%) -1010 -430 79-73-12 (0.57, 52.0%) -50 -1355
66-109 (-1.18, 37.7%) -100 -985 0-0 (0.00, 0.0%) None None 85-83-7 (0.41, 50.6%) -595 -1165
63-130 (-1.31, 32.6%) -2900 1900 0-0 (0.00, 0.0%) None None 93-86-14 (0.79, 52.0%) 185 -1615
70-154 (-1.62, 31.2%) -3285 1795 0-0 (0.00, 0.0%) None None 110-103-11 (0.97, 51.6%) -330 -1800

just a little fo0d for thought, though the 1.5 line may not be that apealing, think in turms of creating a 1 line or somewhere along those lines

TheJackal · May 19, 2009

crazy said:
Jackle, I knew you used the similar things from the same places I look at, ( a few years now)I love your questions you pose too over there, but you are just scratching the tip of the iceburgh. You need to get a little more creative, yet i don't doubt you are one of the few I have encountered that has that ability.

They rarely help you there. Joe tries to be cryptic to sell his services so you are better off on your own. So I hope you don't think everything you read is all I got. (<)< If 67% (http://www.therxforum.com/showthread.php?t=679322) is just scratching the surface then I'm OK with that. :lol:

crazy · May 19, 2009

crazy · May 19, 2009

Oh no jackle i know thats not all you got...I agree with your comments, I've been busting my chops for 2 years and had to find my own ways around it all....I admire your works Jackle, but I've told you that before

tahoeskibums · May 19, 2009

interested... have access to a team with excellent excel and data programming experience. can actually pull data from different sites into a database and put into data web page.

crazy · May 19, 2009

No, scratching the surface was in regard to the post of league and profit you just posted here.

I have no beef with you mate,

TheJackal · May 19, 2009

crazy said:
No, scratching the surface was in regard to the post of league and profit you just posted here.

I have no beef with you mate,

We are on the same page. just trying to steer the conversation from all the "I'm in's" to some real content that we can start to debate.

crazy · May 19, 2009

meaculpa · May 19, 2009

A lot of you guys are way smarter and more capable with these things than I am, so I don't think I can be of much help. Suppose it could be good to look at some teams that are great OR absolutely worthless at covering the spread and look for like the biggest factor seperating the two groups and work away from that point... Relievers come to mind, but suppose team managers or w/e could have a big influence as well.

Mistermj · May 19, 2009

mea culpa said:
While it's good, it's not mindblowing extreme, so to say. Big favorites are covering the spread often either way (about 80%). 44-7 is about 86.3% and thus quite a lot above that average, but still... I don't think it's enough to focus on this exclusively.

Either way, a lot of you guys are way smarter and more capable with these things than I am, so I don't think I can be of much help. Suppose it could be good to look at some teams that are great OR absolutely worthless at covering the spread and look for like the biggest factor seperating the two groups and work away from that point... Relievers come to mind, but suppose team managers or w/e could have a big influence as well.

We have to think in terms of 1000's of games.

51 game trends are completely meaningless. One season trends are probably meaningless as well.

Not trying to bust balls...but I'm not going to do anything unless it's an assumption on 1000's of games. :103631605

I think we really need to boil it down to this...what factors lead to a team beating the run line...over a period of time?

And then it has to be stats that are readily accessible for a program to automate and use.

It must be fairly simple and straight forward.

TheJackal · May 19, 2009

The logical approach:

What are we trying to predict? runs (or more accurately run differential)
What has an affect on runs? Pitchers, teams (lineup), ballpark, home/away, umpires.
How do each of these influence runs? Pitchers - era, teams-ba, hits,avg runs scored, ballpark- over/under park, umpires- over/under ump, strike %, avg runs

plenty of other stats to but I think you get the idea. Just start to break it down a level at a time and then use all the stats together to predict the runs for each team (hence the run differential).

That is the outline (or project plan). Lets start formulating the specifications and then we can put theory to practice and test the thing. (testing is likely a few weeks off but just trying to jump start this thing.)

barman · May 19, 2009

Friendly suggestion that any RL data crunching include results for both +1.5 and -1.5 in all games.

Simply tallying "RL results" can leave the analyst a bit short on info since for many teams "RLs" could be +1.5 some days and -1.5 on other days.

I've learned in Hockey capping that it's very, very helpful to know what a team's results are overall when games are graded at +1.5 and also at -1.5

Thanks to any and all who contribute useful information to this burgeoning project

bar

geeze · May 19, 2009

.Should we consider adding a -1 line also for comparison?

Mistermj · May 19, 2009

barman said:
Friendly suggestion that any RL data crunching include results for both +1.5 and -1.5 in all games.

Simply tallying "RL results" can leave the analyst a bit short on info since for many teams "RLs" could be +1.5 some days and -1.5 on other days.

I've learned in Hockey capping that it's very, very helpful to know what a team's results are overall when games are graded at +1.5 and also at -1.5

Thanks to any and all who contribute useful information to this burgeoning project

bar

I think the run line analysis article is really confusing people.

They broke it all down according to the ML odds...not the Run line...so it is all figured in. The highest dogs win by more than 1 run at a little over 70%, which is in range of the rate that the highest faves win by more than 1 run.

That sounds surprising...until you consider that it only includes dogs (and faves) that win the game. Picking +200 winners is the hard part and I have no intention of trying it.

I really think we are getting off track worrying about analyzing the run lines.

We need winners that win by more than 1 run...thats it...the run line odds will take care of itself.

Run line analysis is a red herring that will side track us from the main goal in my opinion.
===========
Probability of winning by more than 1 run given that the team wins the game straight up.

League Wide: 72.5%
+200 or more: 70.2%
+190s: 77.0%
+180s: 76.4%
+170s: 67.4%
+160s: 75.3%
+150s: 72.9%
+140s: 70.9%
+130s: 73.5%
+120s: 68.0%
+110s: 72.4%
+100s: 72.9%
-100s: 68.9%
-110s: 70.6%
-120s: 72.4%
-130s: 72.4%
-140s: 74.7%
-150s: 74.5%
-160s: 75.3%
-170s: 71.9%
-180s: 70.8%
-190s: 77.5%
-200s: 76.2%

Favorites that are -120 or better beat the runline at a higher percentage than the league average (73.6%)

Run Line Analysis for Baseball (19 May 2009)

http://www.easybaseballbetting.com/run_line_analysis.html

barman · May 19, 2009

My bad....I skimmed the thread too quick. Thought you might also be seeking some data that could help identify when a team is in a good spot to be wagered at +1.5

Though the +1.5 is most often a minus money wager, if it stays at -145 or less, then there are plenty o' good spots that present themselves.

We'll try that elsewhere when the time is right

Mistermj · May 19, 2009

TheJackal said:
The logical approach:

What are we trying to predict? runs (or more accurately run differential)
What has an affect on runs? Pitchers, teams (lineup), ballpark, home/away, umpires.
How do each of these influence runs? Pitchers - era, teams-ba, hits,avg runs scored, ballpark- over/under park, umpires- over/under ump, strike %, avg runs

plenty of other stats to but I think you get the idea. Just start to break it down a level at a time and then use all the stats together to predict the runs for each team (hence the run differential).

That is the outline (or project plan). Lets start formulating the specifications and then we can put theory to practice and test the thing. (testing is likely a few weeks off but just trying to jump start this thing.)

Run differential is the key here in my opinion.

I'm not sure how much we have to add to it...but keep in mind that run differential already includes most of the subsets of stats you are talking about.

In other words...a top run differential team would already most likely be a team with an ERA advantage...and a batting advantage...runs scored ect...thats what makes them a top run diff team in the first place.

Make sense?

I'm lobbying for making run differential as the core stat...because it represents the outcome of all the others.

TraderMike · May 20, 2009

Sorry I posted the article everyone. In my humble opinion, though, I think that we should simply use any RL system developed here with with the regular plays generated by the current system. MJ is right in saying that most of the pertinent statistics are going to be factored into the RL. However, given that we have a reguar system play, finding a set of circumstances where the RL could possibly provide extra edge could help us accumulate some extra units over time; if it doesn't, Rl vs. ML P&L should be approximately the same in the long run anyway. I don't think we should try to develop a pure RL system where, conceivably, the RL could be a system play and the ML would not, at least not to use in conjunction with MJ's system. I think we should try to keep it simple and develop a percentage for the winning team to win by more than one, for a given set of circumstances, to set against probabilty the book assigned to that outcome reflected in the difference between the Ml and RL. If we do this, once we get a system play, we simply flag the games with circumstances we have chosen and compare our number to the book's RL.

Off the bat, I think away national league teams during interleague would be an interesting statistic to look at. Also, I believe the differential between the SP ERA and Bullpen of our projecting losing team could be helpful in determing a possible RL, or even reverse RL play. I don't have the expertise(yet) to acquire and process all of this data, so I'm sorry to say that my ability to contribute will be limted to analyzing figures other people were able to produce. Thank you to everyone with the resources and know how that have volunteered to help analyze this data. BOL.

SpreadOffense · May 20, 2009

Mistermj said:
I think the run line analysis article is really confusing people.

They broke it all down according to the ML odds...not the Run line...so it is all figured in. The highest dogs win by more than 1 run at a little over 70%, which is in range of the rate that the highest faves win by more than 1 run.

That sounds surprising...until you consider that it only includes dogs (and faves) that win the game. Picking +200 winners is the hard part and I have no intention of trying it.

I really think we are getting off track worrying about analyzing the run lines.

We need winners that win by more than 1 run...thats it...the run line odds will take care of itself.

Run line analysis is a red herring that will side track us from the main goal in my opinion.
===========
Probability of winning by more than 1 run given that the team wins the game straight up.

League Wide: 72.5%
+200 or more: 70.2%
+190s: 77.0%
+180s: 76.4%
+170s: 67.4%
+160s: 75.3%
+150s: 72.9%
+140s: 70.9%
+130s: 73.5%
+120s: 68.0%
+110s: 72.4%
+100s: 72.9%
-100s: 68.9%
-110s: 70.6%
-120s: 72.4%
-130s: 72.4%
-140s: 74.7%
-150s: 74.5%
-160s: 75.3%
-170s: 71.9%
-180s: 70.8%
-190s: 77.5%
-200s: 76.2%

Favorites that are -120 or better beat the runline at a higher percentage than the league average (73.6%)

Run Line Analysis for Baseball (19 May 2009)

http://www.easybaseballbetting.com/run_line_analysis.html

Definitely need 1000's of games for trends (not trends based on days of the week or night/day games or any of that bullshit, etc, but rather based on stats, umps, etc) and results to analyze. We need to find a good, efficient way to backtrack data. Anything less than that and it's no where near normal and you're just picking numbers out of hat in terms of percentages.

From the numbers above (what's the sample size on that MJ, by the way?), IMO, we should start by only analyzing faves from -140 to -160 and figure out a good way to pick those winners. **If you hit simply game winners at a 55% clip (not all that difficult when betting on -150 faves), and your RL hits 74% (around what it is at -140 to -160) then that's a good profit**

So in my opinion, if we can find good times/opportunities to bet the runline, we might be able to hit that at around 80%. If the average for -140 to -160 is 74.8% RL hits, then it seems to me that if we find a way to determine better than average RL cover opportunities based on trends and stats (what I'm thinking this thread might be able to accomplish with some of the minds we have in here), 80% is not out of the question. And if we find a way to increase that RL hit % to 80 as stated, you only have to hit 50% (of -140 to -160 faves no less) for a profit.

Let me know if anyone followed that. I think we should get started on tracking data for better than average RL covers (through trends or whatever, etc) to determine if we can hit that a better %. Next will come the picking 50% game winners at an average of -150, which should be a less complicated process....

***Rx Community Baseball System***

New member

New member

New member

New member

New member

New member

New member

New member

New member

New member

New member

Member

New member

Honey Badger Don't Give A Shit

New member

Member

Honey Badger Don't Give A Shit

Member

New member

New member

Forum statistics

Share this page

Rx Community Baseball System