Here's an Idea

I like working with baseball stats trying to come up with formulas and such. I had the thought that using Hits/Run would be a decent measure of how teams tended to score their runs. A team that had a high Hits/Run would be a team that is more station to station requiring a string of hits to score, like San Diego (2.29) and Atlanta (2.33), highest in the league. Teams with a low Hits/Run were more efficient with their hits, needing maybe just one to score with a long fly, OAK (1.70) and TOR (1.78), being the lowest two. The correlation between Hits/Run and Percent of Overs is -.562, pretty good for just a basic ratio.

The OU Record for those 4 is:

71-41 Under for ATL and SDN

65-52 Over for OAK and TOR

I pulled stats into a spreadsheet from Baseball Prospectus and used 2 other stats, OBPxSLG and BP's True Average. OBPxSLG provided a slightly higher correlation (.568) that OPS so I used that instead. True Average had the highest correlation to Overs at .615.

I took those 3 stats and made a simple formula:

( ( True Average*(.615)+OBPxSLG*(.568) ) * 100 ) / (Hits/Run) = OU Rating

The way I see it, the higher the number the more likely the team is to go Over, I called this the OU Rating. Then I added the OU Ratings for the 2 teams in each matchup today, and called the sum the OU Score. For today, 6/6, these were the OU Scores

Oak/Bal  26.38
ChA/LAA  25.01
LAN/Col  24.55
SLN/Tor  24.24
Mia/ChN  23.26
Cle/Tex  23.18
NYN/SFN  22.91
Bos/Det  22.81
Hou/Min  22.57
Sea/TBA  21.66
Atl/Ari  19.97
NYA/KCA  19.91
Was/SDN  19.88
Phi/Cin  19.32

So here's what I was thinking

On Mondays and Fridays I want to start a 3 game chase on whatever series has the highest OU Score to go Over and for the lowest OU Score to go Under. I only say Mondays and Fridays because I have too much else going on to pay attention to when every series starts, and Mondays and Fridays typically have only game 1's. At least, it will be easier to know which games are game 1's.

The money management I will be using is to flat bet the 3 games. I'm very risk averse when it comes to laying juice.

**Important**

I have done absolutely NO backtesting on this, I'm simply testing a theory, and am only starting with a bankroll that I am okay with losing should this not be as good an idea as I believe it to be.

I'm not going to start this until Monday June 9th, but I'm posting it now so that if I've completely gone off base with something, someone can point it out to me before any money gets lost. But just for fun the games for this weekend's chases would have been Oakland and Baltimore to go Over once, and Philadelphia and Cincinnati to go Under once.

I would love to get any thoughts, positive or negative that anybody has regarding this idea. Best of luck to everyone this weekend.

CAP

good look

The problem is, you are assuming the trends you spit out will continue. You need more clues as to when to hop on board or not.  I don't have all the answers, but I didn't need a formula to tell you that Tor hits a lot of HR and SD does not.

That's fair. But isn't that just called handicapping? Figuring what stats are useful for projecting future results? If you like the Over in a game, it's likely because you think some trends will continue into the current game. As far as the home runs go, I get it. But wouldn't the fact that Toronto hits a lot of home runs be one of the factors you consider when betting the total on their games? I'm not just looking at home runs, though. Seattle has 48 home runs, better than just 7 teams. However they have the 5th lowest hits/run, 1.93.

I do appreciate your thoughts and respect your opinions, thank you for giving them.



Since this is just an exercise really I made a tweak to OU Score number, so that it incorporated at least some of the current game conditions, by dividing that number by the game total. And instead of chasing a specific series, I'm going to make it a chase on whatever games are the high and low for each day. One final adjustment is that I'm not going to play them separately, I'm going to play a 2 team parlay on the Over and the Under for that day. Eliminating the juice and reducing the risk. That way, if either of the first 2 days of games hit, they should produce a profit, and the only losses would be in the event of a 3 day streak, losing 3 units. Like I said before I'm only using a very small bankroll for this.

I'm still going to start on Monday the 9th though.

June 9

Atlanta / Colorado U10.5

Combined - 21.57
Total - 10.5

2.05


Washington / San Francisco O6.5

Combined - 22.79
Total - 6.5

3.51

Parlay:
ATL/COL U10.5 (-120)
WAS/SFN O6.5 (-120)

1 to win 2.36

June 9

Parlay:
ATL/COL U10.5 (-120)  UNDER
WAS/SFN O6.5 (-120)  OVER

1 to win 2.36

YTD 1-0 (2-0) +2.36 Units

I don't know why I called this a chase system since I'm flat betting it and doing it every day.

Nice start.

June 10

Atlanta / Colorado U10

Combined - 21.40
Total - 10

2.14


Washington / San Francisco O6.5

Combined - 22.89
Total - 6.5

3.52

Parlay:
ATL/COL U10 (-110)
WAS/SFN O6.5 (-105)

1 to win 2.73

YTD 1-0 (2-0) +2.36 Units

June 10

Parlay:
ATL/COL U10 (-110)  OVER
WAS/SFN O6.5 (-105)  UNDER

1 to win 2.73

YTD 1-1 (2-2) +1.36 Units

Good luck buddy

BCap,

I like the ideas you have put forward here and am following the system.  A question about your last tweak.

Do you think tweaking the OU Score by multiplying it by the Line Total will help or hurt in the long run?  It seems by doing so, it throws your original formula partly out the window.  The lowest OU Totals are not the plays for yesterday or today.

SD vs PHI OU Total is 18.47 today but the OU Score is 2.64
ATL vs COL OU Total is 22.51 today but the OU Score is 2.25 (because ot the line total of 7 and 10 respectively).

Canis

You're right it does make the lined total worth the same as the OU Score in the end. For today I squared the combined OU Scores to make that number more important, then divided by the total. Using the squared method I come up with SD/PHI and OAK/ANA. OAK/ANA is so much higher than all the others, squaring it just made the gap larger.

June 11

San Diego / Philadelphia U7

Combined - 17.91
Total - 7

45.80


Oakland / LA Angels O8

Combined - 27.00
Total - 8

91.09

Parlay:
SD/PHI U7 (-110)
OAK/ANA O8 (-115)

1 to win 2.57

YTD 1-1 (2-2) +1.36 Units

In you other thread you have Oak under is right??

Yeah, the trends say Under but we'll see.

June 11

Parlay:
SD/PHI U7 (-110)  UNDER
OAK/ANA O8 (-115)  PUSH

1 to win 2.57

YTD 2-1 (3-2-1) +3.93 Units

BetOnline paid it off as a win so I'll take it.

June 12

San Diego / Philadelphia U8

Combined - 17.74
Total - 8

39.36


Detroit / Chicago O7

Combined - 23.38
Total - 7

78.07


Parlay:
SDN/PHI U8 (-110)
DET/CHA O7 (-120)

1 to win 2.50

YTD 2-1 (3-2-1) +3.93 Units

good luck

BCapp,

I am using data that I update every morning from here:

https://www.baseballprospectus.com/sortable/index.php?cid=1662625, _

I perform your calculations using R, H, OBP, SLG and TAv from that table.  I am also using your constants of .615 and .568, yet my result are off from yours by a few points.  I am wondering if we are using the same days data, or if I am doing something wrong.

Any help is appreciated.

Thanks,

Canis

Canis,

Shoot me an email at bcapriotti88@gmail.com and I can send you the copy of the spreadsheet I'm using and we can compare to see why we have different numbers. I'm not an expert with it so it may just be something I didn't explain correctly.

Cap

June 12

Parlay:
SDN/PHI U8 (-110)  OVER
DET/CHA O7 (-120)  UNDER

1 to win 2.50

YTD 2-2 (3-4-1) +2.93 Units

June 13

Chicago / Philadelphia U8.5

Combined - 19.93
Total - 8.5

46.74


Colorado / San Francisco O7

Combined - 24.32
Total - 7

84.46


Parlay:
CHN/PHI U8.5 (-125)
COL/SFN O7 (-115)

1 to win 2.37

YTD 2-2 (3-4-1) +2.93 Units

BCap,

I have everything worked out now with the calcs.  Thanks so much for the help.  Interesting to note that a half run difference in the Total for the COL/SFN game (you 7, me 7.5) is enough to switch my Over game to NYY vs OAK.   Will have to think on how to make these numbers not so sensitive.

Canis

Yeah, those top 3 are all pretty close. It's still a work in progress though. Eventually I want to incorporate starting pitcher hit/run data as well. Thanks for checking on that.

Canis,

I just discovered an error I made in my sheet. I was using .615 as the constant for OBP*SLG instead of TAv and .568 on TAv. Therefore the numbers you had were likely correct, and I'm revising mine now. Too late to change today's picks but glad I caught it.

Cap