Baseball semi-arbitrage system

**Ominous** · 06-06-09, 04:25 PM

DOnt have that much great info for you, but i can say this.

1. DONT CHASE any serious system or capper should not chase losses. This will only increase your deviation and possibly create an illusion of a winning system from lots of small wins with the risk of a big loss lurking around the corner.

2. Your "system" or possible "semi arbitage" seems to simple Im not saying it cant work, its just that most general and simple systems in sports usually dont work so well. However, if your data truly does back it up... the edge predicted in your oringinal post does however seem quite bizzare and I doubt it can be true.

3. Trying to eliminate certain games to find better bets definently seems like a good idea. O/U line would be an obvious stratingpoint but I would guess that it is already factored into the RL odds. Anyway gl and if you have more data to support your theroy please feel free to share

**MonkeyF0cker** · 06-06-09, 05:54 PM

Look at totals, Lenk.

**TomG** · 06-06-09, 11:14 PM

I tried this before and it doesn't work because when games are tied 3-3 for example near the end of the game, then most of the time they end in a 1 run game. This happens a lot more than you think. I found the trick was not to bet the underdog ML and the favorite -1.5, but to take it further. Bet the dog +1.5 and the favorite -2.5. That way you make money on 1-run victories which happen more often and you just have to dodge 2-run games which is more rare.

**shhhhh22** · 06-07-09, 02:15 AM

Originally posted by TomG

I tried this before and it doesn't work because when games are tied 3-3 for example near the end of the game, then most of the time they end in a 1 run game. This happens a lot more than you think. I found the trick was not to bet the underdog ML and the favorite -1.5, but to take it further. Bet the dog +1.5 and the favorite -2.5. That way you make money on 1-run victories which happen more often and you just have to dodge 2-run games which is more rare.

I really like this middling strat... I wonder if anyone has any data to back this up. I tend to agree 2 run games will be far less common than 1 run games.

Great thread guys. I love it when this group is so on topic! We all benefit!

**scurd8** · 06-08-09, 12:58 PM

Looks like Dodgers just had a decent run of 1 run games at home lately.

06/06/09 covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/teams/team2958.html W covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/results/2009/boxscore291440.html covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player42256.html (R) covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player43412.html (R) W -114 U 9 -125 06/05/09 covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/teams/team2958.html W covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/results/2009/boxscore291439.html covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player40531.html (L) covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player40638.html (L) W -122 U 9.5 -123 06/04/09 covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/teams/team2958.html L covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/results/2009/boxscore291438.html covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player42863.html (L) covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player43488.html (L) L -102 U 8 -116 06/03/09 covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/teams/team2968.html W covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/results/2009/boxscore290633.html covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player41055.html (R) covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player42884.html (R) W -191 U 8 -118 06/02/09 covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/teams/team2968.html W covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/results/2009/boxscore290632.html covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player41452.html (R) covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player40814.html (L) W -117 O 7.5 112 06/01/09 covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/teams/team2968.html L covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/results/2009/boxscore290631.html covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player43236.html (R) covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/players/player43412.html (R) L -161 U 9 -113

**Ominous** · 06-08-09, 11:49 PM

Just some random input. I do defenentliy not believe the +1,5/-2,5 could work.

In order for it to work, one of the lines i.e. Either the +1,5 or the -2,5 line needs to have value. I mean, at least one of the legs needs to be set wrong. It would seem like pinnacle expands thier +-1,5 odds to genereate the +-2,5 odds for instance. That means that the -2,5 odds is a function of the +1,5.

Betting +1,5/-2,5 is basically the same as betting +1,5/-1,5 in regards to value.

Like Ive said before, I do not belive -1,5 and ML could work either but here I am not as sure. There is more factors that differ in the setting of the ML line and the -1,5 line than there is in the case of +2,5/-1,5

**smallbluedonkey** · 06-09-09, 07:03 AM

Although they occur less often, you are much better off betting the RL for road favorites under this kind of system.

Home favs win by a margin of 1 run in about 18% of cases
Road favs win by a margin of 1 run in about 12% of cases

This is becaues of the fact that the final innings stops as soon as a run is scored for home favs, but road favs never face this kind of immediate end to a game - they can make multiple runs in the final inning without the game ending early.

I ran some quick and dirty analyses and looking at the run total doesn't really have a statistically significant effect on these stats.

The main issue with this system is that you usually only win .2 or .3 units on average for a "win" and you lose 2 units for each game that loses. So on the raw numbers you have a negative EV:

0.25 units x 0.88 = 0.22
-2 units x 0.12 = -0.24

EV = -0.02 units per bet

**cinpls081** · 06-09-09, 07:26 AM

your thoughts having ZERO to do with an arb system. arbing is to bet both sides of a game and if its played you win. What youhave proposed is another gambling system where in the end you will lose money not make money

**u21c3f6** · 06-09-09, 09:01 AM

As has been posted in many other threads, the lines on MLB are very close to the real probabilities. So unless you have an ability to know which way the line will move and bet accordingly and/or find a subset of games that have a higher than normal win %, it will be very difficult to make money on your proposed hedge of Away ML & Home -1.5 RL.

However, if you get the best pricing you can before the game and couple it with live wagering, you will have an environment where you can convert your hedge into +EV. You have to be careful with liquidity. Before the game has a tremendous amount of liquidity while live action not so much. Stay with the more popular matchups and/or major network games for the best liquidity.

Start small, watch and learn. Look to see how the line changes as the game progresses and runs are scored. Also notice the liquidity to get an idea of the maximum you should risk before the game that can be matched during live wagering.

The edge is that the lines during live wagering are priced for how the game will end, not how the game proceeds to get there.

Good luck. Joe.

**hajune** · 06-15-09, 06:00 PM

Originally posted by Lenk

I've been studying something that might BE of interest.

In the last 396 basball games, 292 of them were home favorites. In 247 of them or 84.5%% either (1) the home favorite won by 2 or more runs, or (2) the visiting dog won.

I've been playing around trying to create a semi-arbitrage system where you lose only if the home favorite wins by 1 run.

So, e.g., if the home favorite is -1.5 runs +160 & the visiting dog is +110, you bet $1,000 on the run line & $1,238 on the money line to net $362.

Yes, the odds are real high. In this example 3.96 to 1. But keep in mind, 84.5% of all games win.

At odds as high as 6 to 1 you break even at 85.7%.

What I am looking for are ways to shift the odds of winning in my favor by (1) eliminating games that are likely to fall 1 run or (2) play games that are likely to have a lot of scoring & win by more than 1 run.

A couple of obvious ways to do that is to not play games at San Diego or Wrigley with the wind blowing in.
Also eliminating games where both teams score little & give up few runs (e.g. SF has scored & allowed the fewest runs year to date. When they play STL (3rd lowest, 2nd lowest in the NL) it would be risky.

I'm looking for the less obvious ways to eliminate games or select games to play. This is where I need help.

I don't have enough data points yet, but it looks pretty rare that a home favorite wins by just 1 run two consecutive days. So one thought is to play following a 1 run game.

I'm wondering what your thoughts are.

Thanks a lot.

Play teams/games that score a lot orf runs.
Play games where the ERA of both starters is high.
Play games that has at least one bullpen with ERA of 4.00 or more
The more runs scored the less likely game will be won by 1 run

**RickySteve** · 06-15-09, 06:24 PM

Not a winning plan.

Sexy font though.