Get the Wizard's free newsletter!

Monthly, 100% spam-free

Betting Major League Baseball

Last update: May 4, 2009

Introduction

The analysis below is based on 19,008 games played during the 2003 to 2008 seasons. The lines were from the Stardust until 2006, and the Las Vegas Hilton, starting in 2007. In both cases, a “dime line” was used, meaning the difference between the line of the two teams was 10 points. For example, +140 and -150. For the Stardust years, after lines of +165/-175, the breakage went to 15 cents (+170/-185). For the Hilton years, after lines of +180/-190, the breakage went right to 20 cents (+185/-205).

Totals

A good place to start is looking at totals. I recall before the 2007 season a friend told me that overs would be a good bet that season, because steroid use was on the rise. The following table shows the average number of runs from 2003 to 2008. While 2007 was higher than the average for the period, it could easily be a random blip. Then again, maybe the steroids made the pitchers stronger too.

Average Runs in MLB
Year Avg. Runs
2004 9.54
2005 9.24
2006 9.69
2007 9.63
2008 9.27
Total 9.46

What is more important is knowing which is the better bet, the over or under. In the 19,008 games in the sample, the over won 46.71% of the time, the under 48.33%, and the actual total fell on the line 4.96%. Flat betting every game, laying 110, which is standard, you would have lost 5.87% on overs, and 2.77% on unders. Laying 105, you would have lost 2.82% on overs, and 0.68% on unders. It does not surprise me that the unders did better. In any sport, square bettors tend to favor betting the over, creating value the other way.

I should give the caveat that I assumed the bettor laid the same price on every game. In reality, one way or the other is usually favored, resulting in unbalanced lines on the total, for example under +110 and over -130. My data did not show the specific lines on each side of the total. However, I assume that over 19,000 games the imbalances would average out.

Money Lines

Over the 2003-2008 period studied, home teams won 54.5% of the time. They also scored 0.16 more runs, despite often not finishing the last inning. If you had bet every road team over the five-year period, betting one unit when getting odds, and betting to win one unit when laying odds, you would have lost 3.09% of money bet. Betting the home teams, you would have done much better, losing only 0.96%.

Not surprisingly, underdogs did better than favorites. It is no secret that the square public likes betting favorites in any sport, creating value the other way. Overall, the house edge on favorites (laying 106 or more) was 2.48%. On "pick" games (laying 105) the house edge was 2.38%. On underdogs (laying 104 or less, or getting odds), the house edge was a low 1.04%.

That begs the question; if home teams and underdogs are only slightly bad bets, what if you combined them, and looked at just home underdogs. The following table shows all six ways of crossing the road vs. home team and underdog, pick, or favorite, as well as the larger groupings.

House Edge on Money Line Bets
Bet Sample Size Exp. Value
All road 19008 -3.09%
All home 19008 -0.96%
All underdog 18605 -1.04%
All pick 806 -2.38%
All favorite 18605 -2.48%
Road underdog 12580 -2.07%
Road pick 403 -10.86%
Road favorite 6025 -4.25%
Home underdog 6025 1.10%
Home pick 403 6.10%
Home favorite 12580 -1.74%
All 38016 -1.92%

I would not read much into the 6.10% advantage on home picks, due to the small sample size. However, the home underdogs show an advantage of 1.10% over 6025 bets. The standard deviation on that average is about 1.5%, so it could just be that the home underdogs were lucky.

Run Lines

Unfortunately, I don't have any data on historical run lines. However, I have done analysis on what a fair run line should be, according to the money line, and whether it is on a road or home team. What I found is that the run lines are pretty well set, at least at the off-shore sports book Pinnacle. Sometimes you will see what looks like a good bet laying the 1.5 runs, but it always seems to be on a game with a low total. A low-scoring game is more likely to end in a margin of victory of 1, which makes a -1.5 spread lose on both sides, and +1.5 win on both. The following table shows the probability of a margin of victory of 1, according to the total.

Probability of MOV of 1
Total MOV=1 Probability
6.5 27.27%
7.0 31.66%
7.5 33.25%
8.0 27.18%
8.5 28.23%
9.0 27.47%
9.5 26.56%
10.0 26.46%
10.5 26.18%
11.0 25.78%
11.5 23.17%
12.0 17.48%
12.5 25.00%
Total 27.57%

Conclusion

To conclude, here is my basic strategy for betting baseball:

  1. Bet 10-cent money lines only. A list of who offers 10-cent lines in Vegas can be found in my Sports Betting Appdendix 1
  2. Between unders and overs, bet unders.
  3. Between road and home teams, bet home teams.
  4. Between underdogs and favorites, bet underdogs.
  5. Better yet, bet home underdogs.

©1998-2009 Wizard Of Odds Consulting, Inc. All rights reserved.   Privacy/Terms   Contact   Advertise   About Us   Links

The Wizard's other sites: Wizard of Vegas, Wizard of Macau, Math Problems The Wizard recommends: The Bear Growls,   Casinomeister, Online Casino Suite