Finally, correlated MLB parlays

The previous post indicated that there were a couple of MLB combinations of odds, location, and total parlays that could come close to being profitable if blindly followed. Following is a table of all 8 combinations of these factors.

Games	Fav-Loc-Total	Bet	Net	Ret/$
12,954	Fav-Hom-Ovr	$12,954	-$1,117	$0.914
6,282	Fav-Vis-Ovr	$6,282	$26	$1.004
12,954	Fav-Hom-Und	$12,954	-$473	$0.964
6,282	Fav-Vis-Und	$6,282	-$819	$0.870
Games	Dog-Loc-Total	Bet	Net	Ret/$
6,282	Dog-Hom-Ovr	$6,282	-$990	$0.842
12,954	Dog-Vis-Ovr	$12,954	-$723	$0.944
6,282	Dog-Hom-Und	$6,282	$253	$1.040
12,954	Dog-Vis-Und	$12,954	-$713	$0.945
76,944	All	$76,944	-$4,555	$0.941

It turns out there were two combinations that would have been profitable over the past 8 seasons if you had blindly bet $1.00 on each. Had you bet on every visitor that was a favorite and over the total, you would have returned $1.004 for every dollar wagered. The same bet on all home team dogs and under would have returned $1.040 per $1.00 wagered.

Does this make sense, or is it simply the result of data mining stumbling across anomalies in past data? There is a reasonable explanation on why this should occur. It’s because of the rules of baseball at the end of a game. If the home team is ahead after the top of the 9^th, then they don’t bat in the bottom. Hence there will be one less inning for them to score and slightly fewer total runs. If the game goes into extra innings, it ends when the home team gets ahead. Meanwhile in extra innings, the visitors can score an unlimited number of runs. These two combines to give us the results demonstrated above.

So, should you blindly wager on every home dog and under parlay? I wouldn’t. Here’s why, a chart of returns on these wagers per year.

Games	Year	Bet	Net	Ret/$
769	2010	$769	108.3475	$1.141
747	2011	$747	33.4439	$1.045
816	2012	$816	50.6865	$1.062
812	2013	$812	28.5946	$1.035
740	2014	$740	-1.4718	$0.998
770	2015	$770	-19.3694	$0.975
835	2016	$835	9.6334	$1.012
793	2017	$793	43.2103	$1.054
6,282	ALL	$6,282	$253	$1.040

Some years would have been quite profitable, others not. In fact, had you started in 2014, you would have suffered through two long losing seasons. Too much variability for my tastes.

However, these results do factor in the system that I currently use for wagering on MLB. There is a nice post in the “Handicapper Think Tank” forum on sportsbookreview.com. It’s in the topic “Basketball Math - Off/Def - Net Efficiency”. The post was made by a new comer named “Larkman” on June 4^th.  It provides a couple of Wikipedia links worth reading.

My next post will be on answering a question about hedging raised in the “Handicapper Think Tank” forum. I will indicate on Twitter, @ole44bill when I post this.