Simple MLB divisional underdogs system - big sample size and ROI

[PunterBoss]

Hi

I'm new to this and I don't know much about baseball, however I think I have come up with a MLB system which looks promising. 

I read someone's thoughts somewhere that because teams play heaps of games against opponents in their own division, the underdogs have a better chance of winning than in non-divisional games since they get used to their opponent's playing style and can work out how to best beat them.

Step 1: let's see how divisional underdogs fare - p:division=op:division and line>-105

This gives us a 0.9% ROI from around 9500 games. This drops to a flat 0% since 2007, however it still indicates there is some bias towards the dog.

Step 2: bet the divisional underdog only if they're coming off a win (teams coming off a win are more confident and more likely to perform well in their next game) - p:division=op:division and line>-105 and p:W

OK now things start to look pretty good! 4.4% ROI from some 4300 games - a huge sample size! This drops to 2.3% from 3000 games since 2007, but still a really nice return from a solid number of bets.

Step 3: only bet the underdog if their opponent has a WP of less than 50%. My reasoning is we're getting plus odds against a team with a losing record - must be on the whole a good thing, right? p:division=op:division and line>-105 and p:W and o:WP<50

Now we have an ROI of 8.8% from 1549 bets! Since 2007 it's gone 9.0% ROI from 1046 bets. This system has been profitable every year since 2005 except for 2011 where it broke even.

Seems pretty good to me and I would love to hear what you guys think. Also, I don't think I have 'fitted' the data too much (if that's the right terminology) but can't be sure since I'm just starting out with this stuff - thoughts?

[PunterBoss]

Woops - I made a mistake in my post. It should be division=o:division, which is how I tested it initially..

So the final query is division=o:division and line>-105 and p:W and o:WP<50 giving us 1549 bets at 8.8% ROI. This seems to be exactly the same result as I had below.. hmmm. Confused!

Have I messed up or am I on the right track? Thanks!

[Jesse Larkins]

looks solid to me.  It appears to be an excellent October dog which I love

---

Date: Wed, 8 May 2013 05:34:24 -0700
From: damien....@gmail.com
To: sportsd...@googlegroups.com
Subject: Re: Simple MLB divisional underdogs system - big sample size and ROI

--
You received this message because you are subscribed to the Google Groups "SportsDataBase" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sportsdatabas...@googlegroups.com.
To post to this group, send email to sportsd...@googlegroups.com.
Visit this group at http://groups.google.com/group/sportsdatabase?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Dr M]

Punter Boss,

Welcome aboard!!  You have already made a great contribution.

FYI...

The reason p:division gives the same results as division, is that the team didn't change divisions.  That is, p:division is the division of the team in the previous game and division is the division of the team in the current game. 

Ed 

[PunterBoss]

Thanks Ed!

Yep, that makes sense. Cheers for the explanation.

[Jar Jar 21]

Great system guys and EK, it looks like adding your "better than 40-percent" parameter reduces the number of games in play (well, this year anyway) while boosting the return. 

Just a head's up... You can save yourself space by using these shortcuts: 

DIV and D and p:W and 40<o:WP< 50 and season

DIV means divisional games and D accounts for the underdog status of our play-ON team. 

Thanks for sharing, 

JS

[Paul McDonald]

Great system.  Any suggestions on how to change this to show teams that have this potential scenario in an upcoming game?

[Eser Orhan]

hi paul i don't know what your problem is. if there is a game in an upcomming schedule then it will show up with blank fields for the result and date ect. here is an example for today's MLB
http://sportsdatabase.com/mlb/query?sdql=+HD+and+tA%28W%29+%3C+.42+and+month+%3C+7+and+division+%3D+o%3Adivision&submit=S+D+Q+L+!
Diamondbacks fields are empty. So this is a potential play. Some systems only come up once the odds are defined by the oddsmaker. Others are free from that.

[Paul McDonald]

Thanks Eser.  I didn't notice that line at the bottom.  Do most sports show the potential daily play if it exists?

[Eser Orhan]

Yes they do, as long as the database is up to date.

Am Mittwoch, 8. Mai 2013 14:17:07 UTC+2 schrieb PunterBoss: