Originally <a href='https://www.sportsbookreview.com/forum/showthread.php?p=6421329'>posted</a> on 09/17/2010:

The advice in here is 100% dead wrong.

Yes, parlays are sucker bets in general, but, like you said, if you can find correlation, you have a +EV wager on your hands. Anytime a single variable manipulates the outcome of two events, there is correlation. In this case, if one teams offense performs abnormally well, there is good chance they cover the spread and the game goes over. Conversely, there is even stronger correlation that if the dog plays good defense and the game remains tight, the under and the dog have a good chance of winning together.

Take a look at this thread in the think tank, or try searching the think tank for correlated parlay.
http://www.sportsbookreview.com/forum/handicappe...ed-parlay.html

I have ran database queries on this exact thing several times both NFL and NCAA, and the Dog/Under bet has a high chance of winning historically, well over 30%, which beats the required 27.78% required win rate on +260 odds (of course, you need to beat 28.57% on +250). Favorite/Over has a slightly lower win rate, but from what I can tell, it still beats the +260 break even point of 27.7%.

Now, to compare the 2 team and 4 team parlays from a shopper's perspective, use the SBR parlay calculator:
http://www.sportsbookreview.com/Betting+Tool...alculator.aspx

The 2 team parlay of 2.5:1 has a 12.5% premium vs the standard 10% paid for 2.6:1
The 4 team parlay of 10:1 has a 31.25% premium vs the standard 18.75% for 12:1

Stick to the 2 teamer, because the premium is lower, and I doubt you could find a correlation between 4 simultaneous events.

Note that books charge a premium because they can often be exposed by very long odds so they do this to protect themselves (and to keep their theoretical hold very high). I know several sharp players that unload on 2 team moneyline underdog parlays at very long odds (think +700 and higher). It works well when dealing with free plays and bonus's, not necessarily cold cash.