NFL Parlays Huge Spreads and O/Us 4-0 this weekend

[brendanb438]

Hey.

I was wondering if anyone else has been making these correlated parlays? For yesterday on games with a spread >or = to 25% of the total the plays of fav+ over and dog+ under went 4-0. I haven't been playing these every week but I have gone like 9-1 this season on these.

Anyone have an easy way to check the last couple seasons to see how well these parlays work? I am guessing that just like in college with the favs/overs and dogs/unders of > 33% that these hit quite a bit. Since totals are lower in the NFL I play these at the 25% mark instead.

Any input is appreciated. Looks like around 5 games might qualify for this week also.

-Brendan

[ProfessorBen]

Brendan,

how do you decide to take the fave+over or dog+under side of this? Wouldn't the dog+under on Buff have lost?

[brendanb438]

You play both fav+over and dog+under just like a lot of people had been playing on the college games.

1 unit to win 2.6 units on each side. So you either win 1.6 units or lose 2 units on these plays.

-Brendan

[brendanb438]

The 4 games this week were.

Pats/Bills -16 o/u 47
Pitt/NYJ -10 o/u 40
KC/Indy -14.5 o/u 41.5
Car/GB -10 o/u 37

Betting each fav/o and dog/u for 1 unit and you would have been +6.4 units.

2 favs+overs won and 2 dogs+unders won.

-Brendan

[kyleb]

Brendan,

I've been doing a bit of research on correlated parlays, actually. It shouldn't be too hard to put the data together from Covers and backtest it for a couple of years. Hopefully I'll find some more time to gather the data with what free time I have left. Thanks for shining a light on it.

[Thremp]

[ QUOTE ]
Brendan,

I've been doing a bit of research on correlated parlays, actually. It shouldn't be too hard to put the data together from Covers and backtest it for a couple of years. Hopefully I'll find some more time to gather the data with what free time I have left. Thanks for shining a light on it.

[/ QUOTE ]

Samples are relatively small IIRC and checking edge on these is pretty ew.

[brendanb438]

I am gonna try and get some info so I can go back 4 or 5 years to see. The games that Q in the NFL will be a small sample size which is a downside to this. I bet a few of these at the end of last season and they were hitting at a good clip also.

The only time this has been mentioned on some of the other SB forums the regs who had been hammering the College correlated parlays were being hush hush about these also and I got a couple PMs to just let it go. I figured I would let those with good NFL databases over the last few years have this info to see if anything is there to exploit.

[Jazzy3113]

I think you are just trying to find a hidden pattern that is not there.

[Rustjive]

Somewhat on-topic and amusing:

YOUR WAGER
HAS BEEN REJECTED

You have at least one combination of selections on the same contest that is highly correlated. This may include a favorite and total with a high spread and a relatively low total OR an underdog and total with a high spread and a relatively low total.

SELECTION CONTEST
NEW ENGLAND -22.0 PHILADELPHIA AT NEW ENGLAND
OVER 50.5 PHILADELPHIA AT NEW ENGLAND

[NajdorfDefense]

[ QUOTE ]
I think you are just trying to find a hidden pattern that is not there.

[/ QUOTE ]

We've had this discussion on this board at least 2x. There is no +EV there. Some books ban these parlays due to cfb games where you can have Okla -42, Total=48 and in those cases there is +EV on a correlated parlay, so they ban them all rather than waste time figuring which is which.

There is no +EV on AFC-14, Total =56. NFL is not college LDO. You can run hot over short sample size, but you're simply data-mining. Besides, what if the value is Fav/Under? Or Dog/Over? Then you're explicitly making -EV bets.

Your matrix of payouts is: +1.6, -2, -2, +1.6.
Wouldn't a 2-team "IF-bet" have higher EV anyway? If AFC-14 wins 20, then Over 56 wins 40/[38 I guess]. I haven't looked at IF-bets in years so I'm not 100% sure. I know you can do If-bets on game[s] at same time or even If on 4pm with 2nd leg on 1pm.