How much of an advantage is this correlated parlay?

**Bluehorseshoe** · 08-31-10, 06:32 PM

It's getting brutal in here.

**George7904** · 08-31-10, 06:39 PM

Why is it getting brutal? What do you mean by that? Justin says (in previous posts) that anything above a 38.5% ratio is advantageous. This is over 70%. Am I misinterpreting your comment? This out is a clueless local.

**IrishTim** · 08-31-10, 06:49 PM

What percent of the total is the spread?

**George7904** · 08-31-10, 06:54 PM

72%. 37.5/52

**Justin7** · 08-31-10, 07:39 PM

guessing quickly, the two good boxes (+37.5/und 52 and -37.5/ov52) will each hit maybe 34%. The dog will do slightly better.

If those numbers are good (and I didn't do a DB search to guess on those) and you get 2.6:1 on 2 teamers,
EV = 0.34 * 2.6 - 0.66 = 0.224, or 22.4%.

**George7904** · 08-31-10, 08:18 PM

Sorry, to ask another question (I'm not being lazy, I just don't know where to start with the math). How did you come up with the 34% hit ratio?

**Justin7** · 08-31-10, 08:29 PM

Originally posted by George7904

Sorry, to ask another question (I'm not being lazy, I just don't know where to start with the math). How did you come up with the 34% hit ratio?

It was a guess.

If you want a better guess, look at a database. Compare how often a college game with a total to line ratio of 1.5 (or 1.4 to 1.6) hits dog/under, and fav/over.

**IrishTim** · 09-01-10, 05:42 AM

Originally posted by George7904

72%. 37.5/52

Correct, now you need to run a database analysis as Justin said to generate an expected winning percentages for the parlay. From there, calculating your EV is a trivial task.

**Bill the cop** · 09-01-10, 08:55 AM

Originally posted by IrishTim

Correct, now you need to run a database analysis as Justin said to generate an expected winning percentages for the parlay. From there, calculating your EV is a trivial task.

Here ya go. With a LTR of 1.39 for the full game the F/O covers 34.6% of the time, the D/U covers at a 29.5% rate (using a LTR range of 1.3 to 1.5).

**Flight** · 09-01-10, 09:24 AM

Bill, did you use a historical game database to get your numbers? If so, what league and what years did you query?

Thanks.

**Bill the cop** · 09-01-10, 09:31 AM

2000-2009, database of 6338 college games with lines posted on Don Best.

**Flight** · 09-01-10, 02:16 PM

Nice database, wish I had Don Best's closing lines.

I'm surprised to see F/O cover more than D/U. In the query that I ran, D/U was around 38% and F/O was around 26%. (Only looked at 2008 and 2009 NCAAFB though).

**dj_destroyer** · 09-29-19, 10:59 AM

Originally posted by Bill the cop

2000-2009, database of 6338 college games with lines posted on Don Best.

Sorry for bumping super old thread but anyway this database has been kept updated? I have a local who doesn't have a clue about correlation. I'm wondering how you and Flight could have such different numbers. Wondering what the real ratio is.

**Believe_EMT** · 10-07-19, 08:19 AM

looks like Filght only used 2 years of data which suggests a much smaller data set and could present wildly different results. might only cover a handful of games