A Under, B Over, and then A>B, I love the Super Bowl

[Prop Man]

There isn't much profit in these three different bets, but I still have a lot of fun making them

Hayes Under 5 tackles +120
Harrison Over 6 tackles +110
Hayes +1 tackle vs Harrison +100

all at the same book (but lines have now moved).

[admin]

Interesting mental exercise:

Is it possible for this group of bets to be -EV?

[The Chaperone]

Of course it's possible. If the actual predicted number of tackles was something like 1 for Hayes and 20 for Harrison for example, you would pretty much go 1-2 every time Harrison doesn't suffer an early injury...

[Prop Man]

I think Chap has it reversed, but I get his gist.

[The Chaperone]

Yeah I screwed something up there, but yeah, while it is unlikely, there are definitely scenarios where this combo is -EV.

[PerpetualCzech]

What if your betsize on the +1 prop was 2 units and the others were 1 unit?

[Prop Man]

I wonder what the exact relative betsize is that reduces variance the most. Kim Lee?

[ComptrBob]

Yeah I screwed something up there, but yeah, while it is unlikely, there are definitely scenarios where this combo is -EV.

You did screw the example up, but you had the right idea. The idea is to split the first two props, but have Harrison with more than one tackle than Hayes.

E.g.
If true average is Harrison 7, Hayes 5.5, then you go 1-2(LWL).
If true average is Harrison 5, Hayes 3.5, then you go 1-2(WLL).

The more difficult question to answer is:

If all 3 props are fair lines (assuming say a 20 cent line), are there any probability distributions for the two player's tackles that result in an overall -EV for equal size bets?

[ComptrBob]

What if your betsize on the +1 prop was 2 units and the others were 1 unit?

From my examples above, if the two OV/UN are "mispriced" as stated, clearly weighing the +1 prop as 2 units is a mistake.