Interception Odds?

[dmolition]

If a QB has averaged 11 int in his 14 games and
and is playing a team that averaged 22 int in their last 16 games,

how do i get the expected interception for this particular game, im having trouble normalizing.

[Justin7]

Expected INT = QB Int rate * Def int rate / League avg int rate

What is the league average?

[Thremp]

I approve the new J7 message answering "Log5" when "Log5" is the correct answer. Little worried we'd be making a power ranking, but all is well.

[Justin7]

I approve the new J7 message answering "Log5" when "Log5" is the correct answer. Little worried we'd be making a power ranking, but all is well.

Thremp,

Why don't you explain your approach? Do you have any empirical evidence that one approach works better on an interception prop?

[gotemcoach1523]

Go steelers!!!

[lemart5]

what is the league average for expected INT?

[Thremp]

Thremp, Why don't you explain your approach? Do you have any empirical evidence that one approach works better on an interception prop?

Do you understand what I said?

Serious question. If so, do you understand what you said?

Hint: They're not different.

The last sentence then remarks on how you've upgraded your knowledge since the last iteration of this style question came around.

[dmolition]

Thanks for the response,

just want to check up the math,

So if i understand correctly a QB int rate is 11 int in 400 passing attempts multiplied by 100
so its int rate would be 2.75

Same goes for defense i imagine, so lets say defense int rate is 2.6

and league average could be 2.2 (the league average for QB int rate right?)

Exp INT = (2.75*2.6)/2.2
Exp INT= 3.25

Thanks again.

[Thremp]

Justin7,

Do you have empirical evidence that Bayes Theorem is incorrect?

[Justin7]

No.

[Thremp]

Thremp, Why don't you explain your approach? Do you have any empirical evidence that one approach works better on an interception prop?

Could you point out why you view what I wrote as "my" approach? I only see one approach possibly suggested in this thread. Once explained in process and once by name. There seems to be some misunderstanding.

[dmolition]

What method would you use Thremp?

[Thremp]

dmolition,

Was I unclear at some point? My first post pretty clearly explains the methodology I'd use. And my approval of answering a question correctly in an authoritative tone, as opposed to incorrectly in an authoritative tone.

[dmolition]

dmolition,

Was I unclear at some point? My first post pretty clearly explains the methodology I'd use. And my approval of answering a question correctly in an authoritative tone, as opposed to incorrectly in an authoritative tone.

I take it from your first post you would use some kind of Log5 analysis for the expected QB int rate.

I know log5 analysis from baseball but that is all, any recommended books where i can find the basics of log5 probabilities so i can come to the conclusion on my own then?

Thanks for the help.

[Thremp]

http://en.wikipedia.org/wiki/Odds_ratio

Its from that, which is apparently a derivative of Bayes Theorem.

[dmolition]

http://en.wikipedia.org/wiki/Odds_ratio

Its from that, which is apparently a derivative of Bayes Theorem.

Thanks i'll check it out.

[sonnyhowitzer]

what a game.

[sonnyhowitzer]

steel too many turn overs odds on win 11%

[CyberSleuth]

Interesting thread, and thanks for the link.