Prop Question

**TPowell** · 11-11-10, 01:28 PM

Not worried about the price right now, just worried about which formula is mroe accurate

**rm18** · 11-11-10, 02:10 PM

just bet Roddy White, common sense works a lot better.

**TPowell** · 11-11-10, 02:35 PM

they have the same amount of TD's on the year, Atlanta's pass defense is probably top 5 worst in NFL, and White isn't at 100% though

**rm18** · 11-11-10, 02:42 PM

Not saying it has a huge edge but I am very sure White has over a 50% chance, them having the same amount of TDs is not that statistically significant White is the better player, and Baltimore is more stout against the goalline run.

**TPowell** · 11-11-10, 02:45 PM

Atlanta is very stout against the run period and is TERRIBLE against the pass. Anybody evaluate the math part for me?

**Indecent** · 11-11-10, 04:11 PM

Originally posted by TPowell

Anybody evaluate the math part for me?

The first solution looks pretty straight forward, nothing that sticks out to me.

I'm not convinced that dividing the player td catches by quarterback td passes will provide you anything you can work with, at least in the context of this problem. It's certainly possible to normalize using something other than the expected points, but I'm not sure this is the right approach.

**TPowell** · 11-11-10, 04:15 PM

by straightforward do you mean effective?

**donjuan** · 11-11-10, 04:27 PM

I'd go with the first one, but you may want to use more years of data (assuming no major changes in schemes, players, etc.), particularly with something like TDs.

**donjuan** · 11-11-10, 04:28 PM

Also that number is going to get you his expected TDs for the game, not how often he scores a TD before Boldin.

**TPowell** · 11-11-10, 04:30 PM

Originally posted by donjuan

Also that number is going to get you his expected TDs for the game, not how often he scores a TD before Boldin.

that was my other big question, are these type props crapshoots timing wise? Just because a guy is 66% to another guy's 35% doesn't mean the first guy is the much better bet.

**saratoga1927** · 11-11-10, 06:04 PM

what's the line on neither scoring a TD?

**TPowell** · 11-11-10, 06:13 PM

no line on neither. If neither scores, it's a push

**Flight** · 11-11-10, 06:33 PM

I use the first approach all the time, albeit only for Yes/No will he score a TD. Since it's against Boldin, you will need to measure against him somehow (probably using the same approach). And then I'm not sure how you would handle the "who scores first" part of the prop, perhaps just discard it and compare directly each players probability of scoring at all.

I recommend only using data from this season. I would also look to see if anyone scored 2 TDs in a game and only count it as 1.

In your second method you take 8/8=1 and then discard it.... were you planning on using that for something?

**TPowell** · 11-11-10, 10:39 PM

Originally posted by Flight

I use the first approach all the time, albeit only for Yes/No will he score a TD. Since it's against Boldin, you will need to measure against him somehow (probably using the same approach). And then I'm not sure how you would handle the "who scores first" part of the prop, perhaps just discard it and compare directly each players probability of scoring at all.

I recommend only using data from this season. I would also look to see if anyone scored 2 TDs in a game and only count it as 1.

In your second method you take 8/8=1 and then discard it.... were you planning on using that for something?

the 8/8 was just to show the formula. Still not sure how to price these type things, I actually like the 2nd way best because it shows how often the QB throws him the ball TD wise and when you combine that with defensive ability I think it's worthwhile. It puts things in a broader prespective

**That Foreign Guy** · 11-12-10, 11:04 AM

This can be reworded "Given a TD is scored by one of these two players, what are the odds it is from [chosen player]"

Unless TD scorer is not evenly distributed (early TDs are more likely to be from passes or similar) then [white expected TDs] / ([white expected TDs] + [boldin expected TDs]) gives you the % of the time white scores first of these two as it gives you the probability that any random TD scored by one of these two players is scored by White.

How you calculate the expected TDs is obviously the tricky part.

Is this a common prop? I've spent very little time looking at NFL markets and haven't seen anything similar for soccer.

**luegofuego** · 11-12-10, 12:09 PM

There's goalscorer match-up props available sometimes. Sadly at square books usually, and only for the biggest games.

**rm18** · 11-12-10, 05:04 PM

White was the play but Boldin won just how gambling goes sometimes

**That Foreign Guy** · 11-13-10, 09:23 AM

Originally posted by luegofuego

There's goalscorer match-up props available sometimes. Sadly at square books usually, and only for the biggest games.

I'll have to have a look for them. There's more data points for soccer and I am more familiar with the stats.

**TPowell** · 11-15-10, 12:37 AM

used #1 to cash Adrian Peterson and Arian Foster TD props today