Super Bowl Prop: Is this one +EV

[xyz]

Just saw this one on 5dimes:

PIT +8½ & game under 51½ +109

This is effectively a 6 point teaser. The +8.5 crosses the 3 and the 7. The under 51.5 is slightly better than 6 points, since some books have the total at 45. At +109, it seems like a +EV bet. What do you think? I remember last year Justin posted some prop picks for the Super Bowl. Can we get that again this year please? Thanks.

[mminkovski]

Yes, I like this teaser. But how did you get these at +109?

[rm18]

+EV

[LT Profits]

Yes, I like this teaser. But how did you get these at +109?

He said 5 Dimes, so they must have it under props.

[Justin7]

If 2.5 is the fair price, that leg should cover about 74% as a BS teaser.

Total... teasing totals is usually a terrible idea. Without the help of a push chart, I'll guess that the total leg will cover 63% of the time.

Odds of winning = 0.74 * 0.63 = 0.4662, so fair price is about +114. Looks -EV.

[ccy8888]

pit all the way no need to tease or buy point

[xyz]

If 2.5 is the fair price, that leg should cover about 74% as a BS teaser.

Total... teasing totals is usually a terrible idea. Without the help of a push chart, I'll guess that the total leg will cover 63% of the time.

Odds of winning = 0.74 * 0.63 = 0.4662, so fair price is about +114. Looks -EV.

Good thing I didn't put too much on it. I see that the problem was I didn't know how to value the 6 points in the total.

[rm18]

Good thing I didn't put too much on it. I see that the problem was I didn't know how to value the 6 points in the total.

well first you are getting 6.5 points not 6, and also I think 45 is too high. I would definitely bet under 51.5 at -200 and even a little higher.

[blackbart]

five dimes is very knowledgeable about teasers, so i would conclude it is not a good deal. However it sure looks tempting, which is the idea i guess

[HauntingTheHoly]

I'm really curious as to HOW people cap props. Do they use a database on the Internet somewhere?

Justin7 last year made a super bowl prop video saying that the "No" on overtime was fair priced at -1100, if I remember correctly. Does this mean that I should bet "Yes" if I can get it at +1000? +800? Or maybe on the day of the game the "No" could be bet all the way down well below -1100?

Another interesting Prop that would require capping - which I feel unable to do - "Will the game be decided by exactly three points? +375 for yes, -550 for No. What should I DO to begin understanding where this line is relative to the fair price?

[bztips]

I'm really curious as to HOW people cap props. Do they use a database on the Internet somewhere?

Justin7 last year made a super bowl prop video saying that the "No" on overtime was fair priced at -1100, if I remember correctly. Does this mean that I should bet "Yes" if I can get it at +1000? +800? Or maybe on the day of the game the "No" could be bet all the way down well below -1100?

Another interesting Prop that would require capping - which I feel unable to do - "Will the game be decided by exactly three points? +375 for yes, -550 for No. What should I DO to begin understanding where this line is relative to the fair price?

You need an NFL push rate chart. For example, Justin's book has a push chart showing that 9.5% of NFL games end with a spread of exactly 3. That corresponds to a fair no-vig line of about 952, so the +375 is hugely -EV, while the -550 is +EV. But before you go out and bet the house on No, you might want a push chart that also accounts for the total.

[HauntingTheHoly]

You need an NFL push rate chart. For example, Justin's book has a push chart showing that 9.5% of NFL games end with a spread of exactly 3. That corresponds to a fair no-vig line of about 952, so the +375 is hugely -EV, while the -550 is +EV. But before you go out and bet the house on No, you might want a push chart that also accounts for the total.

Hey, thanks for that info. I need to look in to push charts, then. How do you obtain your push charts? And did the chart convert 9.5% to 952 for you, or can you calculate it another way? At first glance I thought I could move the decimal point two spots to the right (950 roughly = 952) but that's clearly not right, eh? lol. Shows where I'm at.

Lastly: When I see a prop that I want to look into, how do I know what kind of chart to search for? For example I see "Will there be a special teams or defensive TD scored in the game?" +145 for Yes -175 for No.

[Flight]

"Will there be a special teams or defensive TD scored in the game?" +145 for Yes -175 for No.

Several ways to skin that cat

Most obvious is to look at all NFL games this season and count the number of games there was a special teams or defensive TD and divide my number of games sampled. This gives you a percentage which you can convert to a moneyline.

Or you could narrow your error by selecting only games played by Pittsburgh and GB, but this also reduces your sample size... so you need to strike a balance between several methods for the best results.

There are errors with this approach (using the average or arithmetic mean instead of expected value to start), but this is a tutorial for starters. I'm guessing you would have trouble with even this crude method, so really no sense in driving on about advanced methods (I mean no offense, just that most people that ask about the methods rarely have the data necessary to proceed)

[Flight]

a push chart showing that 9.5% of NFL games end with a spread of exactly 3

Close, but be careful

9.5% of NFL games with a spread of 3 push

If you were to look at all NFL game scores as you say, the numbers are different and not as useful.

[Flight]

five dimes is very knowledgeable about teasers, so i would conclude it is not a good deal.

5Dimes regularly offers teasers and parlays in disguise in the props section and I have caught them several times being off by more than 15 cents. The limits are like $100-$250 so I would not consider them unbeatable. They are worth investigating, especially if time is not important.

[HauntingTheHoly]

Several ways to skin that cat

Most obvious is to look at all NFL games this season and count the number of games there was a special teams or defensive TD and divide my number of games sampled. This gives you a percentage which you can convert to a moneyline.

Or you could narrow your error by selecting only games played by Pittsburgh and GB, but this also reduces your sample size... so you need to strike a balance between several methods for the best results.

There are errors with this approach (using the average or arithmetic mean instead of expected value to start), but this is a tutorial for starters. I'm guessing you would have trouble with even this crude method, so really no sense in driving on about advanced methods (I mean no offense, just that most people that ask about the methods rarely have the data necessary to proceed)

I follow all of this quite well. I did manage to make it through Stats 1 & 2 with A's.

When you say most people asking about the methods "rarely have the data" - if you're talking about the NFL records - you are correct. But I suspect you're talking about a full understanding of how accurate (or inaccurate) particular methods may be. That part is something I can look into myself. But how can I quickly access such detailed information on NFL games throughout the season, is there a big user-friendly database somewhere on the Internets? [sic]

[bztips]

Close, but be careful

9.5% of NFL games with a spread of 3 push

If you were to look at all NFL game scores as you say, the numbers are different and not as useful.

Good point, my bad.

[HauntingTheHoly]

@bztips - it seems your point should still be good in this case, since the line is in fact 3 points, no?

[simplydusty]

@Haunting. Check out killersports.com for a public database

[wrongturn]

The Super Bowl games may need its own push chart. Considering PIT is a great defense team, I would think the prop is +EV, my 2 cents only.

[JustinBieber]

It looks like your 2 cents are actually worth 0 then. Or maybe something like 0.9 if you want the EV.

[Flight]

But how can I quickly access such detailed information on NFL games throughout the season, is there a big user-friendly database somewhere on the Internets? [sic]

Not really. You will need to build your own. At least that is what I recommend to people investigating these things. Sometimes (for props) you can just copy and paste stats from say espn into excel and do analysis there. But something like this would require more information than the box score (scoring summary from certain websites may tell you who did the score and you could maybe figure out whether it is defensive or special teams).

My personal NFL database, for example, does not contain the correct information to price a specialteams/defensive TD prop. It does contain the information to correctly price overtime.

Eh, maybe someone here will cough up a dB, but I wouldn't count on it.

[Flight]

@Haunting. Check out killersports.com for a public database

Hmm, this used to be sportsdatabase.com, but I see they have perhaps changed their stance slightly to be a more gambler oriented (probably realized there's a market for data).

Haunting, you may be able to write an SDQL for solving some of your questions. For instance I just wrote this one for OT

season>2006
count: 2134

season>2006 and overtime=1
count: 132

132/2134 = 6.2%


I don't much like their output format (I had to the SU numbers to get the count, and they have a max row of like 50 on their output) hence why I like having my own db. But there ya go, sportsdatabase.com gives you data for free.