MyTurn's week 5 CFB (9/28-9/30)

[MyTurn2Raise]

I wish I knew...I expected this line to be near two touchdowns. I've tried to figure out why it only moved from an opening of -5 to -7.

One reason is the 20-5-3 ATS blah blah blah stuff. Personally, that kind of stat usually makes me vomit in my mouth a little. While an interseting sidenote, my opinion is that those numbers are worth very little. They try to show that there is a systematic bias that underrates a team in the public's eye. They also suppose that this underrating will continue.

I don't buy it. Honestly, if lines are set so that they are accurate and have a 50/50 chance of ending up on either side, you'll see some teams go on odd streaks. Flip a coin a few hundred times and see how random heads and tails really are.[/rant]

Real Reasons to choose Iowa:
-they are undefeated and ranked #14
-Iowa is playing in a good home environment
-tOSU was unable to really establish the run last week and Iowa has a very solid defensive line
-Drew Tate doesn't usually make many turnovers
-tOSU's achilles heel has been stopping a straightforward rush attack

My response:
-big deal: beating Montana, Iowa St, Syracuse, and Illinois doesn't impress me
-tOSU is ready for a tough road environment more than any other team
-tOSU's supremacy out on the corners will cause the Iowa defense to adjust that way. Then, tOSU can blow up the line and mediocre linebackers.
-Iowa lacks big playmakers on offense. Tate hasn't been able to throw downfield since his injury either. He's going to face more pressure than he has all year; it's always interesting to see how a qb will respond to the first few head knocks of the season.
-Iowa lacks the straightforward rush attack to expose the Buckeyes weakness. This hurts doubly so since the Tate offense thrives off play action. Play action only works with the defense bites...uh oh.


So, I'm still searching for game analysis that shows why the Sqawkeyes will win. Perhaps, I'll venture over to fantasyland (ie the Hawkeyes fanboards) to see what their deluded fans think.

[MyTurn2Raise]

oh man...I went to hawkeyenation.com and I'm a bit more dead inside for it

some potential tOSU issues

How Iowa will win [img]/images/graemlins/confused.gif[/img]

you can go there and check it out for yourself

the only thing I saw that concerned me is that they correctly identify Tressel as being quite conservative with a lead and on the road at times

[Jibba]

[ QUOTE ]
Auburn is -13.5 on Mansion right now. -107 pricing in the exchange. If you like it this may be a good time to grab it.

[/ QUOTE ]

Thanks. I did just that.

[MyTurn2Raise]

[ QUOTE ]
I really almost feel like Michigan State falls under the Mike Tice rule. This team is impossible to figure out on a week-to-week basis. I feel comfortable fading Illinois, but man does it feel uncomfortable to bet on Mich. State after what had to be one of the worst choke jobs I have ever seen.

I agree with you on tOSU to cover @ Iowa. I figured this line would be around 12, and I may have bet it at that.

I've been to the last two Purdue games, and I attended all of the home games last year. I can tell you that the pass D is better than last year, but still not very good. The corners are bigger and shouldn't be dragged for ten yards by Samardzija (sp?) every play. Quinn had over 400 yards in the game last year, mostly on quick passes. I think Purdue covers even on the road, but I am an alum so take it for what it's worth. If you recall, Purdue went three years only losing by more than 7 once until last year.

[/ QUOTE ]

yeah...Michigan State and Arizona State are my favorite Jekyll and Hyde teams. I probably would've stayed away from the Spartans, but I always pick the Illini and have done quite well on their games over the last decade. One thing that concerned me about Sparty was John L's press conference today....that guy is a tool. He is still fixated on the ND game. He always looks like a defeated man....bah.

As far as Purdue, I'm sure ND will torch their defense. The corners are too young and too inexperienced. Quinn won't miss all the open tosses that Cupito did. On the other hand, I expect Purdue to have some success against the ND defense. How much is uncertain. Purdue's playcalling looks more like the Boilers from 4-7 years ago instead of the team from last year. It wouldn't surprise me to see the Boilers go off. However, I really think it comes down to whether the Boilers protect the ball and if they are able to get a takeaway.

[iggymcfly]

I think it's more about the preseason hype than anything. Before the season, when the talking heads were going through the schedule, they were all "watch out for Iowa". "Iowa's going to be a tough game for Ohio State" and now that the season's gone for a month, Ohio State looks tremendous and Iowa looks awful, people haven't really adjusted.

Also, the home advantage is worth something. Forgetting ATS records, Iowa's 20-1 SU at home since the start of 2003 with the only loss being a three-point game against Michigan last year. Also, the Hawkeyes tend to play pretty sloppy in non-conference every year, before improving in Big Ten play. Last year, they had the 23-3 loss to Iowa State on a team that went 7-5 and the year before, there was the 44-7 loss to Arizona State on a team that went 10-2.

Still though, the Buckeyes aren't the kind of team that a good to mediocre team can beat on emotion, and the Hawkeyes have too many holes to stay close. This year's team doesn't have the talent on the offensive line or in the secondary to not get completely exploited by the Buckeyes, and even if Iowa doesn't lose another home game this year, I still agree that they should lose by three possessions to the Buckeyes.

[ML4L]

[ QUOTE ]
I wish I knew...I expected this line to be near two touchdowns. I've tried to figure out why it only moved from an opening of -5 to -7.

[/ QUOTE ]

This post got out of hand, so skip to the last paragraph for the short version...

Your posts regarding this subject seem to indicate a philosophy that is somewhat similar to what mine has traditionally been. When you see a match-up that you really like, you incorporate those things into what you think the line "should" be. This also manifests itself in the form of "predictions" of what the score "should" be, similar to what iggymcfly seems to be fond of doing.

The reality is that (and posters who are in the know, feel free to correct me if I am way off-base here) the lines generally do not incorporate all but the most blatantly obvious match-up issues. Most lines are a product of a "power rating" context. That is, tOSU is probably the best team in the country right now, or certainly in the top 3, generating a rating of X. Meanwhile, Iowa is perceived to be in the top 15, generating a rating of Y. If you look at how many points an elite team is generally better than a very good team, the answer is around 8-10. Throw in 3 for homefield, and you have your line. I'm certain that it gets a little bit more involved than that, particularly on big games that are sure to have a lot of action, but it's pretty easy to get close to 95% of lines by using that methodology strictly.

Some people handicap this way. They worry more about "value" from the line than anything that they expect to happen on the field. If you are good at this method of handicapping, you will hit 55% and grind out a profit.

My approach has always been that, if you want to beat sports betting, you have to find games where specific player vs. player and unit vs. unit match-ups are going to cause the power rating approach to be inaccurate. It is my personal opinion that this approach will allow you to find games where the line is "off" by a touchdown or more. That is, if Iowa were hosting tOSU, USC, and Auburn, they would probably be a 5-7 point dog in each game. But, hypothetically, maybe they should be a 13 point dog to tOSU and a pk with Auburn...? It is also my personal opinion that a handicapper who is good at identifying spots where specific match-ups skew a line greatly can hit 60% over the long-term.

But, the key is that, if you choose to take the "match-up" approach, you have to be aware that "value" is what drives the line in most cases. From a value perspective, the line for this game is right on the money or, if anything, skewed toward tOSU. Hence, people who do not share your opinion about the match-up issues don't mind taking Iowa here.

So, if you choose to take a "match-up" approach, you have to look at lines in two ways:

1) What do I think that the line should be in the "value" context, since that will help me get the best price on the game and predict line moves?

2) Do I think that this is a game where specific attributes about a player/team will have a major bearing on how the game plays out?

This manifests itself along the lines of: I think the line will be -7, but I really like the side of Team X. The first part is a value perspective, the second part is the match-up perspective.

In the past, I have been fairly oblivious to part 1. I would guess that a line was going to come at -3, it would actually come -7, and I'd play it anyway because I thought that the match-ups dictated a line of -10. And, this approach has worked well for me for the past two seasons.

But, then I started to get more involved in the "math" side of sports betting. Scalping, middling, the way that books work, etc. It seemed as though nobody else on this forum thought about handicapping the way that I did. So, this year, I decided that I would be more committed to incorporating "value" into my handicapping, and I've been amazed at how much "sharper" it makes me feel. I used to be AWFUL at getting good numbers on games; now, I feel like I am usually on the right side of line moves...

The ironic thing is that, the more focused that I've been on value, the worse my picks have looked... But, as I mentioned in my own thread this week, I think things are going to start coming around soon.

Anyway, sorry for the semi-hijack. Here is the synopsis of what I was trying to say:

Given the way that you handicap, you are likely going to be frequently surprised by lines and line moves. But, if you are good at what you are doing, my opinion is that you will continue to be a winner, probably a bigger winner than someone who is better at pricing games but who knows less about football, even though you are not always getting the best of the line.

ML4L

[BobJoeJim]

[ QUOTE ]
Look for an extra 25,000 to be watching this game outside of Kinnick on the video screens that the university and ESPN will be bringing in later this week!
With an estimated crowd of over 100,000 within a 50 foot radius of the stadium - Home-field advantage will be re-defined!

[/ QUOTE ]
*vomits*

[lerxst337]

[ QUOTE ]
oh man...I went to hawkeyenation.com and I'm a bit more dead inside for it

some potential tOSU issues

How Iowa will win [img]/images/graemlins/confused.gif[/img]

you can go there and check it out for yourself

the only thing I saw that concerned me is that they correctly identify Tressel as being quite conservative with a lead and on the road at times

[/ QUOTE ]

Thanks for the links. You are right about Tressel. We Buckeyes often joke around about how vanilla he loves to play on D, particularly with the lead, and how the punt is his favorite play in the book. Those aren't as funny when you're talking about the line. Despite the fact that they have covered every game this year, I would NEVER give more than 3 touchdowns with tOSU. He never runs up the score, always gets in the backups, etc. This team is a little different though..., the offense is good enough that against bad teams they can pretty much accicentally score points, even with the 3rd string QB in (Boeckman is pretty good!).

All of this is pretty much moot against Iowa. I might be concerned if the line was 10 or 11, but Tressel doesn't shut down with a one score margin. He'll play field position to death with a touchdown lead inside 10 minutes, but not with anything less than that. The team is good enough that even in that scenario, we can back door cover (see the Fiesta Bowl last year.).

As an aside, I am not convinced that we can be exposed up the middle on the run. We are, otoh, quite vulnerable to misdirection, particularly early in games. Tressel plays quite boring on D to start. Cover 2 shell, man under. He also encourages his linebackers to be aggressive. Thus, we give up alot of yards on screens and counters. Texas exposed this in the first half against us. All that said, these kids were starting aren't kids. Tressel constantly rotates his defense, so to say that we only have 2 returning starters this year, while factually accurate, doesn't tell the whole story. Most of this year's team has plenty of (quality) experience, and the results thus far have shown it--against some pretty quality teams, no less.

[MyTurn2Raise]

I understand that ML4L and figured as much.

My issue is that power rankings, etc (such as sagarin) are very inaccurate at the tail ends (with both the top and bottom teams). Hence, their utility for measuring match-up involving the top 5 teams are way off. also, the Sagarin predictor for example still gives tOSU a 9 point edge with Iowa's homefield factored. Furthermore, power ratings are pretty crappy in the first few weeks of the football season. The best team Iowa has played is #48 according to sagarin. Iowa's ranking is a big unknown. Are 'value' betters likely to discount the rating when there is so much error behind the number? Do value betters look at the likely distribution of what the 'true' power ranking difference might be?

[dankhank]

[ QUOTE ]


The reality is that (and posters who are in the know, feel free to correct me if I am way off-base here) the lines generally do not incorporate all but the most blatantly obvious match-up issues...

[/ QUOTE ]
[ QUOTE ]

Some people handicap this way. They worry more about "value" from the line than anything that they expect to happen on the field. If you are good at this method of handicapping, you will hit 55% and grind out a profit.

[/ QUOTE ]
[ QUOTE ]

My approach has always been that, if you want to beat sports betting, you have to find games where specific player vs. player and unit vs. unit match-ups are going to cause the power rating approach to be inaccurate.

[/ QUOTE ]

a few rebuttal points. suffice to say i disagree, but it is too hard to organize these ideas into a coherent counter theory:

-sports gambling markets are very efficient historically. year after year you see favs/dogs hitting around 50% each, and over/unders doing the same. furthermore, it is widely accepted that the closing line is a more accurate number than the opener.

-it's hard to say how much "knowable info" is built into a line. honestly, how can you know that only the most blatant info is included? because you see a few bad lines every sunday? it seems impossible to even begin to decipher what is and isn't in a line, because there are so many levels of thinking the bookmaker could be operating at. so i find it easier to assume that everything but amorphous analysis of team ability (as opposed to clear analysis of ability such as that indy has a strong passing offense but is weaker defensively, which anyone can evaluate accurately) is included by the books. also breaking news or inside information is often not included in the line. but i think by giving the book's a lot of credit, it makes the remaining analysis clearer to arrive at, than if you assume the opposite.

-i don't really handicap the games, i let the market handicap for me. so last sunday afternoon when i saw GB drop to +6.5 everywhere, i decided the true line was probably 6.5, and grabbed +7 at the one book that still had some in that ten minute gap, matchbook. never knew the line moved because a lions CB was out. never had to know. if i had known, then i would've spent time wondering if the books had moved on this info already (or not moved), and then further time figuring out how much a missing corner is worth.

or when dave cokin states a lean on his sunday night show and the juice jumps at pinny, but i see a slower shop has a half point better line, i bang it. sure, i can't say confidently i am making a 60% play and i might not even be able to name a single player on the team, but i am pretty sure i'm in the 51-52% range (pinny tells me so), and i might even be 54-55% if cokin's analysis is above average on it.

-in my poker endeavors i have found that the clearest, most repeatable move is the best choice for me long term.

-the term 'value handicapping' is misleading. i am not thinking of getting a good price on something i have no opinion on. i want the right side, it's just sometimes i am using a half point, or ten cents of juice, as my optimal method of finding it.