LOL you're right, I fail at elementary math for a moment there.. Still taking 38. I was thinking 28/20, 28 would be way too low for Michigan lol And yes, usually the Over/Under is divided by the total points. So like tonight's game, the line is -13.5, the Over/Under is 68 so they would post 40.5 for Oklahoma State and 27 for Texas Tech.
Yeah that's what I figured. I was going to say if the Michigan State over-under really was set at 28 despite a 48 total and 27 point favorite, then there's a heavy arbitrage to be had there. So you'll be going with the Over 38 on Michigan State? Or does the actual number dissuade you from it, heh.
Would love to hear your thoughts on the betting pool idea I outlined here: http://www.thephins.com/forums/show...tting-Thread&p=2420618&viewfull=1#post2420618
38 is fine with me, I think Michigan State scores half the points from their defense and special teams alone to cover it.
I will gladly stay away from Michigan State in that game. I'm partial to Wyoming because I followed their staff quite a bit at North Dakota State and had a TON of respect for the way they coached their players on both sides of the ball. It was no coincidence they won three straight FCS National Championships there. Bohl's NDSU teams were a dominant force in the Great West Conference back before any teams were ever selected from it to compete in the FCS Playoffs. He guided his team through a move from the Great West to the Missouri Valley Football Conference which was much more prestigious, and had NDSU in the FCS Quarterfinals by their third year in the new conference. He had them winning the FCS Championship by their fourth year in the new conference, and then again in their fifth year, and again in their sixth year...and I believe they beat every single FBS team they played in those three years. Could be wrong but I believe that's true. Culminated in their beating Kansas State last year. So I've been eagerly anticipating what Craig Bohl and his staff would do at Wyoming and I haven't really been disappointed. They're 3-1 already. I mean, yeah they got Oregon'd...but then, so did Michigan State. It's going to take a while for Bohl to get that program straightened out, don't get me wrong. But I'm not inclined to bet on anyone but Oregon scoring 38+ points on Bohl's defense. Defense allowed 3.9 yards per play to Montana, 4.5 yards per play to Air Force and 4.4 yards per play to Florida Atlantic. I understand none of those teams are good but, that's good defense.
Actually letting the public drive a line up or down in your favor is a good thing. Most sharps pound lines right before kickoff. Unless you spot a fantastic number early ofcourse....all circumstancual.
I was at the gym when I saw the score 14-7, then got home and the touchdowns started rolling for Oklahoma State. 12-3 since last Saturday.
I've checked on the results of the system I discussed which utilizes signals from the users who pick from a large pool at OfficeFootballPool.com. Currently, this system is at 27-22 (0.551) on all games through tonight. However, signal strength needs to be explained. The signal ranges theoretically from 0% to a maximum of 200%. However in all practicality, the maximum signal achieved this season thus far was a 52.3% (Week 1: IND over DEN -7.5) and median signals have been 17.3% thus far. Signals above 35% have a 3-1 (0.750) record Signals above 30% have a 7-2 (0.778) record Signals above 25% have a 12-2 (0.857) record Signals above 20% have a 16-5 (0.762) record Signals above 15% have a 18-10 (0.643) record Signals above 10% have a 22-13 (0.629) record All signals in total have a 27-22 (0.551) record You can see why I have been so intrigued with this idea. And yes, I have verified that these results are actionable. In other words, I see the pool-wide picks in time to paste them into a spreadsheet I've developed which produces an easy to read bet sheet which I can then feed into Bovada (if I were into that sort of thing) and take action on. Admittedly, it's a very tight window. But still. There's nothing theoretical about this.
The second pool I work off is a private pool on CBS Sports. I have some level of historical success data on 329 of the players in this pool, some going as far back as 2009. Currently, this system is also at 27-22 (0.551) on all games through tonight. This signal works a little bit differently simply due to the nature of the pick-weight system. Instead of the signal being a theoretical 0% to 200%, this signal is a theoretical 0% to 100%. However, maximum signal was 2.98% and median is 0.54%. Signals above 2.5% have a 2-1 (0.667) record Signals above 2.0% have a 3-3 (0.500) record Signals above 1.5% have a 7-4 (0.636) record Signals above 1.0% have a 10-8 (0.556) record Signals above 0.5% have a 17-10 (0.630) record All signals in total have a 27-22 (0.551) record Placing action on this system is slightly more cumbersome. You have a longer window between when you find out the pool-wide selections and the start of the games, however you can only find out the pool-wide picks one round of games at a time (in other words you have to check around 12:55pm for the 1pm games, again at 4:20pm for the 4:25pm games, etc). This makes it difficult if not impossible to try and utilize parlays on a week to week basis.
Keep us posted CK on anything you think is worth a wager. I think I need to lay off on Thursday nights. I do better on the weekend.
Going to take -5.5 Temple (vs UConn). Already took -23 Virginia a few days ago, it's at 27 now but it is Kent State, the same team that lost 66-0 to Ohio State. Will post more tomorrow once the final numbers are out and total team points.
Will do. The stuff I get from OFP only becomes available right at 1:00pm on Sunday. I barely have time to hit copy, paste, read the results of my macros and plug it into stuff before teams actually kickoff. However, the OFP data will give me signals for 4pm games as well as SNF and MNF and if those signals are significant enough, I can post them in here.
This is a fallacy. Its a matter of market efficiency. As the market grows, the line becomes more predictive of the outcome. Sharps pound lines right before kickoff, because most sports books severely limit the amount someone can bet into an opening line. Let say that the Cowboys open at -2.5. The public loves the Cowboys, so they get a lot of bets. As soon as the line moves to -3, sharp money is going to come in on the other side. Generally speaking, sharp money is going to pound any number that gives them positive expected value. Also keep in mind that sportsbooks have a huge advantage because they can limit bet sizes. So they will limit sharp bets while giving the public a bad line, then they'll move the line when they're ready to take sharp bets.
Over the last 10 years, teams that have dropped three straight games and come into their fourth game as underdogs are 58% against the spread. The Buccaneers, Raiders and Jaguars all fit this description this week.
Thus far, underdogs are: 30-18-1 That's a 62.5% winning percentage for underdogs against the spread. If there's one weakness in the system I mentioned, it is the system's tendency toward underdogs. The general philosophy of leveraging smart bettor tendencies against not so smart bettor tendencies leads to a consistent bias toward underdogs. Bad bettors tend to bet on favorites. Good bettors don't necessarily go with all dogs but they go with dogs more often than bad bettors, and so if you're sensitive toward the TREND rather than the absolute picks, you're going to end up with a lot of dogs. The last time I tested this was 2012 and this was absolutely the case that year. The system steered toward mostly dogs, and underdogs were killing it during the first five or six weeks of the year. At some point that turned, and the favorites started to come back...and the results of the system suffered predictably. That said, the OfficeFootballPool predictions are between 75% and 85% depending on the signal strength you settle on. So I continue to think there's more going on there than just leaning toward dogs at a time when dogs are beating the spread.
I would be skeptical of any system that is based on market inefficiency. Markets adapt. Any type of system that ignores the current spread is going to lead to ruin.
Betting pools are kind of a different beast than standard betting. As I mentioned before though, I would avoid any type of system that ignores the current spread. EDIT: here is what you could do if you wanted to incorporate expert picks into betting against sportsbook lines. Take the historical data on expert selections and correlate that to the final score or outcome of the game. Then use those results to predict future scores based on expert picks. Essentially an algorithm with expert selections as inputs and the predicted score or W/L as the output. If you're doing W/L, then you need to come up with a win probability for each team. Then use your results against the available lines to determine what (and how much) to bet.
Many of the best sports bettors on the planet wait to pound games all the way to kickoff. I'm talking about the real people in Vegas, the guys who will put a substantial bet on one side to drive the lineup then pounce once it's at the number they prefer. Good sports bettors are aware of how the public will drive a line and THAT is how they make their money. Sometimes a line opens up that they love and they pounce on it before Joe Public moves it against them...sometimes JOE public will move in their favor. Again it's all circumstantial.
If good sports bettors are aware that the public will drive a line, then so are the books. Why would they open with a number that the public drives up?
50-50 is much more difficult when you are moving a line. If half the people have -2.5, and the other half have have +3.5, thats bad for the books. Its nearly impossible to be 50-50 when a book moves a line. If anything, the book will leverage their ability to limit action. They'll open the public favorite at -4. When they're ready to take sharp action, they'll move the line to +3. They'll end up 55-55 before even counting their vig.
Rarely if ever will you see a line go from 2.5-3.5... That opens up middling opportunity, the books are smarter than all of us by 10. And with live betting they realize they can't "limit" the betting as easily.
This is a great clip for all of you who haven't seen it.. [video=youtube_share;limVVER9X1I]http://youtu.be/limVVER9X1I[/video]
12:00 games Vanderbilt TT 14.5 (@ Kentucky) 3:30 games -23 Virginia (vs Kent State) Virginia TT 35.5 (vs Kent State) 4:00 games -5.5 Temple (@ UConn) Will update this post in the next hour and a half, waiting to see if any numbers change to my favor...
Eh I'm sticking with those picks, might have something for Cincinnati @ Ohio State and San Diego State vs UNLV which I will post later.
There's a lot more defense being played in Indiana than I expected. Sent from my Nexus 10 using Tapatalk
Fsu is getting stomped. Might need to hedge with a live bet. Wash +8 Sent from my Nexus 10 using Tapatalk
