Zourah
Zourah
Joined: Mar 26, 2016
  • Threads: 4
  • Posts: 125
September 11th, 2016 at 7:22:23 PM permalink
I enjoy having discussion on these games. I realize Joel appears to be trying to make a buck off of the discussion, but to start a thread talking about how great your picks are, I have a terrible two weeks and then give up and pledge to come back " once the data has worked itself out " ..... I'm sorry that's pathetic.

A cynic might suggest that the guy is afraid of starting 4-15 against the spread or something
AxelWolf
AxelWolf
Joined: Oct 10, 2012
  • Threads: 144
  • Posts: 18133
September 11th, 2016 at 9:07:27 PM permalink
Quote: JoelDeze

First of all, my site was never built or touted as a picks site. I pride myself on building solid tools and utilities and handling big data so that it's easy for people to "research and identify" match-ups that have a high probability of winning. All of the services are research oriented.

The picks section (the smallest section on my site) just takes the highest rated system picks that fall in an x-y ML range criteria and spits out the probable winner. I have to ultimately decide whether or not that team covers the ATS. On a SU format, the winning probability is going to be very high. But, not many people are going to wager a lot of money on low value returns. If you are patient, sure you can build up a 16-19% ROI using a SU format with the ML on my site. Myself, I prefer 42%+ ROI through using all of the research tools I've built.

I don't have any control over whether someone decides to stay a subscriber or not. That decision is up to the person using the tools. The money isn't all that important to me on a subscription based service. It pays the costs of the infrastructure servers, maintenance etc. I find more value in the feedback that I receive from people using the tools. If they have a very solid idea for improving the way something works on my site, I usually implement it in the offseason. I had one guy that liked my Ratings Search tool (which combined historical data) but he wanted to see the ROI on the searches. So, I built it into the tool and now it's very easy to filter any qualifier and see what the ROI return was for that search. Now anyone, including myself, can create an advanced filter search algorithm to build say a 50% ROI or above and use that filter for all games during the season. Basically, it breaks down a 50 game per week format into 3-5 games that have a high ROI and probability. It makes it a lot easier to research 3-5 games than to have to research 50 games. Good idea. Implemented.

I've thought about going to a free format (donation only) service on the site. It would be implementing an honor system of sorts. I'm not sure how that would work out but I'm looking into it. I also do not show advertisements on the site. I hate advertisements but maybe I could incorporate that type of format as well. I just like keeping things clean and easy to use.

Fair enough.

If nothing else you are well spoken and always respectful to everyone. I may not agree with your... what I call Voodo betting systems regarding the craps and BJ .But It's obvious you are a very intelligent person.

I do admire what you built and all it seems/sounds all very impressive and fasonating. Since I don't know how or why you think your sports program/tools/system and whatnot works, can you explain it to us? I'm legitimately interested.

I would like it if you would continue to post up all your picks. If somehow you don't think that's fair to your paying customers. Perhaps you can pick someone(NOT ME) to prevent PM someone your picks who at the end of the season can reveal all of your picks and results.

Would you not agree that 99% of the guy's out there are full of it when it comes to sports and betting system's?

Honestly if you have some way of beating fooball year after year, I would certainly be willing to pay for that information.
♪♪Now you swear and kick and beg us That you're not a gamblin' man Then you find you're back in Vegas With a handle in your hand♪♪ Your black cards can make you money So you hide them when you're able In the land of casinos and money You must put them on the table♪♪ You go back Jack do it again roulette wheels turinin' 'round and 'round♪♪ You go back Jack do it again♪♪
JoelDeze
JoelDeze
Joined: Apr 20, 2016
  • Threads: 49
  • Posts: 467
September 11th, 2016 at 10:01:03 PM permalink
Quote: AxelWolf

Fair enough.

If nothing else you are well spoken and always respectful to everyone. I may not agree with your... what I call Voodo betting systems regarding the craps and BJ .But It's obvious you are a very intelligent person.

I do admire what you built and all it seems/sounds all very impressive and fasonating. Since I don't know how or why you think your sports program/tools/system and whatnot works, can you explain it to us? I'm legitimately interested.

I would like it if you would continue to post up all your picks. If somehow you don't think that's fair to your paying customers. Perhaps you can pick someone(NOT ME) to prevent PM someone your picks who at the end of the season can reveal all of your picks and results.

Would you not agree that 99% of the guy's out there are full of it when it comes to sports and betting system's?

Honestly if you have some way of beating fooball year after year, I would certainly be willing to pay for that information.



Hi AxelWolf,

I also appreciate your attention to detail and not being afraid to speak your mind. You and a host of other people that visit here are very intelligent. I appreciate the kind words.

First, there isn't a single person in this world that can 100% predict a football game. Probability and Prediction are not the same thing. My systems are built on providing probability, and the probability factors strengthen as the season progresses as more data is collected. Probability can help reinforce a prediction but it doesn't guarantee one. So, to answer your question, absolutely I believe that many people do things wrong and therefore the systems they have are not accountable.

I'll do my best to explain how things work within.

The first system I built was called TSRS (True Statistical Rating of Strength). Right around 2005-2006 I was getting quite annoyed with all of the discussion people were having when deciding which team was strongest. I also never liked "human polls" like the AP or USA Today polls. So I worked over a year's time developing TSRS. Here's how it works.

Imagine how raw statistics are handled. For instance, Louisville just ran up over 800 yards of offense against Syracuse. Depending on the statistic you want to measure, how strong was the opposition statistic faced? For instance, how strong is Syracuse's rushing defense and passing defense?

My system converts raw stats into floating point numbers so that they can be adjusted, using a mix of standard deviation and covariance to handle inflation and deflation values. Every team is assigned a base floating point number for every type of statistic that can be measured. It is further adjusted based on the strength of opposition faced. If the team you are facing is weak in passing defense and rushing defense, the strength ratings are lowered to account for weak opposition.

A team's offensive TSRS is a combination of all of those individual strength indexes. The same goes for defense and special teams. Special teams receives a slightly lower index than other ratings.

When I built TSRS I talked with one of my friends, Kenneth Massey (one of the former BCS computer systems) and asked him his thoughts on what I built. He liked what I built and my TSRS ratings are posted on his college comparison ratings site every week of the season.

By the end of the season, teams that are in the top 10 of TSRS are generally going to the playoffs. For instance, last year, Clemson was the #1 TSRS rated team and Alabama was the #2 TSRS rated team. Oklahoma was #4.

The TSRS system also recalculates all opponents previously faced to see if they have strengthened or weakened. This could help or hurt a team's overall strength. An example of this is if you play Clemson on week 2 and they are considered strong (a portion of last season's data is used and combined while limited data is present). By week 8 if Clemson loses 3 games and has poor overall strength, that solid win you thought you had turns out to not be such a great win. Your rating will decline.

This first system has been in place since 2008 so I'm going into my 9th season using this particular system. TSRS accounts for close to 80% true probability on its own each season. It's a solid base forecaster that is used in a lot of my algorithms.

Head -2- Head

Head 2 Head (H2H) was the first scoring system I built. I use regression analysis to determine the true win probability based on every statistic in football. For instance, Scoring Offense might account for a true win probability of 68.3%. Passing Defense might only account for 49.8% probability. I can then take any statistic that accounts for the ability to produce a score or mitigate a score and adjust those based on the true win probability factors.

H2H then is a measurement tool for how much probable scoring each team will have in the game. If a team has a 54.3% H2H probability, that means this team should account for 54.3% of all scoring in the game.

On my site, I break down those probability factors into brackets and provide win results. For instance, 50-55%, 56-60%, 61-65%, etc. I break down the win-loss results for each bracket and also show the full results for the season. I even have a click feature for every year that H2H has been working which will show you the results from every year for that week. So, we are on week 3. Click on year 2014 and you'll see the results from week 3 for that year.

H2H is very solid and provides great probability once you get up to the 65% H2H forecast and higher.

I could go on and on but the other things I have on my site are:

RISK (risk factors based on the 5 highest trends in football) which shows how risky the H2H pick is. If it says 23% then there's a 23% chance the H2H result will fail.

POINTS (I use a historical AI algorithm) that looks back across any game ever played in the past 3 to 4 years to find any team and opponent that represents the teams in the current match-up. It tries to find teams that resemble each other very closely and it averages out all of the scores for those two teams.

LPR (Loss Profile Rating) combines every algorithm I have on my site and uses true win probability built from regression analysis to properly weight all of the systems and create a rating. The higher the rating, the higher the probability.

More importantly, all of the systems I have on my site are measured after every week and after every season. I know exactly how much probability each system accounts for in the overall platform.

A TSRS diff of 30 points might account for 88% true win probability.
A H2H value of 70% might account for 87% true win probability.
The POINTS diff of 7 might account for a 73% true win probability.
RISK of less than 20% might account for 85% true win probability.
And finally an LPR of 24 or higher might account for 96% win probability.

Last, I keep H2H reports, RISK reports, LPR reports, full game breakdowns, momentum charts that show whether a team is weakening or strengthening, full stat charts including PERD (Pass Efficiency Rating Difference), RERD (Rushing Efficiency Rating Difference) and every other stat most sites would have.

I'm only providing a brief snapshot of what I do (I also have ROI search filter tools, Oddsmaker compiled data, etc.). Like I said, the majority of the site is for researching and understanding how a team vs opponent breaks down in a match-up.

Week 1-4 is generally tough because there's limited data collected. 5-7 are solid building weeks and Week 8+ are extremely solid data weeks. For me, I wager the most when I get to the mid to end weeks. I'm very patient and I encourage others to be as well. I've lost about $1,300 on week 2 and gained $530 on week 1. I'll hold off on wagering for week 3 and 4 and wait until more data is collected and things are stronger.

I really should just make the entire site free. I'm not hurting for money and I love what I do. I only charge $20 / month anyways. Some people, as I start to communicate with them, I tend to waive their fees entirely. I know a few retired military men and since I served in the Army, I have respect for those that serve.

I hope some of this information is what you were looking for. If not, PM me and I'm more than happy to let you browse the site and see for yourself whether any of the information is of value.

Take care.

Joel
Its a dog eat dog world out there and Im wearing milkbone underwear . Norm Peterson
SM777
SM777
Joined: Apr 8, 2016
  • Threads: 6
  • Posts: 762
September 11th, 2016 at 10:27:37 PM permalink
Quote: AxelWolf

Fair enough.

If nothing else you are well spoken and always respectful to everyone. I may not agree with your... what I call Voodo betting systems regarding the craps and BJ .But It's obvious you are a very intelligent person.

I do admire what you built and all it seems/sounds all very impressive and fasonating. Since I don't know how or why you think your sports program/tools/system and whatnot works, can you explain it to us? I'm legitimately interested.

I would like it if you would continue to post up all your picks. If somehow you don't think that's fair to your paying customers. Perhaps you can pick someone(NOT ME) to prevent PM someone your picks who at the end of the season can reveal all of your picks and results.

Would you not agree that 99% of the guy's out there are full of it when it comes to sports and betting system's?

Honestly if you have some way of beating fooball year after year, I would certainly be willing to pay for that information.



All you need to know is he adjusted his prediction based on the injury of a college defensive lineman. Defensive lineman are worth 0 points to a spread in college football. Literally 0, like the line doesn't move.

I worked in the back rooms of multiple sportsbooks for over five years, and this my friend, is a fraud. A laughable attempt at trying to persuade the naive sports bettor that he knows what he is talking about.
NokTang
NokTang
Joined: Aug 15, 2011
  • Threads: 56
  • Posts: 1314
September 12th, 2016 at 3:38:55 AM permalink
Turnovers can never be predicted. Unless of course the game is fixed. So without knowing about turnovers you don't know anything. They get the lines correct.
JoelDeze
JoelDeze
Joined: Apr 20, 2016
  • Threads: 49
  • Posts: 467
September 12th, 2016 at 5:51:27 AM permalink
Quote: Zourah

I enjoy having discussion on these games. I realize Joel appears to be trying to make a buck off of the discussion, but to start a thread talking about how great your picks are, I have a terrible two weeks and then give up and pledge to come back " once the data has worked itself out " ..... I'm sorry that's pathetic.

A cynic might suggest that the guy is afraid of starting 4-15 against the spread or something



Hi Zourah,

That is not what I said. I will continue posting my picks over the next couple of weeks. I just will not be personally wagering on college games over those two weeks.

I would rather allow more data to be compiled before wagering money. Sure, I might miss out. But, I'm a patient person. There is a lot of money to be made in the middle to late part of the season.
Its a dog eat dog world out there and Im wearing milkbone underwear . Norm Peterson
JoelDeze
JoelDeze
Joined: Apr 20, 2016
  • Threads: 49
  • Posts: 467
September 12th, 2016 at 7:18:28 AM permalink
Quote: SM777

All you need to know is he adjusted his prediction based on the injury of a college defensive lineman. Defensive lineman are worth 0 points to a spread in college football. Literally 0, like the line doesn't move.

I worked in the back rooms of multiple sportsbooks for over five years, and this my friend, is a fraud. A laughable attempt at trying to persuade the naive sports bettor that he knows what he is talking about.



You are absolutely correct that the defensive lineman injury would not change the line. I never said it would. Show me any where in my post where I mentioned that this injury will cause bookmakers to change their lines. Public perception and betting changes the line. From an injury perspective, key starter injuries have a lot more impact on line changes. Losing a QB or a WR/RB that accounts for 40% of team scoring would cause the public to change their betting and affect the line movement.

I never mentioned that the line would change and why would it? The line and the score outcome are two completely different things. If you truly worked in the back rooms then you should already understand that lines are developed and incrementally moved so that equal action is on either side of the line. Bettors and action affect line movement. William Hill had the line opening at -7 and it moved to -8.5 by game time.

I personally changed "my" score prediction. And, my system had the probable scoring in favor of Clemson (35-20). I "personally" went out on a limb and predicted a much larger score. My system also had the probable scoring covering 11 points for the FSU game. FSU won by 11. That's also posted before the FSU game in Zourah's thread. Here's the quote:

LINK

Quote: JoelDeze

FSU returns 11 starters on offense and will be a favorite to make the playoffs this year. They return 6 on defense as well. Ole Miss only returns 5/5 on offense and defense. When you mention that you don't believe FSU is an elite team I have to question the validity of your research. You have them as an 8 point favorite so I'll at least let you know you are fairly close. FSU wins this by 11+. I have them winning 31-20.



FSU won 45-34 by the way. I'm only mentioning it because you never will.

I'll also supply a nice quote on the subject from some of my friends over at Maddux Sports:

Quote:


Think about the position The betting public will generally assume that the injury of a starting quarterback is a clear sign of the apocalypse. They might not give the injury of a nose tackle more than a passing glance. Its quite possible, though, that a football team will feel a much more significant impact from the defensive injury than the offensive one. The sports betting public generally attaches much more significance to an injury in a skill position than any other. When you consider injuries you have to not only think about what the impact is going to be, but also what the perceived impact is going to be. The difference between reality and perception is often where profit lives.

Its a dog eat dog world out there and Im wearing milkbone underwear . Norm Peterson
Zourah
Zourah
Joined: Mar 26, 2016
  • Threads: 4
  • Posts: 125
September 12th, 2016 at 9:51:12 AM permalink
Quote: JoelDeze

Hi Zourah,

That is not what I said. I will continue posting my picks over the next couple of weeks. I just will not be personally wagering on college games over those two weeks.

I would rather allow more data to be compiled before wagering money. Sure, I might miss out. But, I'm a patient person. There is a lot of money to be made in the middle to late part of the season.



Fair enough... I enjoy reading what you have to say about these games so I look forward to the discussion.
SM777
SM777
Joined: Apr 8, 2016
  • Threads: 6
  • Posts: 762
September 12th, 2016 at 12:19:19 PM permalink
Quote: JoelDeze

If you truly worked in the back rooms then you should already understand that lines are developed and incrementally moved so that equal action is on either side of the line.



Unless you're referring to a small time book like Aliante Station, or Boyd that takes very low limits, that is a categorically false statement that shows how little you know about this topic.

That is the myth of sportsbooks that talking heads tell you. And obviously, you believe it. Games, especially big games are booked to need the sharp side. Billy Walter's runner comes in and bets Packers -4.5, you're going to book to need -4.5 and lose money on the other side should it cover. Big time books in Vegas (the one or two that are left) book to need the side of the sharps. They know what they're betting, because their runners are identified, and have to use a player's card to wager.

Believe me, when Billy Walters bets, the hope for the book isn't to get equal action on the other side. It's to get more action on the other side so your win/loss is on the side of the sharpest bettors in the world. There are no Vegas oddsmakers anymore. They all have their DonBest and Sports Options screen, copy the lines from offshore, and manage risk based on what the sharp bettors play.
JoelDeze
JoelDeze
Joined: Apr 20, 2016
  • Threads: 49
  • Posts: 467
September 12th, 2016 at 1:41:24 PM permalink
Quote: SM777

Unless you're referring to a small time book like Aliante Station, or Boyd that takes very low limits, that is a categorically false statement that shows how little you know about this topic.

That is the myth of sportsbooks that talking heads tell you. And obviously, you believe it. Games, especially big games are booked to need the sharp side. Billy Walter's runner comes in and bets Packers -4.5, you're going to book to need -4.5 and lose money on the other side should it cover. Big time books in Vegas (the one or two that are left) book to need the side of the sharps. They know what they're betting, because their runners are identified, and have to use a player's card to wager.

Believe me, when Billy Walters bets, the hope for the book isn't to get equal action on the other side. It's to get more action on the other side so your win/loss is on the side of the sharpest bettors in the world. There are no Vegas oddsmakers anymore. They all have their DonBest and Sports Options screen, copy the lines from offshore, and manage risk based on what the sharp bettors play.



I'm not claiming to be an expert on this topic at all. I'm happy to listen to different perspectives on this particular subject, especially if you have experience in the industry.

However, I disagree with some of your statements, especially when you start to look at halftime wagers. Most books take the total for the game, subtract the current score at the half, and use that as a starting point. They look at the favorite and if the favorite is tied or behind they increase the total on the second half slightly. Obviously they take into account a lot of factors, but the line is not really what it should be at times. The reason for that is they have to adjust the line based on what they have outstanding for the entire game. This is why a lot of halftime bets are advantageous for players.

And, before you try to state that I'm incorrect, this information comes straight from several good sources (John Avello (Wynn), John Lester (Bookmaker.eu) and Aaron Kessler).

While I do agree with a lot of what you are saying, I don't believe either of us are incorrect. There are too many factors that go into play for this to be one side or another. And, the more games that have to be handled the harder it is to make those adjustments. I've had many discussions with the management at Cantor and I've seen myself hit with no wagers being accepted, limits enforced and full wagers accepted. Ultimately, these establishments still have margins and in my experience, over the years of betting, they act very similar to a trading desk.
Its a dog eat dog world out there and Im wearing milkbone underwear . Norm Peterson

  • Jump to: