I completely disagree with Fez on the market

Just a donkey chiming in. I don't care of sample size for years past or years in the future. It's 2010- 2011 NFL and what is winning NOW is all that counts for me. Been a member here since day 1. From my ears all I hear, once things go south, is " Can't Be Beat." " Too Tight" The $ 25 / month is like buying a daily newspaper to me, but going forward, what should a member at this site expect in the future? I live on Miami Beach, so live M bets mean nothing to me. Quarter or half time bets that can only be made in LV are worthless. As a paying member my expectations were never that high and for the most part kept my mouth closed and ears and eyes open so I could learn. The learning part is nill here. I'm stupid to think FEZ is going to teach me the proper math on bets. I guess long term that hurts him and others that have that talent. No problem. So I ask without malice, other than Shooter and Big Papa, what do I get or can learn to " fish for myself."

[QUOTE=IceTea;34243] Again, if people want to digress into talking about long term success and maximum win rates, feel free to do so. Personally, I disagree with Fez and others there as well, as I've seen guys who are hitting at long term rates of multiple seasons in the NFL. I'm talking over 56% for 3+ years. I guess these guys are freaky incredible and the best in the world? Personally, I don't think so, I think they are just very good, but I'm sure there are more out there than people think. Especially guys like IrishTim who thinks the market is so efficient its impossible to have long term success like that. My friend, what happens when someone does achieve that success? Do you go "poof" like dandelion spores? But all we are talking about here is the 2010 NFL season. 30 plays, 60 plays, 90 plays, it does not matter to me. I am seeing guys beat it and Fez says that's not the case, so one of us must be lying.[/QUOTE] Well A) you need to define long-term rate. Anything under 2000 plays is unreliable. If someone goes 2-1 this season, does that mean they're a long-term 66% handicapper? Ignoring sample size is ludicrous. If you don't like my first coin flipping example, cf. the famous Warren Buffett coin-flipping example.

What can one expect to get from this site for $25 per month? That in itself is worthy of a new thread. I'm sure the answer is a bit different for everyone, but we all want to win. As far as this thread, 6 pages is enough already.

[QUOTE=bigrobbie;34248]More disturbing to me is that I see the Fez "sticky thread" has been removed. Does that mean we won't see much of anything from him for a while? I hope not.[/QUOTE] The sticky is removed when the week is over so it can sink. There hasn't been one made for this week yet, ostensibly because he hasn't found any WA plays yet.

Got it Joel - thank you. Fez must be online, as I see he started his sticky thread for this week. No picks yet, but let's hope for some good stuff. We're up to 7 pages for this thread, so I won't post in it anymore. Again, thanks.

[QUOTE=Fezzik;34194]I'm THE GUY WHO hit a documented 65% over a 2 years period in the Hilton Contest, 170 plays! Ok, soft lines. "Only 62% vs. market numbers". So Mr. Members could use MY results to laugh at my feelings......of course he would have to ignore a bunch of other Fezzik data points.......[/QUOTE] Just a quick response to this and other posts. Fezzik is correct, many small samples like Members presented can be data mined to produce statistically insignificant results. Of course those who are incapable of any statistical analysis will continue to be wrong and thus insist statistics don't matter. Now, by construction, the subject here is NFL success rates. And not coincidentially, NFL records are the most subject to data mining primarily because of very small sample sets and relatively large variance. Anyone who thinks 500 picks is a large set in a statistical sense is really ignorant. I brought up FJ's CFB record as a contrast and a check on data mining his NFL record. One check to see if a result is a statistical outlier is to look at data for similar endeavors. Of course, CFB is not the same as the NFL, but clearly FJ isn't some sports/football clairvoyant as his CFB and all sports record shows. It is hardly irrelevant. Finally, Sundown is correct that discussion of one year's record especially in the NFL is really going to have such statistical variance as to make it almost worthless and a waste of time.

Great Thread-Don't Close It [B]This thread presents some of the great problems that sports gamblers face with touts and one another: 1.How do you know if someone is getting lucky or is a good handicapper? 2.How do you separate winning handicappers from lucky ones? 3.How many picks does it take to figure this out? 4.How can you verify accuracy of when and what numbers plays were released at? 5.Is it even possible to answer these questions with any certainty?[/B] When everyone wants the pride (and income) associated with being a winning handicapper, it's easy for all data to get fudged. Fudging plays a half-point here and there is worth thousands and thousands of dollars to a tout. That's why they're (all of us, actually) hyper-aggressive in getting that extra half-point on their releases, even if it's only at one obscure book. Computer Bob, love him or hate him, and with his skeptical attitude(well-founded most likely) is probably the most important guy on this site. He's keeps the records as honest as possible imo and points out any discrepancies. Touts usually hate him because he's a threat to their bottom-line. Computer Bob analyzes Fairway Jay. It should be taken as a compliment. At least he has a half-belief he's a winner. Why waste time scrutinizing all the scamdicappers? We already know their full of doo-doo. Point is, we probably can't figure out anything with absolute certainty in this medium except some of what has happened, and that the reality to winning at sports betting can be sobering at times. Grinding out play after play, with slight edges you're not even sure exist many times, and with the price of tout plays that making losing money virtually the only absolute lock you can count on for the average bettor is the reality most people don't want to face. It can be a tough business. It's no wonder to me why even winning handicappers seek to make money outside of betting. No variance income in sports betting? Sounds good to me. It helps to understand this stuff is my opinion. ***************************************************** One more thing:I can produce results over 30, 60, 90, 120, or even 500 plays to show almost anything I want. A not-quite so widely available half-point here or there can turn 1 out of 3 break-even handicappers into 56%ers over any random 500 plays. It's that simple. And that's before you pay. If you're paying money for these plays and getting any lines that aren't at least 1/2 point better than the release number, you could easily be throwing away money. Touts also make a ton of money off of super short-term results. When you have 7 touts handicapping spreads and totals on 4 different leagues NFL, NCAA football, NBA and college hoops, that's 56 different columns. It's a mathematical certainty that 3 or 4 of these "columns" will be going 15-5 or 22-8 virtually all the time for 50% cappers.

[QUOTE=ComptrBob;34263]Just a quick response to this and other posts. Fezzes is correct, many small samples like Members presented can be data mined to produce statistically insignificant results. Of course those who are incapable of any statistical analysis will continue to be wrong and thus insist statistics don't matter. Now, by construction, the subject here is NFL success rates. And not coincidentally, NFL records are the most subject to data mining primarily because of very small sample sets and relatively large variance. Anyone who thinks 500 picks is a large set in a statistical sense is really ignorant. I brought up F's CB record as a contrast and a check on data mining his NFL record. One check to see if a result is a statistical out lier is to look at data for similar endeavors. Of course, CB is not the same as the NFL, but clearly DJ isn't some sports/football clairvoyant as his CB and all sports record shows. It is hardly irrelevant. Finally, Sundown is correct that discussion of one year's record especially in the NFL is really going to have such statistical variance as to make it almost worthless and a waste of time.[/QUOTE] Is there a correlation between the number of plays and market conditions changing? I used to think that you needed thousands of plays before a certain handicapper was worth following. However, is someone like Dr.Bob still worth tailing in football? The yearly sample size (For Some Handicappers) is small in the NFL & CFB and for some handicappers 500 plays could be 5 years.

[QUOTE=grux;34274]Is there a correlation between the number of plays and market conditions changing? I used to think that you needed thousands of plays before a certain handicapper was worth following. However, is someone like Dr.Bob still worth tailing in football? The yearly sample size (For Some Handicappers) is small in the NFL & CFB and for some handicappers 500 plays could be 5 years.[/QUOTE] For those with statistical models, certainly a market changing to "better" (closer to yours) lines would reduce your plays. Also, lines getting "worse" would produce more plays. Tailing football cappers is generally more an article of faith that the capper can adjust his model as time goes by and be a winner than having a 95% confidence level that the margin of error in his long term win rate is something in the range of +/- 1%. As an example, the 95% confidence level figure for a +/- 1% error range would statistically take tens of thousands of picks to achieve.

Hey fez On that Jay will be under 50% rest of year NFL picks 2-0 Today.