Why is these two groups of folks can never get together? And it seems you can never find someone who has both sides.
It would seem to be a marriage made in heaven.
It would seem to be a marriage made in heaven.
-
-
most guys here are just addicted to watching sports and don't truly have deep player/team knowledge, they only think they doComment -
Because the 1st group of people win money, the second dont.Comment -
Actually, if you include "insiders", such as former athletes/coaches with connections and game-day experience or certain press folks, I think the second group could probably pump out some +EV. Otherwise, it's likely a crap shoot.Comment -
They can't get together because modellers say stuff like "All publicly available team / player info is already reflected in the line and can be reflected mathematically." and "I have to correct my efficiency ratings for field-specific biases" while sports-expert guys say stuff like "me like da bears. Them motivation much gooder than browns." / "I remember that Al Bundy once caught ran for four touchdowns in a single game so his team will destroy the other team". / "This is the lock of the century, I am betting four hundred dimes, the dog, the cat, my house, my neighbours house, and all the change I found in the bottom of my couch on it"
I do listen to the few genuine insiders I know though (a couple bookies, a horse trainer, and a guy who works for a soccer team) because that is often not public information yet. I also anyone who actually knows more structurally about a sport than I do (for example a massive Volleyball fan that can explain to me that injury to a team's libero should still affect their offensive production estimate). Everyone else is a mouth breathing member of the unwashed masses who only exists to ensure sportsbooks don't go broke.
I'm exaggerating slightly but have had a bad week with people giving me "helpful" tips.Comment -
Originally posted by JustinBieber
dreamer 
Comment -
Modelers= People that can spot +EV opportunities.
Folks with deep player knowledge= People that create +EV opportunities.Comment -
Another pleb surfaced. there must be a full moon or something.Originally posted by CFA
Comment -
Dreamer? Im doing pretty well, I think a lot of other people are too. We arn't selling you locks of the week for $35 either.Comment -
98% of all gamblers lose $$$ , that includes 98% of people in the think tank.Originally posted by JustinBieberComment -
Glad your winning and happy.Originally posted by JustinBieber if not its because your buying into nonsense. Only you know the truth, glad I could help. 
Comment -
How do you know? Seems like a wrong premise. There are a few modelers on this board but AFAIK nobody with a truly "deep player/team knowledge".Originally posted by Flying DutchmanComment -
Didn't I just say that?Originally posted by Data
Comment -
While we're at it, let's also try to figure out why don't doctors and bums don't often hang out?Comment -
What you said sounded more like "cannot get along". To answer your clarified question, the insiders cannot get openly involved with the gamblers due to the risks of losing their jobs. So, any get-togethers of this type cannot happen on a gambling board but only at the places where the modeler masquerades himself as a statistician.Originally posted by Flying DutchmanComment -
Well, I'm a (runway) model masquerading as a statistician.Originally posted by Data
Comment -
Originally posted by ForgetWallStreet
Comment -
Here's an example of a guy who has both.
Comment -
Like most precious synergies, they are best left to incubate where you won't be looking.
Comment -
Modelers have the advantage over someone with deep player/team knowledge because a modeler has a handle on BOTH teams in any given contest while the player/team knowledge is normally limited to a single team. I can tell you every single player on the 49ers roster, what scheme we're likely to run, who's likely to play, etc, etc... but I'd be lucky if I could name over 50% of the starting players on any given opponents team.
Now an ex-player or someone else with insider knowledge could have a huge advantage but that's another issue.Comment -
roasthawg, but someone with deep knowledge has deep knowledge opposite superficial knowledge of modeler.
As I think to win you should be modeler with advices of depp-knowledge-guys (two at least).Comment -
What is the definition of advantage? I am not a modeler and I do not have any deep player/team knowledge. What I do have is experience in finding ways to make profitable wagers/investments due to changing and/or comparative market prices.
So who has the advantage? The real point IMO is that it doesn't matter. The objective is to find a way to be profitable regardless of the criteria and/or method you use to do it.
Some have the opinion that if you don't do it their way, then you are doing it wrong and/or you can't possibly be as profitable etc.
It ain't necessarily so!
Joe.Comment -
Every now and then you get someone very smart posts on here.Originally posted by u21c3f6
sadly they're mostly lost in the mire of nonsense. 
Comment -
Lesee, can I find an advantage as a modeler? Yep, done that.
Can I find advantage as a "deep knowledge" guy? Yep, I've seen folks who've done that.
I think the OP point was, you might really catch lightning in a bottle if you can get both together. Rarely done. We're not talking about which is better.
I think 1) these two types of folks tend to dislike each other in addition to the point 2) it is professional suicide for those who are in any way part of the sport: management, refs, players, ball-boys, cheerleaders, etc. is reason(s) enough for it to rarely be seen.
Plus, if it really works, why tell us clowns among the unwashed masses? Then WE might try to make it happen and erode the market even further.Comment -
Another road block is money. I can see both of them requesting a larger share of profits. Most nerds don't have good personal skills and the connections to get in touch with an insider or former player.Comment -
You know come to think of it, the way I have seen syndicates described on this forum, if they really operate in that way then I would imagine that they probably do avail themselves of both angles!Originally posted by Wrecktangle
Also note that the OP said folks with deep knowledge, and I took this to not necessarily mean insiders -- so from my perspective we all have varying levels of "substantive" knowledge about the sport we are trying to handicap or model.
I studied a little bit of social science in college (probably just enough to hurt me!) and I was taught by those folks an axiom that I have also seen expressed on this forum and that is:
in order to be worthy of any credence statistical correlations or models have to make sense on the substantive level.
In other words if you want to avoid data mining you have to first start from a theory of why you think there is a relationship between two or more statistics and then investigate.
The other (bad) option is to just massage your database until you find a model with a high R square or whatever the case may be -- this is data mining and if you try to model with no substantive knowledge of how the sport works then you run the risk of data mining.
Conversely, a punter who's strength is deep knowledge probably can't help but have a few stats in her head like maybe how have they done ATS at home over the last 5 contests -- so it is a question of degree and it is something that I think all modelers should seek to accomplish -> an integration of statistical power and substantive knowledge
For me it is a question of degree .................modeling with absolutely no player/team knowledge yields a prime suspect for data mining
Comment -
You don't have to start from a theory, but when you arrive at a good statistical relationship you have to be able to grok if it works or not.Comment -
you should always start from theoryOriginally posted by That Foreign Guy
pour over enough numbers and some relationship will pop up like malay butter and the DJIAComment -
Isn't this all about finding the correct correlation between past data and predicting future data? You could have all the data in the world but without knowing the ins and outs of the game it is uselsess imo. Without having a base to start from you would waste so much time trying to figure out the data you had it would not be worth your time.Comment -
If you start with a theory, you set out to disprove the theory. If you operate with any a priori notions of how you might expect, or want, the data to convey a certain reduction then you will probably spin the data in some way to suit those expectations, consciously or subconsciously. Objectivity is tough to attain and practice, its a delicate process.
The fallacy of the general factor of intelligence comes to mindComment -
And then you go "Hmm there is no logical relationship there, I can ignore this".Originally posted by Peregrine StoopComment -
"Follow the Data" - John Tukey, Exploratory Data AnalysisComment -
majority of the ex-players and ex-coaches on tv suck at picking straight up winners IMO and many of them have superbowl rings as well as deep knowledge of the teams. anyone really want help from t bradshaw, shannon sharpe, cris carter or marino handicapping games?Comment -
I'd take the modelers any day. Almost no one has "deep" team knowledge anymore.Comment -
I actually think everyone has access to "deep" knowledge nowadays, so it's all reflected in the marketOriginally posted by PRCComment
