The Huddles "In Season Articles"

[tazinib1]

I'll give you a 5 star ConFac...hold on...just a sec.....................

 

 

**Rips off clothes and runs through thread naked**

 

 

Now THATS ConFac baby!!!

[Tripleshot]

I'll give you a 5 star ConFac...hold on...just a sec.....................

 

 

**Rips off clothes and runs through thread naked**

 

 

Now THATS ConFac baby!!!

 

 

Bah. Five star ConFac would be making your avatar a pic of yourself naked.

[stevegrab]

Bah. Five star ConFac would be making your avatar a pic of yourself naked.

 

 

If he does that I might have to put him on ignore, he'd literally be streaking through every thread he posts in. I don't think we need that, it would really burn the eyes

[jonny2fngrs]

Nobody is spoon feeding you anything. You basically just don't get it when it comes it projections and confidence rating. Without being a bag of dicks I would suggest you pick up an beginner's guide to basic statistics.

 

 

Lol...I am sure I have worked with more statistic models than most, and I can tell you that most quantitative performance models use confidence intervals. No fear, I get it...The Huddle faithful have spoken...these are great projection tools for the obviously more advanced owners that my 20+ years of experience just doesn't get. For those that do understand, I have a question...

 

The weather man stated that there is a 100% chance of rain tomorrow and he is 25% confident that is the case. Should I take an umbrella with me when I leave? Haha..I already know the answer...Stay in and work on my tact and post my shiny silver spoon on ebay

[delusions of grandeur]

If you get it, then you must just be here to bitch about the inexact science filled with differing degrees of variables that is FF projection. I'm sorry I gave you the benefit of the doubt that you actually sought to understand. Have a nice day.

[darin3]

Lol...I am sure I have worked with more statistic models than most, and I can tell you that most quantitative performance models use confidence intervals. No fear, I get it...The Huddle faithful have spoken...these are great projection tools for the obviously more advanced owners that my 20+ years of experience just doesn't get. For those that do understand, I have a question...

 

The weather man stated that there is a 100% chance of rain tomorrow and he is 25% confident that is the case. Should I take an umbrella with me when I leave? Haha..I already know the answer...Stay in and work on my tact and post my shiny silver spoon on ebay

 

 

Bad use of the "chance of rain" analogy. You might want to look up how that all works. It doesn't fit your argument. Just sayin'.

[Gopher]

I haven't really paid much attention to the confidence factor, to be honest. I generally look at projections here, compare them to a couple of other sets of projections, and come up with my own "confidence factor," based on the various projections I see, and a number of other factors (not the least of which is simply my own opinion/gut feeling on the player). I think the Con-Factor might be a good tool to use as a tie-breaker, for people who are weighing players with similar projections (if you don't have your own opinion on the players, which is how I would typically break such a tie), but I wouldn't read into it any more than that, personally.

 

Not to mention, the difference between one star and five stars isn't all that great, is it? In other words, if one star means the writer has virtually no confidence in the projection being accurate, what's the point of projecting? I interpret it to mean that (if you were to assign percentages to those star ratings):

 

* = 50-55% chance that projection is reasonably close to the end result.

** = 55-60%

*** = 60-65%

**** = 65-70%

***** = 70-75%

 

In other words, it's not like one star means 20% (virtually no chance), while 5 stars means 90% (a lock). We should all know that fantasy football is far too unpredictable for anything to be that certain.

 

For what it's worth, I agree with the opening poster, in that as you get closer to the top of the rankings, the confidence factor should trend in the same direction as the projections (to some degree). In other words, if you're top-ranked player has one star, something doesn't add up (you're not projecting that player realistically). After all, projections are really nothing more than a weighted average... somewhere between the player's "floor" and "ceiling" for any given week/season. The confidence factor is what determines how much weight pulls the number in one direction or another.

[jonny2fngrs]

If you get it, then you must just be here to bitch about the inexact science filled with differing degrees of variables that is FF projection. I'm sorry I gave you the benefit of the doubt that you actually sought to understand. Have a nice day.

 

 

I honestly thought I was missing something with both these projections and did come here to figure out what it was. After reading numerous replies that confirmed these projections are exactly what they look like and there is nothing more to them...yes I do get it now. I never expect projections to be exact, I was merely looking for something that is consistent.

 

Bad use of the "chance of rain" analogy. You might want to look up how that all works. It doesn't fit your argument. Just sayin'.

 

 

100% Chance of rain = Projection

25% confindence = ConFac

 

Thanks for playing.

 

Thanks everyone for your help in explaining these projections. I guess it is just a case of "I get it but I don't".

[darin3]

I honestly thought I was missing something with both these projections and did come here to figure out what it was. After reading numerous replies that confirmed these projections are exactly what they look like and there is nothing more to them...yes I do get it now. I never expect projections to be exact, I was merely looking for something that is consistent.

 

 

 

100% Chance of rain = Projection

25% confindence = ConFac

 

Thanks for playing.

 

Thanks everyone for your help in explaining these projections. I guess it is just a case of "I get it but I don't".

 

 

Now you're just dancing. I was trying to help ya out. The weather thing doesn't fit your argument. When a weatherman says there's a "40% chance of rain", it doesn't mean what you think (or applying to this argument). It means it WILL rain in 40% +/- of the national weather service's area for which he/she is making the rain chances at.

 

Linkypoo

 

[jonny2fngrs]

I interpret it to mean that (if you were to assign percentages to those star ratings):

 

* = 50-55% chance that projection is reasonably close to the end result.

** = 55-60%

*** = 60-65%

**** = 65-70%

***** = 70-75%

 

 

Thank you. Although this is not verified percentages, this is the type of info I was looking for....nothing exact just close. Thanks again.

[Gopher]

Thank you. Although this is not verified percentages, this is the type of info I was looking for....nothing exact just close. Thanks again.

 

That's just my interpretation of the star ratings. Doesn't mean I'm right.

[jonny2fngrs]

Now you're just dancing. I was trying to help ya out. The weather thing doesn't fit your argument. When a weatherman says there's a "40% chance of rain", it doesn't mean what you think (or applying to this argument). It means it WILL rain in 40% +/- of the national weather service's area for which he/she is making the rain chances at.

 

Linkypoo

 

 

 

Thanks for the link. I did not know that about weather forecasting and interestingly enough according to that article, one of the variables used to forecast the weather is confidence. My contention all along has been that the confidence should be, or capable of being, "built in" to the projection and Gopher addressed that.

 

We could argue semantics over whether the weather analogy fits in this case but at the very least after reading your article, I think it is fair to say that weather forecasting could easily be used as a like analogy. Thanks again for that link, it was educating.

[BA Baracus]

I would say that con fac is akin to standard deviation if we are talking statistics

[Jackass]

Thanks for the link. I did not know that about weather forecasting and interestingly enough according to that article, one of the variables used to forecast the weather is confidence. My contention all along has been that the confidence should be, or capable of being, "built in" to the projection and Gopher addressed that.

 

We could argue semantics over whether the weather analogy fits in this case but at the very least after reading your article, I think it is fair to say that weather forecasting could easily be used as a like analogy. Thanks again for that link, it was educating.

 

FWIW your point is worth thinking about in my opinion. I think your posts will be better received if you go with the tone of your last post rather than some of your earlier ones. Just my

[rattsass]

I am far from a statistical analyst as you will find. When I saw the confidence factor added I "got it" immediately. This is what we think he will get. This is our level of confidence. I can relate that to the way I set my lineups every week - determining risk/reward. And I appreciated their addition as another level of projecting.

 

I really don't see why this is so difficult or has led to suck gnashing of the teeth. I would rather have it than not, if you disagree you should pretend you don't see that column.

[Papajohn]

 

 

Lol...I am sure I have worked with more statistic models than most, and I can tell you that most quantitative performance models use confidence intervals. No fear, I get it...The Huddle faithful have spoken...these are great projection tools for the obviously more advanced owners that my 20+ years of experience just doesn't get. For those that do understand, I have a question...

 

The weather man stated that there is a 100% chance of rain tomorrow and he is 25% confident that is the case. Should I take an umbrella with me when I leave? Haha..I already know the answer...Stay in and work on my tact and post my shiny silver spoon on ebay

 

 

Then I just feel bad for whomever hired you to do any work with statistic models since all of your post would indicate that you haven't. But never mind, there is no point in continuing this thread (and I do see the irony in me posting then)

[koralkillers]

I really miss the old start/bench column...S1..S2 etc...used to love it when all my guys for the week were S1's....other wise The Huddle is top notch as always....oh..one other suggestion for next year is have the ROS column come out on Tuesdays to allow for Monday nights players to be adjusted in rankings accordingly...

[Tripleshot]

If I ever get some free time on my hands, I have it in my mind to pull weekly scoring info for my teams this year and compare them to predictions from The Huddle, ESPN, and another pay-site to see what shakes out. Methinks it would be interesting.