Talk Fantasy Football

dode74 wrote:It can be used to see if multiple races are comparable (i.e. have different means) with each other.

And it effectively returns a yes or no answer, is the part you neglected to mention (probably because you've done exactly zero ANOVAs, to date). What ANOVA will not do is tell you which of those races are not satisfying the conditions. You may want to consult with google further before claiming you knew that all along but considered it too trivial to state

The "yes/no" is inherent within the part of my statement which you quoted. The "satisfying the conditions" is inherent in the rest of the statement, which you did not quote.

dode74 wrote:The "yes/no" is inherent within the part of my statement which you quoted. The "satisfying the conditions" is inherent in the rest of the statement, which you did not quote.

Let me walk you through this again since you obviously didn't understand when I introduced you to the concept of ANOVA on cyanide's forums. If you run an ANOVA on a set of data then ANOVA gives you a YES OR NO ANSWER FOR THE ENTIRE SET, as in, you say "are any of these significantly different than any others within this set" and it goes "yes" like some sort of cryptic oracle. It does not identify which of that group is significantly different from another. To identify specific outliers you perform different tests that require exponentially larger numbers of calculations and alpha level corrections based on the size of the original set.

That's why I laugh at this kind of thread... plasmoid isn't really taking what you say into consideration which is good, because you don't really know what you're talking about, so you're giving him advice he couldn't possibly put into use even if he were inclined to, which, to date, he hasn't. It's like a comedy routine: people who know nothing about a topic who take turn looking things up on wikipedia and parroting it to one another like it was an academic discussion.

But hey, carry on.. I love a good game of google-based chinese whispers!

Yes, that was mentioned about ANOVA in the video on the link I posted. All you are posting is further clarification on what I said rather than actually saying that anything I said is incorrect, so I'm not really seeing what your point is.

Hi Dode,
ammended to NTBB explained page to take the 1.96 CI modifier into account.
Didn't change all that much, actually.
It did mean that it is less clear just how hard (non CPOMB) bashers crash and elves triumph in high-TV box play.

It also meant I had to move the goal posts a bit for the NTBB defintion of bottom of tier 1 (from 46 to 46.5). It was a fairly arbitrary number anyway, and given the complaints I've had for making more changes in NTBB2014 than originally planned, I preferred this change to making yet more in 2015.

Cheers
Martin

Hi, I noticed the ghouls from undead and necromantic differ in price also have noticed some star players are costed different then on lrb6. Is this on purpose?

I had to move the goal posts a bit for the NTBB defintion of bottom of tier 1 (from 46 to 46.5)

Seems to me that you're shifting the stated goals in order to accomodate your desired changes. Your call, of course.

Hi Dode,

Seems to me that you're shifting the stated goals in order to accomodate your desired changes. Your call, of course.

And you'd be right. For the reasons stated above.

I think I've shown with NTBB2014 that I'm not just ignoring the stats. NTBB2014 brought lots of changes.
Also - an invisible effect of the forgotten 1.96 modifier: Until recently the site stated that I had overlooked Nurgle, and they'd be getting something with NTBB2015 - but with the 1.96 they won't any longer.

Sad though. I Wonder if anyone thinks they're an average or above average starting team. But being such a thoroughly unpopular tournament choice (for obvious reasons) they have played very few games and hence their CI is very wide.

Cheers
Martin

Hi DaCrusha.
Ooops - thanks for spotting that.
Fortunately, in the NTBB document all (intended) changes are in red. So the 80K ghouls are a typo.
But some stars do have altered costs or skills, caused by changes to the player type that they represent.
I'll ammend the pdf asap.
Cheers
Martin

For the reasons stated above.

Personally I think you'd be better off going back to 2013 and starting from there as a basis to make changes from since 2014 was so unpopular and some of the changes have been shown to have no real need. Shifting the goalposts in this way just smells like "I want to keep the changes I made so I will". Admitting 2014 was an error will likely improve the credibility of the ruleset more than sticking with it will.

plasmoid wrote:I Wonder if anyone thinks they're an average or above average starting team. But being such a thoroughly unpopular tournament choice (for obvious reasons) they have played very few games and hence their CI is very wide.

I don't think they're as hard to play as they're purported to be but they reward a mentality that I don't think is common in coaches. Certainly with a high entry cost and no blocking or ball handling skills OOTB I think you'd struggle to call them even an average starting team though.