Math has a well-known Hillary bias.

[Tarc]

Seems there's a bit a bit of horn-tooting today over a new poll where Sanders leads Clinton nationally, but let's dive a little deeper into RCP's findings.

http://www.realclearpolitics.com/epolls/2016/president/us/2016_democratic_presidential_nomination-3824.html

RCP Average 3/16 - 3/22 -- -- 51.5 42.5 Clinton +9.0
Bloomberg 3/19 - 3/22 311 LV 5.6 48 49 Sanders +1
FOX News 3/20 - 3/22 410 LV 5.0 55 42 Clinton +13
Quinnipiac 3/16 - 3/21 635 RV 3.9 50 38 Clinton +12
Monmouth 3/17 - 3/20 391 RV 5.0 55 37 Clinton +18
CBS News/NY Times 3/17 - 3/20 388 LV 6.0 50 45 Clinton +5
CNN/ORC 3/17 - 3/20 397 RV 5.0 51 44 Clinton +7

So the Sanders camp cherry-picks the sole poll that has Sanders up, and it is only +1 which as any Intro to Statistics class will tell you, is like likely a tie, and looky here;

Democrats Evenly Split Over Clinton, Sanders in Bloomberg Poll

The national poll of 1,000 adults was conducted March 19-22 by Selzer & Co. of West Des Moines, Iowa. The overall sample has a margin of error of plus or minus 3.1 percentage points, while the subgroup of 311 Democratic primary voters has a margin of error of plus or minus 5.6 percentage points. The subgroup of 815 likely general-election voters has a margin of error of plus or minus 3.4 percentage points.

Not only does the title debunk their claim, but also the numbers.

[virtualobserver]

1. The Bloomberg poll is rated A+.......

[SFnomad]

3. You don't understand margin of error, do you? n/t

[virtualobserver]

6. You don't understand the value of a current, high quality poll, do you?

[SFnomad]

9. So you don't understand ... that's what I thought

Margin of error is in every poll, including the highest of quality of polls. Margin of error and confidence level is a function of sample size, not quality of poll.

[virtualobserver]

10. Some recent polls have been off by 30-45 points, and all in Hillary's favor....

So for me.....the quality and reputation of a poll carries much more weight.

The margin of error stated for a poll is only a number.....if the pollster isn't competent, the MOE that they claim is as useless as their result.

[SFnomad]

12. So you don't understand and you don't want to understand

Thanks for the clarification

[virtualobserver]

15. I understand that generally, I can't take polls seriously....or you for that matter....

When polling errors have been 5 to 10 times the MOE.....the competence of a pollster is the only thing that I look at....I dismiss all others.

[Bill USA]

11. one poll is one observation or measurement. Using several polls gives you more reliable statistics

[NurseJackie]

2. Funny how that works! :-P

[basselope]

4. +1 > 0

Not much bias there.

[Tarc]

7. Allow me to provide a picture

[basselope]

8. +1 is STILL > 0

So sorry it didn't work out for you.

[Tarc]

13. I'll give you a hint; Sanders isn't really up 1 in any poll

You also ignore the rest, the ones that have Clinton up form +5 to +18.

[basselope]

14. +1

Sorry it didn't work out for you.

[DanTex]

16. LOL. Check back in a month or so. Bye bye Bernie!

[basselope]

17. +1

If Bernie is gone say hello to President Trump.

[hellofromreddit]

18. As long as you're beating others over their heads about math, shouldn't you use the correct formula?

Wilson score interval if you're curious.

What you provided is the naive formula for students to play with in class.

[ladjf]

5. Math, like statistics, is a process.

Each math equation starts from one point and preceeds to a conclusion.

However, unless the basic values of the symbols are defined ,the results are meaningless, or worse,
may be used to mislead the uniformed.

Clintonians and Sanders fans alike are fond of this process. For honesty's sake they should take care
that the basic premises have been define, otherwise the outcomes are fatally biased.

Clinton backers say "math proves that Hillary will win"
Sanders backers say "math proves that Bernie will win".

For the potential voter, it becomes a process of deciding on a product based on the arguments in the
advertisements that were professionally created to lead the viewer to a conclusion favoring the vendor.