When is a positive news story an negative?

Basically when the writer of that story wants to implant subliminal negatives to count any overt positive. Example:

The UK Daily News, who write a lot about American events, lead with this headline: "Trump's approval rating among likely voters soars to his best in 23 MONTHS"

Followed in the articles bullet headliner with overt positives:

Trump's approval rating among likely voters soars to his best in 23 MONTHS at 52 per cent after State of the Union address as border-wall shutdown talks intensify
Rasmussen Reports poll as Trump at 52 per cent approval, his best showing in 23 months and a higher number than his winning edge in 2016
Significant up-swing since government-shutdown low of 43 per cent.
New numbers were collected in the three days immediately following State of the Union address
Asked what Monday's number mean, a senior Democratic House aide confided on background: 'I don't know yet if it's horrible, but it sure isn't good'

Click to expand...

Ah, but the last bullet implants the negative:

Polling average is just 42.4 per cent, including mostly those surveys that are open to all Americans; Rasmussen polls only 'likely voters'

Click to expand...

This suggests this is slanted because "Rasmussen polls only 'likely voters'" but if you include people who do not vote and unlikely to do so it's really only 42.4 percent
and even though the will not vote let's count them because it lessens the poll. And it goes on:

Positive: "Donald Trump's job approval rating among likely U.S. voters hit 52 per cent on Monday in a daily tracking poll" followed by the negative implant.
Negative implant: "conducted by Rasmussen Reports, the polling organization he uses most frequently to promote himself." AHA...

Positive: "That number is his highest since March 6, 2017, less than seven weeks after he took office."
Negative inplants: "They were even at 39 per cent on Monday."
"Overall, 47 per cent of likely voters disapprove of Trump's Oval Office performance. That's a low water mark since November 2, 2018."
(focus on the 47 percent disapprove, don't think for yourself and realize the other side of that is 53 percent approve)
"It's not unusual for presidents to get a polling 'bump' after the high-profile annual address"

"Trump often cites Rasmussen as a rare example of a trustworthy poll, suggesting others are operated by 'fake news' outets that are slanted against him."
For those already negative it's aha, if Trump approves .... well....

And how about those already in that mindset, let's feed them that Trump was not really elected President...
"The president won 46.1 per cent of the votes cast in the 2016 election, prevailing on the strength of a commanding majority in the Electoral College."
and..
Let's cite a polls we can believe: "An average of presidential approval polls maintained by Real Clear Politics now has the president at 42.4 per cent."

The article goes on, while starting with a positive headline for Trump then implanting negatives even concluding with pictures and video of Pelosi against Trump.

So what to do? Read carefully, weigh the creator, the intent and read between the lines whether seemingly good or bad. As to polls, I personally do not place a lot of value in any poll because the outcome can be lead by the tone of the questions.

All that is just my opinion..

Most of those polls are BEFORE the speech as well, so that's probably why the 42% mark isn't really accurate. Rasumussen, while also probably inaccurate because of who the poll, is not nearly as innaccurate as the previous number.