THE new Scottish poll from Redfield & Wilton is largely being billed by the company itself as a bad news story for the SNP, thus blunting the impact of the earlier Ipsos poll which was remarkably positive for the SNP.

But that shouldn't distract us from Redfield & Wilton's good news, with the No lead on the independence question shrinking to a barely-there two percentage points. That's in line with the trend from the Ipsos poll even if it doesn't quite replicate the outright Yes lead reported by Ipsos.

The unionist spin on the persistent contradiction between the clear Yes leads shown by the very sporadic Ipsos and the No advantage reported by the more regular polls from most other firms is well-drilled by now.

The Yes leads are supposedly just eccentric outliers and therefore should be regarded as irrelevant. But polling accuracy is not determined by majority vote, and in fact there are very good reasons why Ipsos polls are carried out less frequently than polls from other firms.

Ipsos conduct their fieldwork by telephone, which is extremely expensive and therefore clients will pay for the exercise only very sparingly. It may just be that the expense of telephone polling produces greater accuracy, and is consistently picking up a Yes lead that polling conducted among volunteer online polling panels is missing.

READ MORE: Scottish independence poll: 54% back Yes in new Ipsos survey

There's no way of knowing that for sure, but by the same token there's certainly no way of knowing that the online polls are closer to the truth, in spite of their greater frequency and the weight of unionist wishful thinking about them.

The SNP will clearly be hoping that Ipsos have the right formula, because the contrast between the two polls on party political voting intentions could not be starker. While the SNP have a substantial - albeit not decisive - lead with Ipsos on both Westminster and Holyrood numbers, they have slipped behind Labour on the Westminster ballot for the first time with Redfield & Wilton, which would suggest they are staring down the barrel of a devastating defeat at next year's General Election.

How can a party make rational decisions about the best strategic way forward when the current state of play could be reasonably favourable, or disastrous, or pretty much any shade of grey in between? Some would perhaps argue they should assume the worst on a precautionary basis, but admittedly that's not an easy call.

Perhaps what will be taken note of is the direction of travel in some of the Redfield & Wilton supplementary questions.

READ MORE: New Scotland poll for Westminster election gives Labour lead over SNP

After encouraging signs in last month's poll that Humza Yousaf was moving out of the zone of being a negative for the SNP, possibly because of his very personal connection with the situation in Gaza and the courageous way he has handled it, this month sees his personal ratings take a tumble.

On head-to-heads with Douglas Ross and Anas Sarwar about who would make the better First Minister, he remains in the lead but the gap has significantly narrowed and is now very small in both cases. That must be treated with caution, though, because there's always the possibility that the trend might be illusory due to normal sampling variation.

More encouraging for the SNP is that Keir Starmer's net approval ratings have also taken a heavy tumble and are now alnost neutral at +1.

It's often been the case in the past that the personal ratings of leaders are a better predictor of election results months in advance, which, if true, favours Labour in this case. But there's still time for the Scottish public to turn decisively against Starmer and the new numbers can be seen as a potentially important step in that direction.

There's certainly no shortage of ammunition for pro-independence parties if they wish to challenge Starmer's standing with the public, most obviously by pointing out his numerous U-turns and consequent lack of trustworthiness.

Arguably a strategy of that sort should now be the highest priority.