Nah, I like the tool predicting the Switch to outsell PS4 3:1 :)

I would say that we don’t necessarily need to adjust the value of each of ones because Gamestop’s way of tracking is not based on actual sales but viewing trend on their website, having the 500M PS4 on their rankings even if it was limited and sold out is pointing to that outcome.

What I would like to suggest is to track preorder separately and do an average based on the numbers of days they where available for preorder, then add the points to the release day totals. This will cause the spike you should expect on the chart without affecting the monthly average.

I think we should also make based on the numbers of skus available, this should prevent false sales spikes like we saw with the Xbox One.

Sold out bundles are already not taken into account.

I dont get people like you and Kerotan being so happy about the inaccuracy of the tool for the last few weeks, everybody should want a tool to be as accurate as possible but instead you guys just want to gloat about how much better PS4/worse Switch is doing.

Well I'd like it to be accurate too but likewise people shouldn't be celebrating their console doing so well on the tracker for months like October or November when it's pretty obvious it does't reflect reality. If we didn't have this tracker we could have proper conversations in October, November and December about who's going to do what, but because we have it and it seems to favor the Switch it's hard to have that conversation because so many are convinced Switch will win October for example. 

Now the guy who works it is doing a good job and it seems for normal months of the year it can be a reliable means but in months like October (BO4 & RDR2) or November (BF) with have to look at the bigger picture. 

I know, I’m using this as an example as how their ranking are working.

1) That doesnt make much sense, if three skus appear on the chart that means that all three skus are selling.  Therefore all skus must be counted.

2) That was not a false spike. The spike came in the last week when there was the $100 discount and the tool reflected that.  Without that promotion sales for the X1 would have been much lower.

There was a spike, but the tool wasn't accurate because it just added the points to the total of the multiple skus appearing, thus also affecting the monthly average in favor of the XB1 which wasn't the case in the end. The tool was more accurate in the months before because this situation didn't really happened.

If we want the tool to be accurate in a ration of 1-0, which the 1 is the leader and the two others are in the decimal under it, we need to lower the error margin, especially in Q3 and Q4 when all the bundle and offers are more prevalent. Since we don't have real numbers with the rankings, we can only speculate on the number of units sold. When I'm suggesting to do an average by numbers of skus on the charts, it's to prevent the massive fluctuation we saw. Skus moving up the charts is more indicative of sales growth than a new sku appearing on the list.

I'll give some examples:

Let's say we make a chart from 1 to 10 as 1 is worth 10 points and 10 is worth 1.

Console 1 has two skus at 1 and 3, which is worth 18 points for an average of 9, while console 2 is holding spots 5 and 6 which is worth 11 point for an average of 5.5. The ratio of console1 would be 1 and and console 2 would be 0.61, which would show that one console is topping the charts while the other is selling at a steady pace, but still at a lower rate than the other.

Now let's say console 2 have a 3rd sku appearing on the chart and grabbing the fourth spot, while the other console is still a the same spots. Console 2 new average will now be of 6 and at a ratio of 0.66 to the leader. The gap is lowering just because we have an other option and selling at the same steady pace as the other before.

Now let's say that the new sku is actually moving up in the chart and even moving to the 3rd spot. Console 2 is now holding spots 3, 5 and 6 for a new average 6.3 and console 1 is holding spots 1 and 4 for a new average of 8.5. Console 2 is now at a ratio 0.75, thus showing a spike in sales.

Is this method perfect ? No, it's still speculations.

But I believe it could help in predicting not only the ranks, but also the ratios between them which is exactly the idea behind the tool.

Tool was solid accurate before, and Switch making so huge difference compared to Switch looked like sure victory for month based on previous accuracy of tool, so you can't really blame people.

On other hand you start trashing tool on another forum saying "They are a waste of time lately trying to predict NPD and I'm not sure why they're taken seriously" and calling its "extremely biased towards Nintendo". But that really just tells more about you than anything else.

We're on vgchartz last time I checked. For all we know that poster on another forum is imitating me. So don't get your jimmies rustled over a post made on another forum. 

So you're staying switch October win looked certain? Jesus christ I'd have put my house on an easy ps4 win. Bo4 + rdr2 + spiderman effect it was never going to lose. Especially when the ps4 is already the most popular console this year.

You see this is the negative thing about the tool I was talking about. People get too attached to it after it had a good run Predicting easier months but then comes the first one where reality is screaming at a ps4 win but people can't see it because the tool says otherwise. This is why i didn't bother debating who would win and just sat back and waited for the results.