By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Sales Discussion - In progress NPD predicting tool

AntonTeiosanu on 29 April 2018
Symbios63 said:
So, going forward, here is a detailed explanation on the changes to come and why I do them :

As mentionned in a previous post :

March Overview (Only 8 days of data) :

Amazon :
PS4 : 99.1 ; XBO : 97.2 ; NS : 155.6 ; 3DS : 0
Best Buy :
PS4 : 184.2 ; XBO : 221.9 ; NS : 163.0 ; 3DS : 6.5
Gamestop :
PS4 : 153.8 ; XBO : 142.8 ; NS : 80.2 ; 3DS : 29.6
Walmart :
PS4 : 261.2 ; XBO : 375.4 ; NS : 168.5 ; 3DS : 45.5
Target :
PS4 : 179.9 ; XBO : 177.4 ; NS : 189.9 ; 3DS : 38.0

Overall :

1 - XBO : 961.6 points/day (1.0)
2 - PS4 832.5 points/day (0.87)
3 - NS : 727.1 points/day (0.76)

NPD results :

1 - PS4 : 354k (1.18)
2 - NS : 308k (1.03)
3 - XBO : 299k (1.0)

So we were dead wrong here. While doing the updates, I noticed two lists out of the 5 are not suited for what we try to do : Walmart list is really strange with items going in and out all the time and ranking not really moving as it should. Target list is more easy to understand but ranking takes into account a long period of time.

So what if I remove them ?

Overall :

1 - XBO : 408.4 points/day (1.0)
2 - PS4 379.5 points/day (0.93)
3 - NS : 368.7 points/day (0.90)

Slightly better, the artifactual lead on XBO is less, but we are still far from the truth...

XBO get a lot of point by having many SKU ranking in top 100, even if not very high on the list. So I changed the point system to award more point to high ranking items (#100 = 1 point, #51 = 50 points, #49 = 52 points, #21 = 110 points, #20 = 115 points, #1 = 210 points) :

1 - NS : 551.7 points/day (1.0)
2 - PS4 519.6 points/day (0.94)
3 - XBO : 506.5 points/day (0.92)

Ok, so indeed, XBO lead is gone but now switch is first... I went with the idea that switch might be too high on amazon and best buy and decided to weight retailers as follow : Gamestop : 1, Best Buy : 0.2, Amazon, 0.5.

1 - PS4 : 237.3 points/day (1.12)
2 - NS 216.9 points/day (1.02)
3 - XBO : 212.6 points/day (1.0)

This does look a lot better !

However, by focusing on the only 5 days for which I have all 3 retailers (I have no gamestop numbers for first 3 days...) the "working weighting" becomes : Gamestop : 1, Best Buy : 0.2, Amazon, 0.8 :

1 - PS4 : 363.6 points/day (1.15)
2 - NS 326 points/day (1.03)
3 - XBO : 315.2 points/day (1.0)

So this is probably how I will go from now on. I will post soon to give April summary with both old and new system...
If you have ideas/suggestions...

Oh my god dude you are such a genius



Around the Network
AntonTeiosanu on 29 April 2018

Wow you are really good at what you are doing



Pionner on 29 April 2018

Damn Symbios. Very intelligent idea with the changes. I usually only look at Amazon, GS and Bestbuy. WM and TG are too random. So there you go.



Formerly ilovegirls69  :(

chakkra on 29 April 2018

Wow, congrats man.  You{re doing an awesome job.



Cobretti2 on 29 April 2018

interesting thread will read all later



 

 

Around the Network
Bofferbrauer2 on 29 April 2018

Sounds great.

Bestbuy at only 0.2, isn't that maybe a bit too low?



The Nintendo eShop rating Thread: http://gamrconnect.vgchartz.com/thread.php?id=237454 List as Google Doc:  https://docs.google.com/spreadsheets/d/1aW2hXQT1TheElVS7z-F3pP-7nbqdrDqWNTxl6JoJWBY/edit?usp=sharing

The Steam/GOG key gifting thread: https://gamrconnect.vgchartz.com/thread/242024/the-steamgog-key-gifting-thread/1/ 

Free Pc Games thread: https://gamrconnect.vgchartz.com/thread/248138/free-pc-games/1/

KilleyMc on 29 April 2018

Awesome job man, can't wait to see the result now that you will have a full month of tracking.



Madword on 29 April 2018

I echo what others are saying, fantastic job, this is exactly what needed to be done to make things more clear and more comprehensive. It may be wrong a few times but as you get more data it should in theory make it a lot more accurate.

Again, thanks for the efforts :)



Making an indie game : Dead of Day!

globalisateur on 29 April 2018
Symbios63 said:
So, going forward, here is a detailed explanation on the changes to come and why I do them :

As mentionned in a previous post :

March Overview (Only 8 days of data) :

Amazon :
PS4 : 99.1 ; XBO : 97.2 ; NS : 155.6 ; 3DS : 0
Best Buy :
PS4 : 184.2 ; XBO : 221.9 ; NS : 163.0 ; 3DS : 6.5
Gamestop :
PS4 : 153.8 ; XBO : 142.8 ; NS : 80.2 ; 3DS : 29.6
Walmart :
PS4 : 261.2 ; XBO : 375.4 ; NS : 168.5 ; 3DS : 45.5
Target :
PS4 : 179.9 ; XBO : 177.4 ; NS : 189.9 ; 3DS : 38.0

Overall :

1 - XBO : 961.6 points/day (1.0)
2 - PS4 832.5 points/day (0.87)
3 - NS : 727.1 points/day (0.76)

NPD results :

1 - PS4 : 354k (1.18)
2 - NS : 308k (1.03)
3 - XBO : 299k (1.0)

So we were dead wrong here. While doing the updates, I noticed two lists out of the 5 are not suited for what we try to do : Walmart list is really strange with items going in and out all the time and ranking not really moving as it should. Target list is more easy to understand but ranking takes into account a long period of time.

So what if I remove them ?

Overall :

1 - XBO : 408.4 points/day (1.0)
2 - PS4 379.5 points/day (0.93)
3 - NS : 368.7 points/day (0.90)

Slightly better, the artifactual lead on XBO is less, but we are still far from the truth...

XBO get a lot of point by having many SKU ranking in top 100, even if not very high on the list. So I changed the point system to award more point to high ranking items (#100 = 1 point, #51 = 50 points, #49 = 52 points, #21 = 110 points, #20 = 115 points, #1 = 210 points) :

1 - NS : 551.7 points/day (1.0)
2 - PS4 519.6 points/day (0.94)
3 - XBO : 506.5 points/day (0.92)

Ok, so indeed, XBO lead is gone but now switch is first... I went with the idea that switch might be too high on amazon and best buy and decided to weight retailers as follow : Gamestop : 1, Best Buy : 0.2, Amazon, 0.5.

1 - PS4 : 237.3 points/day (1.12)
2 - NS 216.9 points/day (1.02)
3 - XBO : 212.6 points/day (1.0)

This does look a lot better !

However, by focusing on the only 5 days for which I have all 3 retailers (I have no gamestop numbers for first 3 days...) the "working weighting" becomes : Gamestop : 1, Best Buy : 0.2, Amazon, 0.8 :

1 - PS4 : 363.6 points/day (1.15)
2 - NS 326 points/day (1.03)
3 - XBO : 315.2 points/day (1.0)

So this is probably how I will go from now on. I will post soon to give April summary with both old and new system...
If you have ideas/suggestions...

Great job ! My suggestion:

- This year Amazon is often sold out for PS4 slims and or Pro SKUs. For instance currently only the Pro BF2 bundle is available on Amazon. You should take that into account by greatly reducing the amazon weight. Also it will improve Switch ranking cause Switch is almost always the best selling console on Amazon oddly.



Symbios63 on 29 April 2018
Bofferbrauer2 said:
Sounds great.

Bestbuy at only 0.2, isn't that maybe a bit too low?

It might, we only have few days of data for march which may lead to mistakes. I will adjust again next month if needed.

globalisateur said:

Great job ! My suggestion:

- This year Amazon is often sold out for PS4 slims and or Pro SKUs. For instance currently only the Pro BF2 bundle is available on Amazon. You should take that into account by greatly reducing the amazon weight. Also it will improve Switch ranking cause Switch is almost always the best selling console on Amazon oddly.

Yeah, out of stock items might be a thing to consider in the future, but it is not that easy to quantify. Let's wait and see the results for April before making other changes.

 

And thank you everyone for the encouragement.