Fun With Numbers: Amazon Tracking vs. Sales (12/21-27, 2015)

When I was wrapping up last week’s post, I thought the delays in the Nash database on account of the holiday season were almost over. Turns out, after the BD data update covered in this post on February 3rd, the corresponding DVD update took 15 days. Hoo boy. Maybe there were errors in processing that took time to clean up, or getting every retailer under their umbrella to summarize reports on bargain bin DVDs liquidated this week just took that long. Either way, it’s here, looking much the same as it has in previous weeks.

Raw Data:




Formula Recap:

#10k-#40k: S=10-(R-10000)*(9/30000)
#2k-#10k: S=30-(R-2000)*(20/8000)
#1k-#2k: S=70-(R-1000)*(40/1000)
#500-#1k: S=220-(R-500)*(150/500)
#200-#500: S=250-(R-200)*(30/300)
#150-#200: S=800-(R-150)*(550/50)
#100-#150: S=1000-(R-100)*(200/50)
#50-#100: S=1300-(R-50)*(300/50)


-DBZ:F DVDs are now outranking the limited edition versions counted in the same total, likely the cheaper-price edition taking over on account of time and late-holiday gift buying. In response to this, I switch to using the DVD rank datapoint for the chart above comparing average ranks and sales totals.

-Since the formula’s accuracy is kind of a main goal of this project, I’m continuing last week’s exercise where I compared formula predictions with actual totals. Apologies for the lack of internal consistency, but I’m swapping chart formats again in a minor way – the chart comparing last week’s formula with sales used average dailies. I’m switching them to weeklies now because doing so doesn’t involve an unnecessary set of division (weekly totals are weekly by default, expected totals are the sum of seven days’ worth of data run through the formula above). Additionally, I’m including the VHD and AoT alternate editions’ estimates, just to show what they are.

As before, DVD estimates (3x the BD formula) are in grey and BD estimates are in black:


Of the 27 applicable datapoints, only two releases see an estimate less than half or more than double the actual value. These two are One Piece: Strong World on DVD (overestimated) and Arietty on DVD (underestimated). DBZ:F on BD, which was severely underestimated last week, has a total slightly more than half the actual value. Hal, which showed a near-perfect estimate-actual match last week, is a bit underestimated as well. Every other expected value is at least reasonably close, and some (Ponyo, Naruto: The Last on BD) hit the nail on the head. Another fairly strong week showing a fallible formula, but one with generally strong predictive power.

Get ready for a bunch more posts with this bare-bones style – there’s one interesting thing that happens to these releases in the whole month of January 2016 (Marnie’s Oscar nomination), and who knows how long updates will take to get to February and the much-more-interesting Eva 3.3333333 launch.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s