> ## Bounty Question: Create an analysis on NBA Top Shots moments and attempt to uncover any correlations between a specific category and sales volume. This week, focus specifically on the play_type column in the flow.core.dim_topshot_metadata table. What play_types do users hold most, and how much volume in sales do these play_types generate? Include any other transactional metrics you can think of.

> # Introduction

Moment™ NFTs are officially licensed NBA collectibles that celebrate the game’s epic highlights from the most incredible basketball stars. They are the core parts of NBA Top Shot, capturing some of the best plays from across the NBA.

What separates them from regular highlights? Each one is made unique through NFTs on the blockchain. In other words, your Moments are your Moments.

Moments include exclusive collectible details of your favorite players like:

• On-Court Video Highlight

• Guaranteed Authenticity by the NBA and NBPA

• Moment Type, Tier, Series, and Serial Number

• Highlight and Player Stats

• Badges

  \

[Refrence: NBA Top Shot Support]

Source

> # Methodology

The following two tables were used to check the basis of the answer to this question:

> flow.core.dim_topshot_metadata > > flow.core.fact_events

flow.core.dim_topshot_metadata table:

In this table, each row corresponds to a unique NFT_ID from the Top Shot collection. Each of these NFT_IDs represents a Moment and each of these Moments has a play_type among 9 play_types. This table was used to determine how many of these moments have what play_type.

flow.core.fact_events table:

This table was used to determine how many of these Moments have been minted so far. For this purpose, transactions that met the following conditions were filtered as Moments minting transactions:

> EVENT_CONTRACT = 'A.0b2a3299cc857e29.TopShot' > > EVENT_TYPE = 'MomentMinted'

The point in these two tables is that there are a number of Moments in each of these tables that are not in the other table. The number of each of these absences was also obtained. Of course, the following reasons can be put forward for these absences:

• Moments that are present in the dim_topshot_metadata table and absent in the fact_events table can be due to two reasons:

> > 1. ==The time limit== of the data in the fact_events table (data is available from April 22 to July 8) > 2. Moments are ==not minted==

• Moments that are present in the fact_events table and absent in the dim_topshot_metadata table can be attributed to the ==incompleteness== of the data in the dim_topshot_metadata table.

In the following, the data of the fact_events table was considered as a reference to answer the rest of the questions. But only data that existed in both fact_events and dim_topshot_metadata tables was used. Moment minters were assumed to be holders and it was checked how many of them sold their Moment. Sellers are removed from the list of holders (even if they bought one of the Moments in the days after the sale of their own minted Moment and still have it).

\

• There are about 1.06 million Moments in the dim_topshot_metadata table, while the number of minted moments in the fact_event table is about 1.043 million==.

  \

• But the noteworthy point is the high number of missing moments of each table in the other table, which is a very significant figure (more than 10% of all moments) and can associate the accuracy of the obtained results with a certain percentage of error.

  \

Considering that the fact_event table was included as a reference table and moments were checked that were in both tables (dim_topshot_metadata and fact_events), the number of moments is less compared to both mentioned tables (less than one million).

• Another point is that out of the 9 play_types in the dim_topshot_metadata table, only 7 of them could be found in the fact_event table, and the other two (which were among the least popular paly_types in the dim_topshot_metadata table itself) are not present in these results.

• Rim ranks first among these 7 play_types with absolute authority in all parameters. In terms of the number of mints, the number of holds, and the number of sales, this play_type holds the first place with more than 40% of the shares.

  \

• Rim in all three parameters is at least three times larger than the second rank (i.e. 3 Pointer).

  \

• The interesting thing about these pie charts is that the proportions and percentages have not changed much in all three charts and even the first to seventh ranks are the same in all three charts. (except for the shift of the fourth and fifth ranks between Mid Range and Block in sales volume)

\

• In the daily review of the number of mints, it can be seen that most of the mints were done in a limited number of days. 7 days can be considered as the main days of mint moments, which are the number of mints of these days in order:

> > 1. Fourth of May > 2. The twentieth of May > 3. The 31st of May > 4. Second of May > 5. Second of June > 6. May twenty-fifth > 7. The twenty-second of June

• Considering that out of the 7 main days, the first 4 days and a total of 5 days are related to the month of May, you can choose the month of May as the "Mint Moments Month".

  \

• In the number of sales, it can be seen that there was a great boom in this field in May, and after the first days of June, these busy days have given way to less busy days.

  \

• It can be seen in the number of sales that there was a very good boom in this field in May, and after the first days of June, these busy days have given way to less bright days.

  \

• The highest number of sales occurred on May 6. All the top 5 days in terms of sales are in this month. It might be better to name May as "the month of moments".

  \