Part 2: Whose the Auction Settler?

Does the winner settle the auction or does new bidder settle the current action to start the next auction?

To answer this question, we would need to take a closer look at the AuctionSettled transaction data:
By comparing tx_json:receipt:logs[4]:decoded:inputs:winner and from_address of the transaction, we can identify if the Auction Settler is a newBidder or the winner.

Looks like 64.7% of the time newBidder are the ones who are settling the current auction while winner settles the remaining 35.3%. Seems like new bidders are always excited to start the bidding on the next Nouns once the current auction is over.

After looking at a couple of AuctionSettled transaction, there were a couple of addresses that keeps popping up which prompted my interest in this graph.
This graph shows a count of the no. of times each address settled an auction and in what capacity (newBidder or winner) did they settle the auction in.

By comparing the above 2 graphs, we can see that for the top auction winner of 10 wins 0x696ed7b26f4b019ceec78dc8b9140ad64a6f354b, he (or she) only settled auctions 2/10 times (as the winner).
For the next winner 0xf8d24586f793fab81248f0b23f47e5b6bc0a6c9a, he only settled the auction 0/6 times he won. For the 3rd auction winner, he settled the auction 2/5 times he won and another 6 times as a new bidder.

Part 1: What are the valuable traits?

Comparing the metadata of NFTs and their respective mint or sale price on secondary markets, what are the more valuable traits?

Looking at Mints

The graph above plots the token price (in ETH) over tokenID (proportional to time).
It isn't surprising that the token #1 had the highest mint price of 613.37ETH.
The next top sale is #69 which is also expected too. Though I'm not too sure about the subsequent ones with the numbers 4 & 7...

At first glance, it does seems like the tokenID has the largest impact on the mint price.

Traits: Body

The graph below plots the probability of each trait occurring, its respective body ID# and the median price of that NFT.
Body IDs with high trait probability but yet high median price are deemed more valuable to the market as higher trait probability should correspond to a lower median price and not higher.

For Body, it looks like #2, #20, #27 have slightly better median prices than the rest

Trait: Accessory

A similar approach is also applied to the Accessory Trait. The results are rather even distributed with 2 exceptions:

1. Accessory #26 with an exceptionally high trait probability but slightly above average mint price
2. Accessory #95 which is likely due to Noun #1

Trait: Head

Similar to Accessory trait above, most of the points are evenly distributed except for 3 exception:

1. Accessory #90/#158/#186 seems to have similar if not higher median price as compared to the other data points.

Trait: Glasses

Even though the highest median here is trait 1, I think the more interesting traits are #18 and #19 as they have significantly higher probability of occurring but yet they have higher median price as compared to the other traits.

I do agree that there are some caveats with using Trait Probability as a way of determining the rarity of a trait, as Nouns are voted into existence (instead of generating randomly) but the statement that even though there are more of such trait in existence, the price is still higher than average still holds.

Looking at secondary market sales

Similar to what we did for the mints, we first look to see if there's any high sales for specific tokenID such as TokenID #40.

Trait: Body

Body #18 has the highest occurrence with an average sale price whereas Body #29 has the highest median sale price among all the other traits.

Trait: Head

Head #186 has changes hands the most times while the sales price remained average.
Head #216 has the highest sale price compared to the other traits.
Head #74 also has a comparatively higher sales price as compared to the other traits of similar rarity.

Trait: Glasses

Glasses #18 was sold for the most as compared to other traits. Glasses #7 were also sold for more as compared to the other traits of similar rarity as shown below.

Trait: Accessory

Accessory #56 have been sold the most times but its sale price remains average.
Accessory #47 is slightly rarer than #56 but it has been sold at a much higher price as compared to other traits of a similar rarity.
Accessory #136 was sold for the most out of all the accessory traits.