PIP 10: PieDAO DeFi++ Index

Following up on the discussion kicked-off during the last community call and the multiple inputs collected from the community, we’d like to formalize a proposal for a PieDAO DeFi++ Index, representing a balanced allocation of relevant assets within the DeFi ecosystem.

In order to cope with the current Balancer limit of 8 assets/pool, the suggested approach would include sub-indexes based on assets’ current Market Capitalisation.

The proposed DeFi++ Index at launch would include a balanced allocation of the followings, weighted in as per reached consensus (e.g 70/30, 50/50):

  • LargeCaps (assets with current total Cap >= $200M)
  • SmallCaps (assets with current total Cap < $200M)

Note: This approach would be flexible enough to allow the inclusion of additional sub-indexes in due course, like a “discovery” MicroCaps, for instance, to be easily weighted in the DeFi++ Index.

Below a summary of assets identified to be part of the respective indexes, based on both their fundamentals, the general community sentiment and more specific inputs collected

image

LargeCaps

Assets represented within the LargeCaps have been weighted in as a function of their current Market Capitalization, capped by a max weight of 25%.
-> Currently 6 assets identified, leaving room to the potential inclusion of 2 additional LargeCaps in the future.

LargeCaps Index would be initially priced at $1,000.
Rebalancing of underlying assets would take place as per agreed frequency (e.g. monthly, quarterly).

image

SmallCaps

Assets represented within the SmallCaps have been weighted in as a function of their current Market Capitalization, capped by a max weight of 30%.
-> Currently 4 assets identified, purposely leaving room for the future inclusion of up to 4 additional SmallCaps, given the development pace of the DeFi ecosystem.

SmallCaps Index would be initially priced at $1,000.
Rebalancing of assets would again take place as per agreed frequency (e.g. monthly, quarterly).

image

DeFi++ Index

The DeFi++ Index would be defined as a weighted composition of both LargeCaps & SmallCaps indexes, with weights defined as per Governance input in order to factor in the overall market outlook, the risk propensity, etc.

DeFi++ Index would be initially priced at $ 1,000, and trading as a function of both the LargeCaps & SmallCaps indexes.
The DeFi++ token would be pegged to 1/1000 of the DeFi++ Index.

You can find here the full spreadsheet.

Below a visual example of Index performance over the past few weeks, assuming a weighted allocation 70/30 between LargeCaps & SmallCaps.

image

image

image

image

5 Likes

Do you have any proposals for the token symbols and names?

What was the starting date of the backtesting performance 6/24 or earlier?

Some ideas
DSC / DLC
DEFISC / DEFILC
DTOP / DMID / DLOW
DLARGE / DSML

I also would like to find some semantic name.

I think we should stick to the ++ terminology.

1 Like

Backtesting started as of 06/23, since BAL data wasn’t there before

For the one including SM and LC I totally agree, how would you call the sub pie tho?

A bit tricky…maybe:

MacroDeFi++
MiniDeFi++
MicroDeFi++

2 Likes

FWIW, I don’t think backtesting from 06/24 means a lot. In fact, I would personally say this is a completely useless metric. My reasons for thinking this are:

  • Crypto assets are highly volatile and performance over a few weeks means very little.
  • We are in the middle of a very clear DeFi bubble so current performance likely won’t reflect future mid to long term performance.

As a DAO made of people who have a good understanding of the space, the choice should be focused on fundamental value investing imho, so these backtesting graphs should beat no significance on the choices we make.

While I am here, let me repost something I said on the Discord regarding a potential DeFi++ index.

This will probably be for later down the line, but i think DeFi++ should be more flexible than just a Balancer pool. I’m imagining a smart contract that would hold a balance of different DeFi tokens, and dynamically put them to work on their native protocols.

For example, if KNC is part of the Pie, the DAO would stake it on Kyber, if REP is part of the Pie, it should be ready to be used to dispute outcomes in the Augur protocol Augur, if COMP is part of the Pie, it should be used to vote on proposals, etc. This better expresses the full spirit of a DAO owned DeFi index fund imo, as opposed to “just” putting it in a Balancer pool – not to mention the possibility of impermanent loss.

I don’t know what this entails at a technical level, but it sure sounds interesting to look into.

3 Likes

Thanks for the feedback. As already pointed out on Discord I definitely agree on the fact that current performance tend to be a useless and potentially dangerous metric to consider for the selection of assets to include in the DeFi++ index: within the above proposal, assets included were exclusively selected based on their fundamentals.

My bad for improperly using the term backtesting since in fact no optimization analysis was performed (i.e graphs were provided as a mere example of what the index tracking may look like from a hyphotetical t0 when index is established).

Appreciate your suggestion. Reminds me of what we’ve been referring to as DeFi Governance Pie.
Definitely worth exploring more down the road. The modular structure currently proposed for the DeFi++ Index would potentially allow its inclusion as a ‘sub-index’ (next to LargeCaps, SmallCaps, or MicroCaps - for the discovery of emerging DeFi projects with strong fundamentals).

As addendum to the above proposal, we’ve been running some analysis to compute the scalability of the DeFi++ index at growing liquidity levels.

Results are summarized in the heat map below, showing for each underlying asset the % of its Total MarketCap that would be hypothetically absorbed by the DeFi++ index.

At current values, even by projecting the total liquidity of the DeFi++ index to $ 100M, there would be no major issue for any of the assets, with only 1 SmallCap reaching 10% of its Total MCap as value locked in the index.

NOTE: the analysis considers current fixed prices for all assets, so in reality the % of MCap locked would anyhow result sensibily lower when taking in consideration the pressure on asset’s price resulting by the growing demand triggered by the Index growth.

2 Likes

VOTE #202 for PIP-10 DEFI++

https://mainnet.aragon.org/?#/piedao/0x109b588a4f2a234e302c722f91fe42c5ab828a32/vote/202/

https://gateway.pinata.cloud/ipfs/QmQu5RYHqL94mGzLdifuQE4X7yG9SrmBafvnweJrbLXtwr

3 Likes

It’s been 21 days since the original allocation was proposed. Should we start discussing any new entries? What’s our criteria for determining which tokens go into the portfolio? I know that ‘strong fundamentals’ tends to be vague and subjective criteria so am looking to formalize this process more.

1 Like

Let’s put together an analysis framework template to use it as a reference, a great examples could be something like the Delphi research reports.

currently working on a more concise spreadsheet where new assets proposed could be assessed in their fundamentals, to serve as basis for further community discussions

1 Like

Rebalance Proposal

This proposal changes the model calculation related to the market cap and adds 1 token for DEFI+L and 2 tokens for DEFI+S for a total of 13 tokens exposure in DEFI++.

Model chance

The proposal suggests using the average 30-day value to calculate market up and smooth up weights, more suitable during euphoric markets.

Exposure rebalance

  1. Adding MLN from @MelonProtocol and PNT from @pTokens on DEFI+S
  2. Adding YFI to DEFI+L

The comparison of weights can be found in tables below and in the spreadsheet.

Screen Shot 2020-08-24 at 6.38.40 PM

Vote in the DAO.

1 Like

From the last proposed allocation for DEFI+L have been introduced a few new elements:

Sentiment Score
A sentiment score was introduced to assess for all underlying assets of the Pie what’s the community perspective on the 3 following pillars:

  1. Innovation (Solution & Roadmap/Timeline)
  2. Functionality to the DeFi Ecosystem (Uniqueness & Composability)
  3. Growth Potential (P/E Ratio, Volume & Outstanding Supply)


Each of the above items is graded on a scale from 1 to 12, and the resulting Sentiment Score for each asset contributes as per a defined weight (now = 20%) to the definition of the total allocation of the asset in the Index.

Changes in underlying assets

  1. Swapped LEND for AAVE. Given the ongoing migration from one token to the other, we are going to effectively track the price of AAVE in the index, while temporarily still referring to the 30days MCap and Supply of LEND for the sake of dimensioning this asset allocation.
  2. Introduced UNI - Uniswap
  3. Removed KNC - Kyber Network, since its MCap felt out of the range considered for Large Caps.

DEFI+L Pie Allocation (8)


We’d therefore like to move further on with a final proposal for the allocation of our DEFI+L Large Caps index: all numbers can be explored in detail on this spreadsheet https://docs.google.com/spreadsheets/d/1kbpy3vpYjx6grnpeSxFu6CdCNt7HQvq2pOXlpNvBW_A/edit?usp=sharing

Resulting DEFI++
According to the above allocation of DEFI+L, it could be eventually minted also her Majesty the DEFI++ Pie with the resulting allocation below:

Screen Shot 2020-10-05 at 17.15.08

DEFI++ Assets (4)

3 Likes

This is iteration seems great, just to clarify this would be with trading enabled with 0.1% fees correct?

When is this fee charged? Every tx made by DEFI++ or minting/burning DEFI++?
And I assume, fee is going into PieDAO vault?

1 Like

I meant swap fees in my prev message but we can include streaming fees already on TVL.

Streaming fees would be annualized fees which can be charged at any time (0.1 fees called after 6 months will charge 0.05 of TVL in the pie) by anyone and go directly to the DAO.

Info: https://medium.com/piedao/piedao-smart-pools-v2-balancer-grant-6bcefda47649

1 Like