Major Portfolio Update: Part 1

The market has changed, and the crypto portfolio has to change with it.

The Yen Carry Trade will either continue to unwind (pain) or the central bankers will respond (money printer go brrrrrr). Either way, we’re entering a new paradigm.

We’re going to do it in two parts. Part 1, today, is about identifying the tokens that handled the Yen unwind crash the best. We’re going to use some simple math to narrow our focus, shorten the list, and prepare ourselves to make trades later this week.

The narratives of tomorrow are being shaped by the price action of today. I want to know which tokens crashed the least, which recovered the fastest, and what kind of volatility popped off along the way. I want to build an intuition for what the market likes tomorrow, not yesterday.

Summer is fast ending. We need to be ready. Part 2 will go through portfolio changes and rebalancing — I’ll evaluate the portfolio broadly, incorporate what we’ve learned today, and start making trades.

There’s a decent bit of spreadsheets and data and high school math in this post. Skip to the end if long division isn’t your thing.

Spoiler alert — SOL gud. BTC gud. ETH…the ethereum ecosystem had a pretty shit week.

Strength is never a weakness

The goal of this analysis is to identify the crypto tokens that crashed the least and recovered the most. I want to prioritize crashing the least slightly higher, because a smaller draw down favors greater compounding growth over time. Remember, if the price falls 50%, then it must then recover 100% to get back to where it was. That’s a big bounce just to get to breakeven.

Here’s the plan:

• Analyze 100 of the largest crypto tokens by market cap. This ensures each token has plenty of liquidity and is less susceptible to market manipulation/shenanigans.

• Evaluate their price decline from Wednesday July 31, just before the BOJ raised rates, to the bottom on Monday morning, August 5. I chose 8am EDT — close enough for horseshoes and hand-grenades.

• Evaluate the bounce from the bottom for each, from August 5 to end of day Aug 8 (I ran these numbers on Aug 9, thus arbitrary date).

• Use a little math to assess each token’s resistance to decline and ability to bounce back. I’ve come up with 5 metrics; more detail below.

• Develop a Combined Strength Score that better includes the nuance of the price action than a simple weekly return. For each token, I’ll use z-scores to normalize all 5 metrics and sum them up.

• Once we have the metrics sorted, we can look at the best performers fundamentals, narratives, and tokenomics, and then buy and sell. (coming in part 2, later this week)

Grab your calculators, nerds

All the data used in this analysis was sourced from the CryptoCompare API, powered by CCDATA. Their large market cap list is different than CoinGecko’s. I’m not sure what’s going on there, but worth noting as an issue I will need to revisit.

Each metric uses prices and returns at July 31, August 5, and August 8 to isolate the time around crash and recovery.

There’s nothing special about the metrics below — I just made them up in my attempt to analyze the price action from different perspectives. This is definitely not rigorous, probably not predictive, definitely not guaranteed, and most definitely not financial advice. I would not recommend using any of these for any other situations.

(skip the formulas section and jump to the conclusions)

1. Recovery Strength Index (RSI)

This metric combines both the resistance during the downturn and the strength of the recovery.

A higher RSI value indicates a strong recovery relative to the decline. If a token barely falls and then recovers strongly, it will have a high RSI. Tokens that fell significantly but didn't recover well will have a low RSI.

2. Performance Composite Index (PCI)

This metric considers both the resistance to the initial decline and the strength of the recovery, assigning equal weight to both.

This index combines the effects of both periods. A token that resisted the initial drop and then recovered will have a PCI greater than 1. Less than 1 means the price has not recovered to its original level.

3. Recovery to Resistance Ratio (RRR)

This metric gives a more straightforward comparison by subtracting the percentage decline from the percentage recovery.

The bigger the RRR value the better. Pretty straightforward (but remember, a 50% decline requires a 100% recovery to breakeven. So an RRR of 0 does not mean breakeven).

4. Weighted Performance Index (WPI)

None of the metrics so far give extra importance to strong performance during crash itself. The Weighted Performance Index (WPI) is my attempt (the PCI is close, but gives equal weight).

First, we normalize the decline and the recovery sides of the crash, give the decline a bigger weight (because we want to prioritize smaller draw downs for more compounding) and then calculate the Weighted Performance Index (WPI). I chose 70% weight to the decline (α), 30% to the recovery (β).

To normalize the crash we subtract 1 from the percentage decline. The smaller the decline, closer to 1 (and therefore higher). To normalize the bounce, I divided by 100%. This is an arbitrary choice. The maximum observed recovery in the data set is barely over 50% — plenty of room to spare.

5. Combined Volatility and Recovery Score (CVRS)

This metric is a simple way to include volatility in a metric and emphasis the intensity of the recovery. Divide the return during the market bounce by the total volatility experienced during both periods.

This score ranges from 0 to 1, where a score closer to 1 indicates a strong recovery relative to the overall volatility (both decline and recovery combined). A high CVRS indicates a token that not only recovered but did so relative to its total movement (both down and up).

Combined Strength Score - 1 Metric to Rule Them All

The final metric brings together the other 5 into a single score for each token. This method normalizes each metric to account for different scales and then sums the normalized scores.

Each of the Z’s in that formula are calculated by subtracting the mean and dividing by the standard deviation of each metric.

Results — Here are the tokens to watch

Here’s the data in a Google Sheet if you want to play with it.

Many of the tokens at the top of the list are stablecoins (USDC, USDE, FDUSD, etc), and some are tokens associated with centralized entities that I’m not interested in investing (ie. LEO —> Bitfinex). If I have time I’ll update this to exclude these. I don’t think that change will impact the results enough to matter.

And there are others that I don’t know anything about — FTN is a new one to me (their twitter has 200k+ followers but is only getting 2k impressions per post… sus ¯\_(ツ)_/¯ ).

Here’s the household names and a few others I’ve written about previously:

Of the big names, SOL and BTC continue to lead the way. Solana in particular performed incredibly well. It remains my top pick across all coins.

Ethereum and the ethereum ecosystem took a beating this week. The bottom of these rankings are full of major ethereum projects. LDO, MKR, and ENS are all core projects in the ETH ecosystem, and all three took punishing drawdowns with little to no recovery. Even among the memecoins ETH struggled. ETH-native PEPE scored the worst of all memes on the list.

The price action, momentum, and sentiment/narrative are all very negative everything ethereum right now (and this is despite the ETF launch). I don’t know if that means sell everything you’ve got or buy it all as a contrarian take. The ETF seems to be going well too. Wild times.

WIF was punished in the rankings for it’s wild volatility, yet still landed in the top quartile of tokens (excluding the stables, etc). It’s price fell by the most of any coin on this list, and it also posted the highest percentage return on the rebound. No other coin fell farther or climbed higher. If you can stomach the vol, WIF as a Solana beta play remains very attractive to me.

AAVE and XRP stand out as strong performers, each scoring well on the decline and the rebound. XRP’s price was boosted by a favorable outcome in the courts last week; hard to say how it would have performed without that confounder.

AAVE is an OG DeFi protocol that’s like a decentralized bank. It’s as battle tested as they come, generates lots of real revenue, and there are no token unlocks to worry about since the protocol has been around for several cycles. The price however has performed unremarkably for ages - like another OG coin high up on the list, XLM.

Portfolio Update

No updates this week (yet). Next newsletter we will be doing a comprehensive review and update. Expect several sizable changes in positions as I rebalance for today’s new market conditions.

Based on today’s results, the plan for part 2 is to investigate and consider AAVE, XRP, SUI, and FLR with others possible as I expand the analysis to include fundamentals and longer term momentum.

Market Vibes