🛑Considerations

When Pair-trading, there are a number of factors to consider, including:

• Choosing which asset to long

• Choosing which asset to short

• Selecting leverage

• Entry level and when to take profit

• Liquidation levels

• Fundamental analysis

• Technical analysis

More advanced factors to consider include:

• Net funding costs

• Slippage

• Correlation between the two assets

• Rebalancing the long and short exposure (beta)

• and many others...

Let’s take a look at some of these in more detail. Firstly, with regards to choosing which asset to long and which to short. It should be common sense, traders want to long the asset they think will outperform better, and short the one they think will underperform.

For example, if a trader believes that $ETH will outperform $LINK , the way to express this is to long $ETH / $LINK as the pair trade.

The chart will automatically populate and show traders how this spread/pair has evolved over time:

Pear is a permissionless platform, meaning users can select any combination of longs and shorts subject to available liquidity. This is incredibly powerful as the number of permutation scales very quickly with new assets added.

3 underlying assets available = 6 possible combinations of pairs to trade

We can generalize this using mathematics:

4 underlying assets available = 12 different permutations

20 underlying assets available = 380 permutations! 🤯

So clearly, there are a lot of different pairs (pears?) to pick from.

Once a user has selected their preferred assets, they need to consider the entry level and whether they consider this a ‘good’ entry. Pear Protocol will never give financial advice and just like how users have permissionless power to choose any combination of assets, they must also exercise discretion as to the whether they think the pair is at an attractive level. In the example above, the LINK/ETH chart is the inverse of the downward trending ETH/LINK charts.

Spread prices as of 17/10/2023 for reference:

ETH/LINK: 212

LINK/ETH: 0.00472

These levels are calculated from ETH ($1586) and LINK ($7.48) respectively. The prices are a calculation of the spread, there is no direct oracle available for these. For example, for the calculation of ETH/LINK , the full calculation is coming from:

ETHUSDT/LINKUSDT = 1586 / 7.48 = 212

Please note, for ETH/LINK to go from 212 to say 233.2 (+10%), there are a multiple ways this can happen!

For example, ETH could stay flat at 1586, but LINK could move downwards to 6.8. The spread would now go from 212 to 1586/6.8 = 233.2, resulting in a +10% gain for the user (excluding leverage).

Alternatively, ETH could go up to 1800, and LINK could move also move up to 7.72. The spread of the two assets would be 1800/7.72 = 233.2, also resulting in a +10% gain for the user (excluding leverage)

Lastly, ETH could go down to 1500, and LINK could move down to 6.43. The spread of the two assets would be 1500/6.43 = 233.2, also resulting in a +10% gain for the user (excluding leverage).

It’s this dynamic, where traders can make money regardless of whether the broader market goes up, down or sideways that makes pair-trading so powerful. The only thing that is important is that users are right on their narrative, (i.e. that Asset X will outperform Asset Y).

Even better, in the example above, with 10x leverage, the user would have made a +10% * 10 = +100% gain on a pseudo-market neutral trade.

However, given this path dependence, traders must choose leverage carefully. Because traders are not in control of the path that the two assets may take to get to a gain, and the pair trade consists of one long perp and one short perp, users must carefully consider avoiding being liquidated on any one of the legs (in Isolated Mode).

In the example above where ETH went up to 1800 (and the user was long), the gains there offset the losses on the LINK move to 7.72. However, the trader shorted LINK at 7.48 (LINK went up +3.2%). But if the trader had done this trade with 50x leverage on Isolated Mode, they would have been liquidated on the short leg.

This is one of the disadvantages of isolated margin product, which sources liquidity from GMXv2, since all their positions have isolated margin, meaning collateral from the winning position (the long) cannot be used to post more margin against the losing leg (the short).

On the other hand, integrations with Vertex and SYMM natively enable cross-margining, meaning users can lever up more comfortably. For those platforms, monitoring Account Health and how much the trader has in their margin account before Partial/Full liquidations is important. Given this path dependence of both assets, we can only ever give an 'estimated' liquidation level.

A quick note on some of the other advanced factors:

Net Funding

Traders will likely pay funding on one leg, and receive funding on the other leg. We have provided the ability to dynamically observe the net funding rates directly on the platform when you chart any 2 pairs.

Slippage

Sometimes one asset will have more liquidity than another leg. Users must consider slippage when entering and exiting a trade, especially once you are trading significant size outside of the top 20 cryptocurrencies.

Correlation Between Two Assets

Pair trading has a rich academic history, grounded in statistics. Generally speaking, people look to trade two assets that are closely correlated but where they either expect some deviation between the two assets, or a mean reversion of some sort. Pear is working on tooling to analyze correlation between the offered pairs.

Rebalancing

Rebalancing the long and short exposure (beta) - Each trade starts off ‘dollar-neutral’ at inception, meaning if a trader deposits $100 of collateral, and have 10x leverage, then they will have total exposure of $1000 (before fees). Of this, $500 will be long exposure, and $500 will be short exposure.

As the trade goes in the trader's favor, the collateral value of the long will rise to say $550, whilst the collateral value of the short may fall to $480 (resulting in a +$50 - $20 = +$30 gain) for the user.

At this stage, the user is net long the market ($550 vs $480). To rebalance their ‘beta’, an advanced trader would take some profit on the long leg and add it to the short leg, but for most traders this will not be a consideration and they’ll just look to close both trades outright at the same time.

Pear is working on the ability to flexibly choose what % you are long and short at inception (e.g. 60%/40% vs. 50%/50%), as well as the ability to rebalance trades more easily.