in

Why providing liquidity in AMMs is a sucker’s trade

In general, I would describe myself as being higher up on the risk curve – I’ll take more profit in return for more risk where I can. But somehow providing liquidity in AMMs always felt like a faustian trade to me. And until this day I couldn’t quite put my finger on it.

But then it hit me.

TL;DR; the problem with LP is that whichever way the market moves, you lose. If there’s any black swan event in the market whatsoever (positive or negative) – you lose. You’re setting yourself up to be a sucker.

Compare that to just hodling where you might buy a high-risk coin and, sure, you’re taking on the downside risk of it going to 0, but at least you’re keeping the upside of it going up 10x. Here you’re giving away the upside, while doubling your downside.

This argument is heavily reliant on Taleb’s black swan idea – which, in short, is that black swan events are way more common in the market than we humans like to think, and not accounting for them is how people get rekt.

Let’s play with some numbers using this [IL calculator](https://dailydefi.org/tools/impermanent-loss-calculator/):

Initial prices: $10 / $10 >> Future prices: $10 / $20

One token moves 2x faster than the other.

You lose 5% of what you could have made.

Doesn’t sound too bad, does it? Especially if the APR is 100%+.

But again, going back to Taleb, black swans are more common than we think. And this is crypto, which is an industry pretty much defined by black swans. So I wouldn’t be suprised to find out one morning that due to an exploit / liquidity attack / something else you never planned for happening one token is 10x the other.

Initial prices: $10 / $10 >> Future prices: $10 / $100

You lose 40% of what you could have gained.

Now the worst part – there’s actually 4 scenarios where this is possible and you’re exposed to all of them:

Initial prices: $10 / $10 >> Future prices: $10 / $100

Initial prices: $10 / $10 >> Future prices: $10 / $1

Initial prices: $10 / $10 >> Future prices: $100 / $10

Initial prices: $10 / $10 >> Future prices: $1 / $10

In another book of his, Antifragile, Taleb talks about fragile vs antifragile investing strategies.

Fragile:

* when you stand to lose all of your capital in return for small % return
* in other words, you keep exposure to negative black swans, but fail to capture positive black swans
* example: you put all your savings into “medium-risk” that promise you a measured 10-20% per year – but during the next financial meltdown you lose 60% of your portfolio.

Anti-fragile:

* when the downside risk is limited while the upside is near-infinite
* in other words, you keep exposure to positive black swans, but hedge exposure to negative blackswans
* example: you put 90% of your portfolio into a stable asset while putting 10% into ultra high risk investments that stand to make 100x if you’re right on them

LP to me is the definition of fragile.

Positive black swan happens – you lose money. Negative black swan happens – you lose money. Any black swan whatsoever happens – you lose money. Come to think of it, the only time you don’t lose money is when the market sits still (in relation to the two assets you hold) – which if you ask me is a tall order for an industry that basically didn’t exist a decade ago.

And all of this for a return of 10-100%.

I dunno about everyone else here, but to me this just feels assymetric in the wrong direction – limited upside, unlimited downside.

It feels like a sucker’s trade.

*Note 1: this is not taking into account 2000% apr shitcoin pools you get on pumps, nor stablecoin pools where there is no real risk of IL*

*Note 2: this also doesn’t consider other, just-as-relevant risks involved in LP, mainly smart contract / hacking risk which is definitely not 0 even for the most trusted platforms*

*Note 3: this became a bit of Taleb post. That was unintentional. I don’t agree with everything he writes, but I do subscrbe to the black swan philosophy, and the fragile/anti-fragile argument*

*Note 4: I might be just another sucker who doesn’t know what he’s talking about. I’m not here to argue – this was just an interesting insight I had that I wanted to share*

What do you think?

10 Points
Upvote Downvote

Leave a Reply

Your email address will not be published. Required fields are marked *

GIPHY App Key not set. Please check settings

14 Comments

  1. Stake stablecoin vs stablecoin and you have no IL and can reap rewards. Plus Bancor and THORchain subsidize IL after 100 days. It’s all about managing the IL intelligently. That’s why Curve finance is so successful.

  2. You’re forgetting that black swan events, though not uncommon, rarely happen over night. You’ll have plenty of time to react to a 10x event in your example.

    The more realistic outcome is that you’ll see your assets move by 100-200%, at which point you’re “only” losing 5-12% through IL and can react accordingly.

    But also, don’t be afraid of IL. Most crypto move along the same BTC baseline indicator, so you would rarely see one coin stay at original price and the other go 10x, unless you’re doing stablecoin pairing.

  3. You’re hedging in liquidity pools. Not the best outcome but not the worst outcome either. Good for me who doesn’t know what the future holds, but bad for you because you can tell the future.

  4. You are only taking a retrospective view as if you had the oracle to be on the correct side of two assets when the market is going to move. No you don’t.

    When you are on the wrong side, using your own computation:

    When positive blackswan, you won’t miss as much as on the token that doesn’t not go up.

    When negative BS, you won’t loss as much as on the token that goes down a lot.

    LP is risk neutral as long as you don’t have a clear idea on the two assets while wanting both, plus giving you rewards for providing a service to market liquidity. It is essentially what banks have been doing in all over the years. Now DeFi makes this tool to everybody.

  5. Agreed. IL is underemphasized and can rekt you. This is why Bancor, integral etc are more attractive.

    I do LPs only if heavily incentivized, not just normal fees. Right now on polygon it can be worth it

  6. You’re muting volatility for fee’s. LP tokens are a completely different asset than just holding a coin outright. If you’re concerned about IL you’re thinking about it the wrong way.

  7. In less words: you’re short volatility

    You can replicate the full payoff of a constant product LP with a portfolio of short options, the underlying, and a bond.

    Let’s put it this way, if there were a way for me to short LP tokens – I would do it in a heartbeat.

  8. Your scenarios can happen if you add liquidity for a top coin (like BTH, ETH) and some shitcoin. Add liquidity to correlated coins (coins with similar weight in terms of market cap, volume, etc.), especially the ones you were going to hodl anyway, and you will be fine.

Some Trust Wallet security questions..

Another DOA Ledger Nano X