5 ms·

A Super Simple Crypto Arbitrage Spreadsheet for Finding Mismatched Prices

37 points by spreadstreet 9y ago

bufferoverflow 9y ago
You don't even need any spreadsheets, you can open any currency and sort by price, then take pairs from the top and the bottom of the list. The problem is getting currencies in and out of exchanges in time and paying all the fees, while still making a profit. While there are fast cryptocurrencies like Ethereum and Litecoin, most exchanges still force you to wait for something like 6 confirmations. Which takes minutes to hours. And forget about doing this with USD, it takes days to move from one exchange to another. If you look at the exchange rates, fast pairs all have very similar exchange rates on different exchanges, because you aren't the only one doing this.
- wbl 9y ago
  That's not what you do. You have piles of each at each exchange and arb between the piles. Long transfers you do rarely to keep piles the same if the trading doesn't.
  - xur17 9y ago
    A lot of the arbitrage opportunities I found in the past existed because it was challenging to get cash in and out of the exchange, so the arbitrage was always in one direction.
  - bufferoverflow 9y ago
    But which "each exchange"? There are dozens of the big ones, fees can get very expensive, and you spread your money thin.
kesor 9y ago
I created a straightforward explanation on how to take Fees into account. Because FEES are super duper important, and often people just don’t understand the simple math required to compute them. The explanation is in this blog post: https://steemit.com/arbitrage/@kesor/the-math-behind-cross-exchange-arbitrage-trading https://steemit.com/arbitrage/@kesor/the-math-behind-cross-e... Another critical thing that creates arbitrage across exchanges is a malfunction at some exchange. One great example is the YoBit exchange, which has about 100+ wallets permanently as “offline”. The price at YoBit might be much higher than on some other exchange, but without a wallet, you cannot move the funds across. Newbies beware!
meirelles 9y ago
IMHO the biggest issue is keeping the money sitting on multiple exchanges, that's very risky, especially these kinds of amounts to make arbitrage meaningful. The counterparty risk is high, crypt exchanges are not like banks or brokerages, you should buy or sell then withdrawal whatever you got asap.
- jnordwick 9y ago
  Hugely underrated comment. Counterparty risk is incredibly difficult to price, and you're going to get adverse selection problems (eg you're more likely to get bids hit on venues that are more difficult to move btc off of). BRTI (CME real time index) is closing in on 10k (@ 9918.33 right now), but I've heard of issues moving money at all for exchanges used to calculate the index.
thisisit 9y ago
Whenever I read an article on "crypto" arbitrage I roll my eyes. For one, people don't seem to account for fees, ever. This guy makes the same mistake in ELIF and mentions it as some kind of "responsible disclosure" towards end. And for god's sake, it is not because "these (NYSE and Nasdaq) markets have had decades of consolidation and mergers" rather the National Market System regulation and NBBO: https://www.investopedia.com/terms/n/nbbo.asp https://www.investopedia.com/terms/n/nbbo.asp Frankly, every time I read article like these I fear for "investors" in the cryptocurrency space.
- spreadstreet 9y ago
  Hi thisisit, Thanks for the comment. If you would have read through the entire article, you would see fees mentioned in the "pitfalls" section. I am glad you brought up the NBBO, as it is an extremely important regulation for newer traders to understand. However, the comment on mergers and consolidations still stands...just look at the Philly Stock Exchange: https://en.wikipedia.org/wiki/Philadelphia_Stock_Exchange https://en.wikipedia.org/wiki/Philadelphia_Stock_Exchange
  - thisisit 9y ago
    I am sorry to sound condescending but if you read my post, a rather short one, I said: "and mentions it as some kind of "responsible disclosure" towards end." Arbitrage money is net of fees between exchanges so your ELIF showing $5 profit is kind of irresponsible giving people false information. If you are concerned about your users do mention the fees in bold letters above fold. On your question on exchanges in US, well read this: https://www.investopedia.com/ask/answers/08/security-market-usa.asp https://www.investopedia.com/ask/answers/08/security-market-... And it doesn't even mention IEX: https://en.m.wikipedia.org/wiki/IEX https://en.m.wikipedia.org/wiki/IEX So no not only two exchanges in existence. And I mentioned NBBO because it nearly removes arbitrage opportunity not because there are only "two exchanges". But if you want to change your goalpost to focus on M&A, I digress.
lordnacho 9y ago
There are more than two stock exchanges in the United States. You also want to think about what you mean by arbitrage. If you can buy IBM at 95 on NYSE and sell it at BATS for 100, you might call that an arbitrage. Your main issue is sending the orders off before someone else does. The problem you don't have is credit risk. Your money on one exchange is as good as your money on another. In crypto this is not quite the case.
- spreadstreet 9y ago
  Hey Lord, Thanks for the comment. I reiterated that it was not an apples-to-apples comparison for stock exchanges...but let's be honest. NYSE and Nasdaq are far and away the leaders. Aboslutely agree on sending the orders off before someone else does. I mention that in the "pitfalls" section.
  - jnordwick 9y ago
    Arbitrary relies on better execution than your competitors. It isn't simply a pitfall or side issue. You really need a way to estimate slippage to make this information realistic.
  - lordnacho 9y ago
    It wasn't meant to be a damning criticism. Here is a chart of exchange volumes. It's far from a two-horse race. There are structural reasons why there will be business for more than just a couple of exchanges, and it's closely linked to arbitrage players actually. https://www.sec.gov/marketstructure/datavis/ma_exchange_tradevolume.html#.Wh3NybSFjyI https://www.sec.gov/marketstructure/datavis/ma_exchange_trad...
itodd 9y ago
I wrote a bot to do this a few years ago. It would grab the order books from a set of exchanges and execute on the overlap. It had support for maker/taker fees as well as deposit/withdrawal fees for rebalancing accounts. I avoided USD trades as it required so much time to transfer. It worked well and it was a fun project to learn how markets functioned. The risk of missing the trade and the risk of hacks ultimately caused me to stop the experiment I had BTC in bter when it was hacked. That more than wiped out all of the profits I had made arbitraging.

37 points by spreadstreet 9y ago

You don't even need any spreadsheets, you can open any currency and sort by price, then take pairs from the top and the bottom of the list. The problem is getting currencies in and out of exchanges in time and paying all the fees, while still making a profit. While there are fast cryptocurrencies like Ethereum and Litecoin, most exchanges still force you to wait for something like 6 confirmations. Which takes minutes to hours. And forget about doing this with USD, it takes days to move from one exchange to another. If you look at the exchange rates, fast pairs all have very similar exchange rates on different exchanges, because you aren't the only one doing this.

wbl 9y ago
That's not what you do. You have piles of each at each exchange and arb between the piles. Long transfers you do rarely to keep piles the same if the trading doesn't.
- xur17 9y ago
  A lot of the arbitrage opportunities I found in the past existed because it was challenging to get cash in and out of the exchange, so the arbitrage was always in one direction.
- bufferoverflow 9y ago
  But which "each exchange"? There are dozens of the big ones, fees can get very expensive, and you spread your money thin.

kesor 9y ago

I created a straightforward explanation on how to take Fees into account. Because FEES are super duper important, and often people just don’t understand the simple math required to compute them. The explanation is in this blog post: https://steemit.com/arbitrage/@kesor/the-math-behind-cross-exchange-arbitrage-trading https://steemit.com/arbitrage/@kesor/the-math-behind-cross-e... Another critical thing that creates arbitrage across exchanges is a malfunction at some exchange. One great example is the YoBit exchange, which has about 100+ wallets permanently as “offline”. The price at YoBit might be much higher than on some other exchange, but without a wallet, you cannot move the funds across. Newbies beware!

meirelles 9y ago

IMHO the biggest issue is keeping the money sitting on multiple exchanges, that's very risky, especially these kinds of amounts to make arbitrage meaningful. The counterparty risk is high, crypt exchanges are not like banks or brokerages, you should buy or sell then withdrawal whatever you got asap.

jnordwick 9y ago
Hugely underrated comment. Counterparty risk is incredibly difficult to price, and you're going to get adverse selection problems (eg you're more likely to get bids hit on venues that are more difficult to move btc off of). BRTI (CME real time index) is closing in on 10k (@ 9918.33 right now), but I've heard of issues moving money at all for exchanges used to calculate the index.

thisisit 9y ago

Whenever I read an article on "crypto" arbitrage I roll my eyes. For one, people don't seem to account for fees, ever. This guy makes the same mistake in ELIF and mentions it as some kind of "responsible disclosure" towards end. And for god's sake, it is not because "these (NYSE and Nasdaq) markets have had decades of consolidation and mergers" rather the National Market System regulation and NBBO: https://www.investopedia.com/terms/n/nbbo.asp https://www.investopedia.com/terms/n/nbbo.asp Frankly, every time I read article like these I fear for "investors" in the cryptocurrency space.

spreadstreet 9y ago
Hi thisisit, Thanks for the comment. If you would have read through the entire article, you would see fees mentioned in the "pitfalls" section. I am glad you brought up the NBBO, as it is an extremely important regulation for newer traders to understand. However, the comment on mergers and consolidations still stands...just look at the Philly Stock Exchange: https://en.wikipedia.org/wiki/Philadelphia_Stock_Exchange https://en.wikipedia.org/wiki/Philadelphia_Stock_Exchange
- thisisit 9y ago
  I am sorry to sound condescending but if you read my post, a rather short one, I said: "and mentions it as some kind of "responsible disclosure" towards end." Arbitrage money is net of fees between exchanges so your ELIF showing $5 profit is kind of irresponsible giving people false information. If you are concerned about your users do mention the fees in bold letters above fold. On your question on exchanges in US, well read this: https://www.investopedia.com/ask/answers/08/security-market-usa.asp https://www.investopedia.com/ask/answers/08/security-market-... And it doesn't even mention IEX: https://en.m.wikipedia.org/wiki/IEX https://en.m.wikipedia.org/wiki/IEX So no not only two exchanges in existence. And I mentioned NBBO because it nearly removes arbitrage opportunity not because there are only "two exchanges". But if you want to change your goalpost to focus on M&A, I digress.

lordnacho 9y ago

There are more than two stock exchanges in the United States. You also want to think about what you mean by arbitrage. If you can buy IBM at 95 on NYSE and sell it at BATS for 100, you might call that an arbitrage. Your main issue is sending the orders off before someone else does. The problem you don't have is credit risk. Your money on one exchange is as good as your money on another. In crypto this is not quite the case.

spreadstreet 9y ago
Hey Lord, Thanks for the comment. I reiterated that it was not an apples-to-apples comparison for stock exchanges...but let's be honest. NYSE and Nasdaq are far and away the leaders. Aboslutely agree on sending the orders off before someone else does. I mention that in the "pitfalls" section.
- jnordwick 9y ago
  Arbitrary relies on better execution than your competitors. It isn't simply a pitfall or side issue. You really need a way to estimate slippage to make this information realistic.
- lordnacho 9y ago
  It wasn't meant to be a damning criticism. Here is a chart of exchange volumes. It's far from a two-horse race. There are structural reasons why there will be business for more than just a couple of exchanges, and it's closely linked to arbitrage players actually. https://www.sec.gov/marketstructure/datavis/ma_exchange_tradevolume.html#.Wh3NybSFjyI https://www.sec.gov/marketstructure/datavis/ma_exchange_trad...

itodd 9y ago

I wrote a bot to do this a few years ago. It would grab the order books from a set of exchanges and execute on the overlap. It had support for maker/taker fees as well as deposit/withdrawal fees for rebalancing accounts. I avoided USD trades as it required so much time to transfer. It worked well and it was a fun project to learn how markets functioned. The risk of missing the trade and the risk of hacks ultimately caused me to stop the experiment I had BTC in bter when it was hacked. That more than wiped out all of the profits I had made arbitraging.