What is crypto arbitrage and is it effective? Cryptocurrency arbitrage allows you to earn on the difference in the value of assets. This type of trading came from traditional financial markets and has become widely used by crypto traders.
Types of cryptocurrency arbitrage
There are two main types of crypto arbitrage:
The first and most common way is to buy cryptocurrency on one exchange at a lower price and sell it on another at a higher price. There are many factors to consider in this approach: spread, fees, transaction speed, price movement speed, trend direction and so on. The second method is more complex, but less risky, because all transactions are made on the same exchange, and the work goes faster. A popular scheme is the so-called intra-exchange triangle, based on the difference in the rates of several cryptocurrencies. Two or more cryptocurrencies are involved in the trade. For example, the first bitcoin is bought, then BTC is exchanged for XLM, which is sold for dollars.
Disadvantages of crypto arbitrage
The method is quite difficult for beginners and requires a certain set of skills. It is necessary to carry out complex calculations and take fees into account: both trading and cryptocurrency deposit and withdrawal fees. It is profitable to engage in inter-exchange arbitrage only when the trader has a sufficiently large deposit since the total fee can be up to 200 or more dollars. This increases the risks significantly. When conducting inter-exchange arbitrage with cryptocurrencies such as BTC or ETH, the rate can change significantly due to fairly slow transactions, which, under high load on the network, can hang for several hours.
How effective is crypto arbitrage
Arbitrage can be effective due to high volatility and rapid changes in the exchange rate of cryptocurrencies. But for the same reason, it can lead to losses as the price can quickly level off or become lower at the time of sale. It is more appropriate to use intra-exchange arbitrage. Trade-mate.io service, which allows you to manage assets in a single window, will help you a lot. Currently, three major exchanges can be connected: Binance, Bitmex, and Poloniex. If you decide to use inter-exchange arbitrage, you can track prices and quickly sell on two exchanges at once. For example, open long on Binance and short on Bitmex without withdrawing funds to save time.
Bitbot: Inspired by AskMike's triangle arb bot, here's the story of my exchange arb bot.
I started typing all this as a comment on AskMike's excellent post , but as it got longer and longer, I figured it was different enough to deserve its own post. I was very active in arb starting in the second half of 2017. My bot did not do triangle arb as Mike's, but rather exchange arb. I limited myself to being on GDAX, Kraken, and ItBit. As I did not trust tether, I did not consider any exchange that relied on it, despite that costing me a lot of opportunities, and as an American, I was blocked from Bitfinex (and didn't want to risk the VPN thing). Exchange arb is easy to do on a small scale, but doing it on a larger scale was harder than it might first seem:
The chance of an exchange losing/bobbling your fiat deposit or withdrawl is greater than zero. More fiat movement = greater probability one will go wrong.
If #1 happens, your money is tied up for weeks thanks to overburdened support staffs. I had GDAX drop a $25,000 wire withdrawal on the ground and swear it was successfully wired. It was not. That money was tied up for months while I fought with Coinbase support.
You could sometimes get around having to transfer fiat if you had incongruent deltas between different coins, however this meant your trading bot needed to be twice as complex.
To do this at any scale you need larger withdrawl limits on each exchange. This takes time. It took Kraken a month and half to verify me to tier 4; I lost out on thousands of dollars worth of opportunities because of this.
I lost quite a bit of money on Kraken's old shitty (now better) trading engine when it would claim an order wasn't submitted, but then 2 minutes later it magically was. ItBit doesn't offer MakerOnly orders, so you'd often get hit with the taker fee when the market data said your order should go in as a maker. But those were the sum total of the gripes. On the other side of things, I was making hundreds a day from arbitrage from October through January (more on good days, nothing on days where I was waiting on fiat, or had reached exchange withdrawl limits). Note that I was not doing true risk free arbitrage as I was not shorting coins to offset the long position on the other exchange; I had an "inventory" of coins waiting on each exchange. This avoided having to wait for coins to transfer to close the loop, I would just reset them after the trades were done. So I had Hodler risk due to that stack. My bot was pretty simple. Listen to the market data on each exchange. Poll for fiat and coin balances on each exchange. If an opportunity greater than a configurable threshold exists and the funds are in place, buy exchange A and sell exchange B. The bot would not attempt to do fiat or coin transfers, even though the APIs did support that, I wanted to handle those manually. The bot would attempt to place maker orders to save on exchange fees, and could dynamically reprice its orders to stay on top of the book if prices moved. My bots were all written from scratch; other than wrappers for exchange APIs, I did not use any existing bots or bot frameworks. I remember one day I the deltas were particularly fat--hundreds of dollars per BTC--and I had a wire deposit for $25,000 pending to ItBit. I had meetings all day at work, and throughout each of them I was barely paying attention, clicking "refresh" on my ItBit account waiting for the that money to hit. Then all of the sudden, BOOM, got the notification from my bot that it had started trading, buying on ItBit and selling on GDAX where the price was much higher. It made $2,200 in minutes. Jesus, those were the days. Then it was a matter of waiting for the wire to flow out of GDAX and back into ItBit so I could do it again. Eventually I switched to buying/selling Bitcoin while selling/buying (opposite direction) Litecoin, which was usually 2% off from Bitcoin's delta. This limited my upside, but saved the hassle of transferring fiat. But now that spread is gone too. So why did these opportunities exist?? Two reasons. One, pre-bubble crypto was mostly a younger man's game, and most younger men hadn't yet saved up large enough sums to have sitting on exchanges waiting for arb opportunities. Thanks to other professional success I was in a position to have a nice amount of cash that I could throw at this. Also the thing I mentioned above about the default withdrawl limits on exchanges. I was limited with just how much arb I could soak up, as was everyone else who was just coming onto the scene. Also I now have a nightmare on my hands for tax reporting. Because I was churning my stack daily in 2017, I have a ton of profitable transactions with the rising price of BTC, which I owe thousands in taxes on. Thousands of trades to report to the IRS. Then that same churn in 2018 resulted in losses on my trading inventory, but I can't pair those off because they're in different years. Another fun note is when I started, the number of Reddit posts that were basically like "oh you won't be able to do this for such and such a reason, if there really was free money to be made people have already grabbed it." Glad I didn't listen to those "experts". I'm sure futures contracts introduced certain arb opportunities when they came out, but I did not want to have to open and maintain accounts with some retail futures broker, and the excess capital requirements to short those futures contracts were very high at first. As a result, I did not pursue any opportunities related to those. Alas, all good things come to an end. By February, things had mostly dropped off, and now they're basically gone, because now institutional money is heavily in the game. Now in order to make money in exchange arb, you need to do enough volume to be in the lowest fee tier, as the spreads aren't even enough to cover taker fees these days, and the pros who do this full time will have better tech stacks than weekend warriors like me. I've pulled most of my funds from the exchanges and put them back into index funds. But man was it fun while it lasted.
So there's been an unbelievable multitude of threads ranging from "Shibes, is Ð1 = $1 possible?" to "Shibes, Ð1 can never be $1, stop dreaming." first of all, never tell a shibe not to dream. second of all, much much more than the ratio of dogecoins to bitcoins to dollars determines the exchange rate of a currency, and that is the subject of today's shibenomics lesson. Everyshibe has probably heard of GDP, but to define it very quickly - the gross domestic product is the value of all goods and services produced in an economy over a period of time. Normally, GDP is calculated per year, but for the analysis in this article and dogecoin in general, i find it more constructive to think of GDP per day. Now, how can we possibly find out what the GDP of dogecoin is? The USD has a whole bureau of economic analysis to do that and they still get it wrong half the time... Herein lies one of the many beauties of the blockchain. The block chain is like a public ledger of every transaction in dogecoin, EVER - which means we can just find the last block with a transaction on Wednesday, January 15th, and the last block with a transaction on Thursday, January 16th, and every transaction in between is part of the GDP for Jan 16th. note: shibes already versed in economics will point out that some of these transactions may not be for goods and services, but rather conversions from USD or BTC, and some of them may be double counted because they are buying an ingredient for something they plan to sell later for more dogecoins - these objections are somewhat correct, but systematically overestimating GDP means you can still measure GDP change over time, which is what we're really concerned about, and conversions from other currencies are our equivalent of exports So, as of now, our GDP on January 16th was a whopping 46,793,497,531 DOGE ($18,078,761 USD). Before you object about pay-it-forward threads and tipping and such, let me remind you - tips are a micro-transaction for a service (the service of making you smile :) ) and pay-it-forward threads are the shibe version of state-sanctioned lottery - which is the service of gambling. Now, gambling and entertainment an economy do not make, but 18 million dollars is a pretty big deal. Only bitcoin, litecoin, and quarkcoin have higher GDPs per day than dogecoin, and among those only dogecoin has a reasonably sustainable average transaction value (~$200 vs >$8000). Ok, so now we know dogecoin has a GDP, and that shibes can feel pretty good about it. Even more impressive, dogecoin GDP has grown from $6.74 million on December 18th to $18 million today - 3.22% per day - at current growth rates, our GDP this time next year would be $1.78 TRILLION (11% of the US economy). Now this is probably unsustainable, but even growing at 20% of our current growth rate for a year would leave us only behind bitcoin in GDP per day. Now let's consider market cap - the first thing that is striking about dogecoin isn't that its 6th most valuable in terms of market cap - its that dogecoin is the only crypto who's GDP exceeds its market cap, and that too by a whopping 62%. This means that for every 1 dogecoin you spent today, 61.7% of that dogecoin was passed on, and then passed on again, and so on. The velocity of money in the doge economy is ludicrous, and it confers a high degree of stability unto our economy, so kudos all around! A fast velocity of money not only helps fight changes in prices, it also makes it possible to post very high GDP numbers without having a large monetary base. Apart from this, its also worthwhile to note that our market cap in USD has grown by around 3.6% per day, while the number of dogecoins added to circulation grows by around 2% per day - so despite all the mining, dogecoin has been appreciating in value quite rapidly in value. On a more theoretical note, it is worth spending a moment to consider PPP (purchasing power parity). In high school economics, PPP means that a big mac in the United States should cost the same amount of real value as a big mac in China, and that the exchange rates will move to reflect that reality. In reality, PPP is more of a goal than a law - it's pretty hard to buy a big mac in China and bring it to the United States so that you get your food cheaper at lunch. In cryptoworld, however, PPP is the law - a bitcoin must cost the same amount in litecoins, dollars, and dogecoins, and the exchange rates will change to reflect that. PPP has so far been strongly determined by the BTC, LTC, DOGE triangle, as there is still not a large volume USD/DOGE exchange to allow arbitrage between the USD/BTC/DOGE (arbitrage is the force that makes PPP a law). Thankfully, a DOGE/USD exchange is probably just around the corner, and so soon we should see DOGE/USD and DOGE/BTC stabilize a bit because of this. PPP also extends to the world of mining - a kh/s mined on one currency will try to be worth as much as a kh/s mined on another currency - the best example of this is multipool. Right now, it is highly profitable to mine dogecoins because of the strong economy & exchange rate and the low block difficulty, compared to other alt coins relatively weak exchange rates and higher difficulties. Each coin has a predetermined global hash rate it will support based on its coin reward and target block time, and uses the difficulty to keep the reward & timing schedule intact. What this means is that as more miners work on dogecoin, it becomes less profitable to mine doge unless its value relative to the other cryptos goes up - and this is the cause for the cycles of highs and lows we see in DOGE/BTC. These cycles will probably not stop after the February 14th block halving, but they will be occuring at higher and higher valuations. It's a point of shibe pride to mention that of all the cryptos, dogecoin probably has the least wealth concentration - the top 100 transactions in dogecoin are only 3.15% of the daily transaction value, whereas for other currencies that number can reach near 50%. In conclusion, it'll probably be disappointing to hear that I have no clue where DOGE/USD or DOGE/BTC will be in a week, much less when it will hit 1 dogecoin per 1 USD. What I can say, and what should be evident from the numbers, is that dogecoin is developing a strong economic foundation unlike any of the other altcoins, and is much less seedy than bitcoin's early economy. Some of the core difficulties of dogecoin going forward are going to be maintaining our ludicrous velocity of money, which means finding and developing new markets for tipping, diversifying our economy away from mainly tipping while keeping a focus on micro-transactions, and creating a more effective store of value besides hoarding coins in a wallet or giving them away in PIF threads and waiting for them to come back. TL;DR - TO THE MOON! statistics used in this article are available on: http://bitinfocharts.com/
Why does the ETH/BTC pair exist on exchanges where both ETH and BTC have USD or USDT pairs?
*Hoping to do this thread sort of like a CMV (change my view), looking for someone to argue the benefit of having ETH/BTC pair (and other pairs with BTC or ETH when USD pairs also exist). I think exchanges like GDAX should remove the BTC/ETH pair. My view is that if an exchange removed the eth/btc pair, we would see much healthier movement and less of BTC “bringing us down with it” type action. Evidence: Having a triangle of pairs like ETH/USD, BTC/USD and ETH/BTC gives advantage to arbitrage bots to make profit off discrepancies between the three prices. Crypto markets are still very inefficient. If the ratio that ETH/BTC is trading at is ever different from the actual ratio of Eth’s price in dollars to Bitcoins price in dollars, bots can make profit. As I see it, this profit comes directly at the expense of average traders, and hurts the ecosystem as a whole. It’s also precisely the reason why most major coin's charts look pretty much exactly the same right now. For a while, every time Bitcoin dipped it brought ETH with it, because when the price changes, arbitrage bots use ETH as a vehicle to accumulate more bitcoin. When BCH started pumping a lot of them switched over, and then those charts started looking the same. The point is that whenever a set of 3 pairs exist on one exchange, CoinX/USD, CoinY/USD, and CoinX/CoinY, it creates arbitrage opportunities for bots. So why have them at all? As far as I can tell, it’s for convenience and to avoid paying fees twice when trading between coins, which I get. But the cost of that convenience is a huge advantage to traders using arbitrage bots. So if you aren’t using an arbitrage bot, this is hurting you. And I think it’s happening on such a large scale as to keep the prices of BTC and ETH very intimately linked, which has counteracted the positive effects that good news for ETH should have had. Maybe it made sense to have a BTC pair when that was the only way to trade for some alt coin without as USD pair, but I don't think it makes sense anymore. Concluding analogy: Imagine there were pairs like this in the traditional stock exchange, imagine you could trade Netflix stock for Amazon stock. The prices of these stocks should have very little to do with each other, but if they were paired, then their prices would DEFINITELY change in relation to each other. It seems obvious to me that that would be a bad thing, why is it okay in crypto?
------------ UPDATE #3 2015/11/05 11:35 am Pacific -------------- Current price 416. Behavior here will indicate the strength of the bull. I expect a pullback to test 400, but if we clear 500 without one, that will be a very good sign. Chance of rollover below 400 is quite low now. ------------ UPDATE #2 2015/11/05 9:40 am Pacific -------------- So far so good, diminishing volatility within the triangle as expected. Rollover would probably have already happened if it were going to. 400 is serving as a kind of ceiling. Look for a break above 400, followed by a pullback test to 400, which if it succeeds (stays above 400), will be a very good sign. Also, random, huge, short (< 1 hour) spikes down in price have happened before in the Bitcoin market on various exchanges. These are exchange blips, certainly if they are not correlated / arbitraged across exchanges, so behave accordingly. ------------ UPDATE #1 2015/11/04 11:04 pm Pacific -------------- More than an hour under 400. So we are definitely taking a break here. Best case: Two day or longer triangle formation with diminishing volatility, followed by the next move up. Just as likely: Continuation of the rollover, crash to lower support, should happen within two days, probably sooner, if it is going to happen. ------------ ORIGINAL 2015/11/04 09:15 pm Pacific -------------- As of this writing, the Bitstamp price is 415 BTC per USD. I believe that we are in the very early stages in the next major run up. If that is the case, based on previous chart action in 2013 and earlier, we should take off again in the next 12 to 24 hours, blowing through the recent high of around 500. On the other hand, if the price goes below 400 again and stays down for an hour or more, the current run is probably not the beginning of the next big thing. I expect that the real blow off top for what is (hopefully) starting now will be at least 5000 USD per Bitcoin, probably more, and we may never see a year long consolidation in price again until Bitcoin nears its ultimate high, somewhere between 100K and 1M USD per Bitcoin, higher if the USD suffers a currency crisis along the way.
------------ UPDATE #4 2015/11/05 12:36 am Pacific -------------- Current price 404. Testing the 400 floor / ceiling now. Action is not violent. Most likely next move is a gradual oscillation up and eventually away from 400 with possible multiple retests of 400. If that happens 500 will present next resistance. ------------ UPDATE #3 2015/11/05 11:35 am Pacific -------------- Current price 416. Behavior here will indicate the strength of the bull. I expect a pullback to test 400, but if we clear 500 without one, that will be a very good sign. Chance of rollover below 400 is quite low now. ------------ UPDATE #2 2015/11/05 9:40 am Pacific -------------- So far so good, diminishing volatility within the triangle as expected. Rollover would probably have already happened if it were going to. 400 is serving as a kind of ceiling. Look for a break above 400, followed by a pullback test to 400, which if it succeeds (stays above 400), will be a very good sign. Also, random, huge, short (< 1 hour) spikes down in price have happened before in the Bitcoin market on various exchanges. These are exchange blips, certainly if they are not correlated / arbitraged across exchanges, so behave accordingly. ------------ UPDATE #1 2015/11/04 11:04 pm Pacific -------------- More than an hour under 400. So we are definitely taking a break here. Best case: Two day or longer triangle formation with diminishing volatility, followed by the next move up. Just as likely: Continuation of the rollover, crash to lower support, should happen within two days, probably sooner, if it is going to happen. ------------ ORIGINAL 2015/11/04 09:15 pm Pacific -------------- As of this writing, the Bitstamp price is 415 BTC per USD. I believe that we are in the very early stages in the next major run up. If that is the case, based on previous chart action in 2013 and earlier, we should take off again in the next 12 to 24 hours, blowing through the recent high of around 500. On the other hand, if the price goes below 400 again and stays down for an hour or more, the current run is probably not the beginning of the next big thing. I expect that the real blow off top for what is (hopefully) starting now will be at least 5000 USD per Bitcoin, probably more, and we may never see a year long consolidation in price again until Bitcoin nears its ultimate high, somewhere between 100K and 1M USD per Bitcoin, higher if the USD suffers a currency crisis along the way.
Triangular Arbitrage is the process of trading three currencies (or other security) to take advantage of a price difference among the three exchange rates in order to make a profit. The above video… Triangular arbitrage means that the bot can execute arbitrage trades on single exchange (intra-exchange) avoiding all the risks involved in arbitrage between exchanges. It is designed to be as lightweight and fast as possible so you won't miss an arbitrage opportunity. All arbitrage opportunities are generated from detailed analysis of various exchanges orderbook records. There is a minimum of 10 level deep orderbook analysis for each profitable signals. This is the key element of arbitrage trading. Triangle arbitrage is done through the buying and selling of three Cryptocurrencies simultaneously to make a profit. For example, you might buy XRP (Ripple) with ETH (Ethereum), then sell the XRP to BTC (Bitcoin) then sell the BTC back to ETH to create a profit. TRIANGULAR ARBITRAGE 1. Let’s talk about Triangular Arbitrage. The Triangular Arbitrage, is the better strategy to you buy and sell the your coins in differents pair, in the same exchange. The Gimmer will go sell your coin and buy the same coin in other quote to have more profits.
Crypto/Bitcoin Triangular Arbitrage Bot - Profit Tracking Binance - Paper Trading - Duration: 5:01. Blockchain Engineer 1,771 views. 5:01. How to use Triangular Crypto Arbitrage for 1CryptoTrader https://ico.1ct.io Automated. Triangular arbitrage trading bot for cryptocurrencies Create a FREE account https://jonnyblockchain.com It is designed to be as fast as possible so you won’t miss an arbitrage opportunity. Bitcoin Trading Challenge 18,769 views. 20:16. Triangular Arbitrage Full Option - Duration: ... Triangular Arbitrage Step by step (super easy numbers to understand) - Duration: 4:47. How To Arbitrage Bitcoin for 20% - Duration: 31:58. Crypto Steve Coates 3,124 views. 31:58. Poloniex Test Bot (Triangular Arbitrage Results Analysis) - Duration: 13:16.