Sam Trabucco is a quantitative crypto trader and the co-CEO of Alameda Research. Prior to Alameda, he worked as a quantitative trader at Susquehanna International Group (SIG), trading bond ETFs and working on their sports betting desk. Sam has a degree in math and computer science from MIT.
How would you describe your process? And how do you balance quant discipline with other inputs?
When I am actively trading, I try to keep an eye on everything. Our automated trading bots are trading on all the big platforms, so I’m constantly looking at what they’re doing – if I notice something abnormal in our fills, I investigate, as often there’s at least an opportunity to maximize if it’s doing something good, and sometimes it reveals a bad parameter or system problem worth addressing. We also have internal tools to monitor the premia of various products, whether any coins have moved a lot lately, big spikes in volume or open interest or liquidations, etc. which I keep an eye on, and which can also alert me. I also monitor some social media, and we have alerts for e.g. whenever Elon Musk tweets.
When I need to change something – one of our positions, an interest rate parameter, a fading parameter, something in a model, etc. – we have an in-house system called Pointer which generally makes that easy. We can control thousands of knobs with Pointer, and, even though our trading is almost all automated, we’re making hundreds of these parameter changes per day. We also use spreadsheets to control some more complicated things like our models.
Some traders have rigorous personal and professional routines. Do you have a daily trading routine?
Not an especially conscious one – I wake up and go through slack, review the list of priorities I maintain for any re-ordering or additions I can remember, and then either work on those or monitor trading actively if the market is relatively busy (which, in crypto, happens something like 50% of the time). Even if I am focusing on something besides watching trading actively, I am at least passively watching it so I am aware if we start doing something interesting, or if the market moves in important ways, or whatever – this helps with maintaining context in the event something does happen. I don’t have any other specific routines, though.
How has your trading had to evolve as your trading notional at risk has increased?
As our capital base has grown (and so has volume in crypto trading), we’ve had to think a lot more than we used to about risk. For instance, what fraction of the open interest for some major contract are we willing to be? How much capital can we justify keeping on each platform given that the data to determine counterparty risk is quite minimal? We’ve developed frameworks for answering these questions (and discounting the EV of each opportunity for how much risk it represents on average, according to our admittedly not perfect frameworks), and we’ve gotten better at it as we’ve gained practice with our capital base growing.
What is/are your edge(s)?
I think I am a strong fast thinker – when a problem arises or changes quickly and it’s important to make a plan immediately and execute, I am quite good. This matters a great deal in crypto trading where market conditions can change very fast and it can be important to make huge calls with a ton of PNL impact in seconds. Along similar lines, I am also “numb” to big numbers to some degree and am good at not being results-oriented in a way which makes me well-suited to making big decisions about lots of money which can sometimes go against me.
Otherwise, I’d say my biggest relative edge – besides things like being quantitative and good at playing games and whatnot – is the intuition I’ve developed for the ways the market works. It’s hard to make explicit, but playing this game for a while has given me advantages moving forward and playing it – I have a better chance at correctly interpreting news than most people, I know what sorts of impacts various market events should have, etc.
How do you think crypto markets differ from the traditional markets you’ve traded?
The biggest differences center on volatility. Crypto markets have crazy volatility as compared to traditional assets – especially the bond ETFs I was trading before. This is caused by a few factors – nascency of the asset class, exchange structure encouraging liquidations, etc. – and they combine to form markets which are super volatile and trade 24/7. It’s a lot of fun!
Lack of centralization contributes big differences, too. Without anyone to mandate prices on two exchanges must be the same, arbitrages are allowed to arise for literally identical products. Futures spreads can also persist because prime brokerage isn’t a thing – in order to put on a zero-delta spread buying cheap futures against spot, I need to have capital in two places, and capital comes at a cost. Data and latency are also of lower quality than in traditional markets – with no one regulating otherwise, exchanges don’t have incentive to improve.
How do you monitor market news & sentiment?
We monitor sentiment in a few ways. Our traders will tend to watch Twitter and other social media from time to time to make sure nothing big has happened. We also have a slack channel called #urgent-trade-news which we encourage anyone at the company to quickly post if they see literally anything which they think could have market impact – that’s been a good addition to our trading and kept as apprised of important events that have market impact on the order of minutes (or less).
In terms of analyzing sentiment, a big strength of ours tends to be identifying spots where the market seems to be over-rating the impact of news – these are spots where we often will make big, longer-term delta bets because we feel that we can predict where the market will go as the market course-corrects. For instance, we did this in mid-July 2021 when BTC got down to the \$30k level after some news – news which wasn’t exactly positive for crypto, but which never really seemed as negative as the market was treating it.
Do you have weaknesses or biases you are aware of?
I am pretty headstrong and likely am over-confident in general. I think this has served me decently well – over-confidence tends to make you try more things than you “should” and sometimes those things end up going well, and that’s definitely been the case for me. But it also causes problems where I over-rate my own opinions and abilities sometimes, and that can be quite costly. I also have plenty of object-level weaknesses – I’m not an especially good coder, complicated VBA makes my eyes glaze over, etc. – which (thankfully) other people at Alameda make up for.
I also find that my natural inclination is to be biased in the direction opposite to my most recent mistake. For instance, if I recently put on some trade way too slowly and missed an opportunity as a result, I might be biased too strongly in the direction of putting it on very quickly in the near future, at the cost of paying a ton for fast execution. I think I’ve gotten decently good at addressing my more object-level biases to the point where, if I am aware of one, I can reduce its impact to the point where I’m not currently cognizant of major object-level biases I have – I definitely do have them, I am just not sure what they are.
Who have you learned the most from?
I’ve learned a lot from a lot of people – at SIG where I got my first training in quant trading, and from the people around me at Alameda. I’ve learned a lot of object-level things from them, of course, but I’ve also gotten a lot better at having a good mindset while trading, not being results-oriented, and optimizing resources given an immense set of problems one could be working on. Everyone here at Alameda has helped me with these things.
What are your pet hates about the market?
Honestly, I do think the crypto markets are the most interesting in the world, mostly because of the ways their idiosyncrasies combine to form a unique and ever-evolving game for me to try to stay ahead of. There are elements that I find annoying – certain segments of “the community” are fairly toxic, and it would be nice if it weren’t truly 24/7 just so there was a time I could take a break without worrying about our positions. But everything together is so fun and represents such a great opportunity that I wouldn’t change it.
Oh, and Tether truthers tagging me on Twitter has gotten a little old.
What do you hope for crypto? And for Alameda?
Alameda: I hope we can continue to remain agile and ahead of the game to the same extent we always have. So much of our success has come because we’re always been on the “cutting edge” to some extent, we’ve been fast to figure out what the most important opportunities are, and we’ve been able to move quickly to take advantage of them, often by doing a lot of hard and annoying work our competitors haven’t done.
We saw that yield farming was hugely valuable, so we spun up one of the world’s biggest yield farming operations over the course of days. We saw that long-term investments were underpriced, so we started a super active VC arm. The crypto markets themselves evolve rapidly, and we need to make wholesale changes to our automated strategies in response to them very often – and we have to do this fast.
This is still going to be the recipe for success as crypto keeps evolving, and we need to keep doing it to stay successful.
Crypto: I hope crypto attains its potential as the future of money in all the ways it seems likely to me that it can via projects like Solana.
Last but not least, what are your strongest market convictions now?
I wouldn’t say that I tend to hold strong long-term market convictions very often. I do think fairly strongly that the market will continue to behave in certain ways – for instance, we’re in a paradigm right now where the market is very responsive to news in the obvious directions, often overly so, to the point where, after the impact occurs, there’s quickly some reversion. I think that’s decently apt to keep up, with the caveat that the effect has been evolving to become faster as the market understands it better.
I also believe that market structure will keep evolving in the direction it has been lately – regulators are increasingly clamping down on some of the more “unsafe” functionalities that crypto exchanges offer, and exchanges are taking steps to reduce the maximum leverage users can attain – both of these effects point in the direction of fewer liquidations and lower short-term volatility. This is what we’ve seen, as big, sharp moves have gotten rarer and smaller (though not non-existent!) – I expect the market to keep evolving in this manner.