Forums  > General  > Best “unknown” quant funds/shops  
     
Page 1 of 3Goto to page: [1], 2, 3 Next
Display using:  

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-01 01:41
A common trope one seems to hear when discussing e.g. Renaissance’s Medallion Fund is that “there are plenty of other funds out there killing it, you just haven’t heard of them.” I guess this is plausible in the sense that there might be very small shops running low-capacity strategies with high Sharpes year over year, and I suppose you’d never hear of those. But what irks me is that whenever I see a statement like this, no example is ever given. If a group is *that* good you would think they might scale up eventually, or somehow or other still make a name for themselves. Or is it simply that they know their strategies’ limits and are happy to stay under the radar?

In short, what are some examples of quant teams, funds, or shops (not pure HF/market-making) that perennially outperform? What are some examples of great “unknown” shops? Are there particular teams at the multi-manager platforms (Cubist, Millennium, etc.) that people might be referring to when they make these statements?

Zoho


Total Posts: 20
Joined: Feb 2018
 
Posted: 2018-07-01 10:14
Millenium actually looks pretty cool
http://www.legendsfund.nl/legends/izzy-englander/

Maggette


Total Posts: 1062
Joined: Jun 2007
 
Posted: 2018-07-01 11:08
Jse Street Capital and TwoSigma.

Ich kam hierher und sah dich und deine Leute lächeln, und sagte mir: Maggette, scheiss auf den small talk, lass lieber deine Fäuste sprechen...

Azx


Total Posts: 38
Joined: Sep 2009
 
Posted: 2018-07-01 12:14
Pan Capital was one of the early adopters of hft with >100% annual returns for more than a decade. Nowadays they have scaled back on the trading significantly.
Pretty unknown when it happened, although they have become more publicly known in recent days (there is even a book about them).

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-01 14:36
I appreciate the responses. However, I work in the industry so I’m quite familiar with the big names like Two Sigma and Millennium and many others. I know it’s vague, but again I was hoping to learn about some places that I’d never heard of. Also, I may be wrong, but my impression is that Two Sigma has not done spectacularly these last few years. They are massive at this point. To be fair perhaps I don’t know about all their funds, but certainly I don’t think they’ve come anywhere near Medallion-level performance.

I hadn’t heard of Pan Capital. Thanks!

NWPhy


Total Posts: 17
Joined: Dec 2005
 
Posted: 2018-07-01 17:18
TGS Management has had a pretty good run according to a Bloomberg article that was written about them. They've operated under the radar for quite sometime. Essentially, they are a continuation of Princeton Newport albeit without Ed Thorp's involvement.

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-01 17:27
TGS is a *perfect* example of what I’m interested in. I hadn’t heard about them before the Bloomberg article, and then here was this legendary group that people caught onto partly via the management’s charitable efforts. Awesome story. Any idea if they’re still doing as well as they were?

NWPhy


Total Posts: 17
Joined: Dec 2005
 
Posted: 2018-07-01 17:40
Not a clue, I'm afraid. I didn't even know that they existed until the Bloomberg article and I've done plenty of research into finding quant funds / prop houses lurking under the radar. I'm sure that there are others on the forum who have more to say on them if they're willing. The Medallion folks even have more to say about themselves than these ladies and gents.

EspressoLover


Total Posts: 337
Joined: Jan 2015
 
Posted: 2018-07-01 18:49
Scalability is a major consideration. Many many groups produce Medallion+ returns but on tiny capital bases. Doing 100% a year on $1mn isn't exactly legendary-worthy. Also regarding Two Sigma and the like, you have to remember that many hedge funds keep separate insider-only funds. They put their best strategies in those structures, and don't report results. This is the same story as Renaissance, but the major difference is that because of its history Medallion is more famous than RIFF.

> (not pure HF/market-making)

Medallion's fairly high-frequency. Last I heard it was doing something like 3% of the ADV on US futures and equities. Given its capitalization, and reasonable leverage assumptions, that would imply sub-hourly turnover. As for market making, I would guess a substantial amount of its order flow adds liquidity. Both in the mechanical and economic sense. Overall I don't think Medallion's that different than a firm like Jump, besides operating at larger scale and pushing the horizons out a bit further.

Good questions outrank easy answers. -Paul Samuelson

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-01 19:11
Great points. I definitely understand the capacity/scalability issue, and it’s true I’m more interested in funds with nontrivial AUM.

Re:medallion, I suppose you’re right that it’s a bit misleading to refer to it as non-pure HF.

Any color on the internal prop funds at places like two sigma, de Shaw, etc.

Thanks for the reply!

HitmanH


Total Posts: 463
Joined: Apr 2005
 
Posted: 2018-07-01 22:54
So - in terms on consistent Sharpe >3, longer term.
Names not mentioned here (buy people know) include PDT, GSA, Verition, Princeton Alpha.
TGS? Voleon?

Those people don't know - so many of these are short term, haven't scaled - so somewhere between HFT prop and hedge funds.

Machina, Quantbot, 1-Eleven, Quadrature, Alphagrep, Maven (given now hedge business), Qube?

EL: "you have to remember that many hedge funds keep separate insider-only funds. They put their best strategies in those structures, and don't report results. "
Really? ;-)

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-02 01:33
Here are some names I don’t know. Very cool :) Thanks!

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-02 01:46
Speaking of quantbot, do you know much about their QMM setup? Is it substantially different from other multi-manager platforms? It sounds as though the idea is to be able to plug in and leverage their technology and resources a bit more readily than at other places. Is the entire fund comprised of individuals or small teams like this, or is there a larger central group as well? Not sure if that makes sense. The QMM situation sounds unique though.

aostvlfix


Total Posts: 1
Joined: Jul 2018
 
Posted: 2018-07-04 02:08
Adding to the ones already named.

On the prop side - Final, Sunrise, HC Tech, Grasshopper, SSW/IAT.

On the fund side - Group Twelve Nine, Elemental, Vector Capital, BGGI, Osprey, Signal, Spark LP, Sigmoid Trading, Z-Flow Advisors.

Problem with TSI is that if you have global macro positions of 5000+ treasury contracts, you can't turn them over without getting raked, it's not that they aren't doing their job but that they've gotten really big.

Jump has long been closer to the pareto frontier than Rentec.

frolloos


Total Posts: 48
Joined: Dec 2007
 
Posted: 2018-07-04 02:45
Any thoughts/opinions on Dutch quant or hedge funds (which probably fall under the category unknown by default)? E.g. Mint Tower Capital, does anyone know the fund?

jslade


Total Posts: 1138
Joined: Feb 2007
 
Posted: 2018-07-05 22:15
Voleon is an interesting example; former DE Shaw guys. Edgestream/Merfin used to be an interesting example from that part of the world (Sandor Straus -former rentech PM, along with Peter Yianilos; sold to someone else). RGM advisors was an early innovator in automated market making, comparable to Virtu or Knight. I think they're still out there doing their thing, albeit at much lower returns. Nobody ever talks about them, mostly because they're in Texas.

There are lots of little prop shops who have historically done well. There are also lots of weird things out there that look like money laundering or fraud.

"Learning, n. The kind of ignorance distinguishing the studious."

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-06 02:21
These are getting pretty interesting. Hope this thread isn't too "gossip-y" for the forum, but I've enjoyed hearing some new names and getting a few insights.

A couple thoughts/questions...

- I recall some hype around Voleon using ML for trading, which didn't really move the needle for me as every fund under the sun is doing some flavor of ML at this point. Their track record doesn't appear to be stellar. Any idea what they're really doing?

- aostvlfix, agreed re: global macro funds. At their size you're so constrained by what you can turn over, and I'm not sure there's meaningful alpha at those timescales that isn't described by a handful of simple "factors" (momentum, etc.). Also, I can honestly say I haven't heard of nearly any of the prop shops or funds you named, nor can I find a web presence for some of them. Are they really *that* under the radar?

- People have mentioned PDT and GSA which both spun out of banks. How about Squarepoint (I believe a Barclay's group)? Any color on how they're doing?

HitmanH


Total Posts: 463
Joined: Apr 2005
 
Posted: 2018-07-06 12:40
Didn't someone buy RGM?
(or it's ex-US operations?)

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-06 13:10
Yep, they're still around in Texas but got acquired by DRW I believe.

jslade


Total Posts: 1138
Joined: Feb 2007
 
Posted: 2018-07-07 03:57
Most low frequency shit using "machine learning" is just Grinold and Kahn, except idiots using some of the stuff you have in R instead of Minitab. There really isn't much new under the sun. Squarepoint is probably more of the same. Frankly, the guys I knew at Blackrock could do the same kind of thing, but they're constrained by regulatory stuff and capacity.


"Learning, n. The kind of ignorance distinguishing the studious."

Zoho


Total Posts: 20
Joined: Feb 2018
 
Posted: 2018-07-07 09:07
@jslade
those who claim they are using ml could do this just to attract labor otherwise people will all go to some kind of googles

kuebiko


Total Posts: 24
Joined: May 2018
 
Posted: 2018-07-07 15:25
Ok so again I pretty much agree with everything that’s being said. But kind of going back to a slant on the original question... which funds named or not named here are doing something genuinely novel? Are they all doing more or less the same thing (conditional on being HF, low-freq, whatever) but to varying degrees of success? Is it a matter of taking well-known ideas and implementing them well (with good execution, etc), or is anyone really doing something cutting edge and one-of-a-kind? I suppose having access to unique data sources would be one example of the latter. But I’m also interested in novel methods and ideas. I realize I’m asking something that most people can’t or won’t answer but i figured it can’t hurt.

u1234


Total Posts: 9
Joined: Jul 2018
 
Posted: 2018-07-09 05:43

Voleon performance is barely a Sharpe of 0.5, but they do have a few large years. Sell themselves on ML. They have some fan base investors but nothing great.

Quadrature is Stat Arb, very complicated process. Ex DE Shaw. Net sharp of 2.5.

PDT too has net Sharpe of 2.5.

GSA has many funds and performance has varied over time. Most of their funds are enhanced Risk Premia, but their GSA Capital partners made an allocation to a FX manager XTX in 2009 till 2015 who had a Sharpe of 27, and performance of that Fund was good. After XTX went on their own, performance has been vanilla.

Princeton Alpha shut down. Not able to make any money.

Qube, or do you mean Cube which is the ex CS Fund. Made a lot of noise. 1bn at launch. Made no money since launch outside of CS.

u1234


Total Posts: 9
Joined: Jul 2018
 
Posted: 2018-07-09 05:46
Squarepoint has 3 funds. The first had a Sharpe of 6, second had a Sharpe of 3, but the 3rd fund is mostly enhanced Risk Premia and struggling. Strategies in the first two funds, range from pure Arb like index cash Arb, Stat Arb, pairs, futures vs spot in European indices, and can be very short term. The first two funds are nothing like Grinhold and Kahn.

u1234


Total Posts: 9
Joined: Jul 2018
 
Posted: 2018-07-09 05:50
Average holding around 3 weeks last time I met. Sharpe is closer to 3.

One of my past employers used to be an investor in Medallion until they kicked out all investors (external). Sharpe was 4 back then with a massive 44% perf and 5% mgmt fee. Sharpe has come down a bit since then.

Still very smart guys.
Previous Thread :: Next Thread 
Page 1 of 3Goto to page: [1], 2, 3 Next