Stat arbitrage, but not a quant shop

[MikeLawrence]

Just got off a call with a higher level employee of a hedge fund who explained to me that while his firm's main strategy was statistical arbitrage and they still use quantitative-like strategies to inform all of their discretionary trading decisions they 'aren't a quantitative finance firm' because it isn't pure algorithmic.

He was very kind, and has been very helpful; I'm not trying to poke fun here, but I want to know what y'all think about this breakdown.
He also implied they only hire developers and 3rd year IB analysts, which doesn't seem to make a ton of sense.

 

[jarryds]

Discretionary trading + hiring IB people probably means it sits within the credit space, maybe high yield. HY is an interesting combination of complex problems that requires some hefty quantitative modeling, but is too nuanced and inefficient to run a systematic strategy. You need people on the floor who actually understand credit and its environment (ex-underwriting desk) and people who are capable of building multi-benchmark pricing model (devs).

How much actual stat arb exists in that corner of the market - I don't know. I have seen it talked about before: convertible arbitrage (which would probably be algorithmic), or arbitrage in CDS premiums and other swap spreads

But regardless of this being an example of credit, it is not uncommon to have quantitatively informed discretionary strategies. I know several other examples of funds that positions themselves to make money off of institutions clearing risk. The people actually trading are ex sell side and they are informed by a team of quants that model their ideas.

 

[MikeLawrence]

A lot of their portfolio is credit, yeah. But they are also listed on the QN list of quant employers so go figure.