Do quants pose a threat to traders?

[d08]

Quants don't pose a threat but machine learning definitely does. ML can do whatever humans can do but better when it comes to pattern recognition. Neither discretionary traders nor automated system traders like myself have a chance against this, we are dinosaurs.

 

[vinc]

Quants ARE traders..even Simons describes himself as a trader..
a natural question arises : do trendline traders pose a thread to breakout traders??
do 5 minute candle traders pose a threat to hourly candles traders,etc,etc..
generally ,traders pose a threat to traders..

 

[d08]

Are quants not, by and large, traders? LOL.

In large firms quants develop models which is a full-time job by itself. They don't do execution. For example running inter-system correlation is completely disconnected from knowing which exchange to execute an order on at a particular time of day with which order type.

 

[spectastic]

Your honesty is refreshing, and your assesment in this case is spot on.


As @believezz already said, 'quant' is a much broader term than just 'quant traders', i.e. traders who use some kind of quantitative analysis as their primary tool for making trading decisions. There are quants who work on option pricing models, who work in risk management, as quant/developers, on credit risk models... the risk is endless.

In fact the number of quants working as traders (or perhaps more accurately in most cases, designing and managing trading systems) is pretty small. I just did a search on efinancialcareers for jobs with the word 'quant' in them. About 1600 results came up. I then filtered by trader, and only 140 survived the cut. And quite a few of those were 'trading' jobs, not 'trader' jobs; i.e. execution analysis for example.



So your second error is assuming that 'quant traders' focus excusively on the high frequency space. Again that is incorrect. Most high frequency firms are small and employ only a handful of people. There are a few large firms, not many. Ranged against that are quant behomoths like AQR, Rentech, Man Group and many more, all in the $10bn+ AUM range, each employing over 100 'quant traders', and all mostly focused on non HFT trading. And that's just simple maths; HFT can't employ large amounts of capital compared to slower strategies.



Yes you're wrong, and I would be delighted to correct you. There is automated and non-automated trading. There is systematic and discretionary trading. These distinctions are not the same. So we have:

1. Automated Systematic trading
2. Non-automated Systematic trading
3. Non-automated Discretionary trading

(of course we don't have 'automated discretionary trading' because you cannot automate discretion! And there are shades of grey here)

All HFT trading has to fall into the first bucket. But at large firms it might be that a trading system makes all the decisions, but some or all of the execution is done 'by hand'. I have friends who run systems in Excel, and then hand trade the positions. It doesn't matter. The important distinction is between trading systems and discretionary trading. The method of execution is less important. If you have a fund manager making discretionary trading decisions, and then using a sell side algo for the execution, ultimately it's still a discretionary trade (since the algo has no discretion).

Having explained that, are slower trading systems 'no different from hand traders' which to make it clearer I will translate to 'no different from discretionary traders'? They are different in many key ways, and the existence of many large quant funds running these kinds of system is evidence of that.



Finally, let me answer your question. The HFT space is to an extent a zero sum game, because the overall improvement in the market which some other posters have talked about is absent. So anyone who executes a trade is effectively paying a tax to the HFT traders. But this is no different from the tax paid to old fashioned market makers and floor specialists 30+ years ago (and actually, for smaller traders things are better because spreads have narrowed, even though for larger traders there is probably less depth in the book). It's a reward for providing liquidity (games with order flow and latency aside). Most people who are making money in markets are mostly earning risk premia, and the HFT guys are no expection.

GAT

Wow thanks!

 

[spectastic]

Quants don't pose a threat but machine learning definitely does. ML can do whatever humans can do but better when it comes to pattern recognition. Neither discretionary traders nor automated system traders like myself have a chance against this, we are dinosaurs.

Are you aware of developments of ai or ml in trading systems currently or for the forseeable future ?

The way I think about it, it's taking a ton of work just to get an ai to drive a car. They're probably already better. But I feel like there are so much more nuances that a trader needs to have that only comes from experience. How would an ai know to even look for certain connections?

 

[userque]

How would an ai know to even look for certain connections?

ML would test most/all possible connections.

Humans give it the data; and the ML crunches the data.

The more data; the more powerful the computing platform; the better.

 

[bone]

I've found that longer holding time frames helps. If you're a legitimate swing trader or position trader the turbulence caused by the algos and bots is largely minimized.

A scalper holding a mouse in his hand is dead meat on the table, IMO.

 

[Tradex]

I read this:
The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

What I found strange is that Jim Simons was very involved in the markets since his early 20s, but pretty much failed as a trader. Considering he is a genius, he only found success in his 50s, as technology had advanced enough by that time (and his staff at the time thought he was on the verge of suicide). He wanted to be like a Soros, but ended up like the people described in your recommended book.

I had a major realization: If Jim Simons had been born 40 years earlier, he would have died a bankrupt, and/or died by suicide. Think about it.

I have the feeling that you are a jetsetter with connections to famous and/or wealthy traders/authors, am I right?

 

[ChipShotTrader]

I have the feeling that you are a jetsetter with connections to famous and/or wealthy traders/authors, am I right?

That can be said. It was the direct result of the schools I attended, and also living in NYC for almost 20 years. There has definitely been some special encounters/events.

For example, while reading the book about Jim Simons, I noticed the name of Robert Mercer who was the CEO of Renaissance Technologies. His daughter's name is Heather Sue, and we took economics classes at Duke University, and even did homework together, while grabbing dinner. That whole time in school, I had zero idea her father was Jim Simons right hand man.

Julian Robertson's son, Spencer, also went to school with me, and I know him. Met his father at the Museum of Natural History, back in the day. Unfortunately, he was winding down his flagship fund, Tiger, and was not too happy.

I can go on for a couple of hours, as it really gets super surreal...

 

[Tradex]

That can be said. It was the direct result of the schools I attended, and also living in NYC for almost 20 years. There has definitely been some special encounters/events.

So you met the daughter of Robert Mercer (a billionaire) and Julian Robertson's son (also a billionaire), while in school, and both these men are well-known traders and fund managers.

No doubt, you are well-connected!

Did they share any trading secrets, by the way?

Just kidding.

Have a good weekend.