Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
We have clearly entered a new market regime that was marked by severe market volatility last week. Things have calmed down considerably this week, but we are still seeing significant moves. I did considerably well scalping ZN futures last week. I walked away with record profits even after making several significant trading mistakes. What is causing such unprecedented moves? I believe that the robots have finally run amok, and that makes now some of the best opportunities for retail traders.
There are two characteristics of this market that lead me to this conclusion. The first is a significant lack of liquidity. Before this move started it was common to see 3k offered at any level in ZN. At some major points such as before options expiration we were seeing 5k and even 10k's put on the DOM. Last week we were barely seeing 2k, and at some points they would only put up 100 on the inside levels. This is not the first time we've seen this. This is exactly the same behavior we saw during the flash crash in 2010. Some market making algorithms start losing so much money that the operators just turn them off completely, and liquidity dries up.
The second characteristic I see is extremely one sided trades. This is particularly pronounced in treasuries where despite huge flight to safety moves up, the overall trader positioning is short in record numbers. We also saw huge moves as the short volatility trade predictably blew up in everyone's face. This coincides with numerous research firms and quantitative analysis with models predicting significant drops in equities if the 10 year yield gets over 3%.
And this is where I get a little puzzled by the current market action. We've known since August when the market started to price in rate normalization that the Fed was going to raise rates. Absolutely nothing has changed since that time. It was literally only a matter of time that rates would get above 3%, and this isn't unusual by historical standards at all. Yet it's only now that we get to 3% that anyone sees this as a problem? We had a pretty good idea of where rates would be long before this latest rally really took off. Where was all the skeptics 6 months ago? It's not like anything magically changed last week.
What I see is a bunch of robots and quants freaking out over something they should have known would happen half a year ago. What's more is that they all seem to be responding the same way to the chaos. Hence we have huge moves up in volatility, and big short term moves up in treasuries.
And this only brings us back to something that has been discussed on this forum numerous times. Yes there are big advantages to robots. They provide liquidity, and are more consistent than humans. However, I believe we have made a mistake in letting them almost completely take over the market. I'm not just saying this as a retail trader mad at being crowded out by algo's either. I worked at a bank in technology writing execution and risk control systems. I'm somewhat familiar with the robots. I've been there when a "glitch" causes millions in losses. One of my jobs was basically to detect when such things were happening to try and cut the cord. So believe me when I say that there are limitations to this technology. There's nobody sitting there saying "hey this trade seems awfully crowded". Even worse there's not very many people left out there that even remember what trade was like during the last time we had rising rates. That was all long before the robots and quants took over. Not that this rate normalization after such an extended period of low rates can be compared to anything in the past.
My recommendation to wall street is to hire more traders. Bring back prop firms, and find people to develop hardcore trading skills. If things get this crazy when everything is supposedly going well, what do you think is going to happen when we're actually in a crisis? While the programmers are trying to figure out what went wrong with their machine learning algorithms I think there's some real opportunity out there.
Right, you are partially correct. I mention that I anticipated a multiple standard deviation down day in the month of Feb back in October (see the Bond forum). You are correct that there is more profit opportunity because the HFT liquidity robots have temporarily shut down because the risk and uncertainty was higher. However, I doubt they took big losses. In fact, there was a lot of evidence the HFT were trading very aggressively in the futures. So, they probably made even more. The environment didn't change to favor discretionary traders: although it was better in many ways. It changed from favoring HFT liquidity providers to HFT aggressive traders.
Possible winners:
1. HFT aggressive traders
2. Macro oriented or speculative traders on long volatility
3. Discretionary day traders and scalpers
Possible losers:
1. Systematic lower frequency traders and swing traders. One reported big losses among CTA's.
2. Short volatility traders (betting on statistical odds)
3, Market order retail auto-trading strategies.
A real concern was the market was trading too fast at times for a discretionary trader but it doesn't seem that anyone cares. There is very little retail trade in futures markets. Most of the trade is institutional and HFT. These are the firms that pay the exchange. I do not see anything really changing toward that way. It would be fairly simple to make it more level and easier for humans to trade. But, I don't think that will happen.
But, I think it is a valid question to ask for the independent trader whether or not it is easier to make profits from the markets statistically vs. opportunistically? It sounds like the "discretionary vs systems" debate but it is framed differently.
There are some good arguments that it might be easier to make money from the markets opportunistically. One argument is to assume markets are mostly efficient but not perfectly. If they are efficient, is it more likely that one would be able to find profitable opportunities by data mining and backtesting or is it more likely that one could find profits by scouring for unique and developing opportunities?
While, I shown that futures markets do not have to be negative sum when we allow for arbitrage and when considered on a participant vs. a contract basis, however there is a cost for every trade you make and with leverage and fees, if you make too many random trades in futures markets your account will go to zero fast. I have built profitable systematic strategies and even have a funded strategy account. Systematic trading can work. One of my strategies has performed well for about 4-5 years after released with no updates and with only one brief critical period. So, do I think historical prices and behavior are relevant? It is relevant. But, is it central? Is it core and central? Are 10 year old prices more central then the events and news today? Relevant, sure but central less likely. And, look at why most traders whether they are discretionary or systematic, they are trying to find rules that work consistently.
What I think of as "cutting edge" trading is when traders formulate original ideas and insights that have value and then fully quantitatively develop them and/or use systematic strategies to realize them. Traditional backtesting is formulated as trying to find a set of rules that work consistently in the past. But, a quantitative outlook, if you can come up with good ideas and then find ways to implement them and keep your trading frequency relatively low-- the question is -- is it going to be easier to make a profit that way?
It is difficult to answer. But, if markets are efficient, if trading might be zero sum, and given that markets are non stationary then all those things together suggest trying to trade the markets using a consistent, finite set of rules is less likely to yield superlative success.
But, right you are right that often there is a risk to reward ratio in anything. It is something that must be understood when making any amount of profit that the traders who capture opportunity earlier then later make the best returns. Most institutions cannot realize the "big returns" because they are too speculative. Right, how many institutions would touch Bitcoin when it was new. They needed more data. The HFT traders don't want to take any risk. They have a monopoly on those trades. Systematic traders, they find the historical trades that no one else wanted. What's left for the independent trader? Maybe to take advantage of opportunistic trades.
Now do you have to be able to predict? Yes but if you can find big R opportunities then you don't have to be perfect. The portrait suggest someone who trades on a lower frequency whether it is day trading or swing trading or whatever - we'll just imagine it is lower then they could trade. The trader is seeking out the bigger trades. They aren't looking to scalp a few ticks.
As an aside, I am trying to be more quantitative in my trading and define quantitative setups. Still building systems and still trying to trade the market on fast frequencies-- scalping, etc. I think perhaps some mix might work best for myself.
I find this argument counter to algos running the show rather interesting, especially the mention of GTAA (Global Tactical Asset Allocation)
Matt Z
Optimus Futures
There is a substantial risk of loss in futures trading. Past performance is not indicative of future results.
Trading futures and options involves substantial risk of loss and is not suitable for all investors. Past performance is not necessarily indicative of future results. You may lose more than your initial investment. All posts are opinions and do not claim to be facts. Please conduct your own due diligence. Use only Risk capital when trading Futures.
1 800 771 6748 local 561 367 8686 email [email protected]
Trading: Primarily Energy but also a little Equities, Fixed Income, Metals and Crypto.
Frequency: Many times daily
Duration: Never
Posts: 5,049 since Dec 2013
Thanks Given: 4,384
Thanks Received: 10,206
Isn't that an argument that contradicts your title? If in normal times the Algo's keep the market in order, but when volatility picks up they pull back and the market goes nuts, isn't it the lack of the presence of the robots that hurts the markets?
Do you think the robots have record short positions or is it PMs and Quants?
I'm not saying that I think your wrong, just that you gave two characteristics that I think oppose what you were claiming.
Not really. You don't really 'turn off', you just widen your spread to compensate for the increased risk.
During the 2007-2008 meltdown, a lot of electronic market making firms did turn off on one particular day, and 1 famed market maker chose not to, and they traded about 40% of the entire US equities market that day. And they were very profitable.
There's no certain argument that passive orders are disadvantaged vs aggressive orders in any scenario. If the hypothesis that "high vol/low liquidity => aggressive makes more money" were true, then this would be an amazing signal because it's so easy for anyone to trade it, and then the aggressive-passive arb would disappear - an apparent self-contradiction.
I fall on the side of this all being market conspiracy theory.
Bottom line is stocks reached a price where you can put your money in short term bonds and get a better return that the dividends in stocks.
As that looked set to unravel, all hell broke loose.
With the drop in liquidity, it doesn't take much size to throw the market 5 levels either direction.
Trying to read too much into that is futile. What I would recommend is that people who think this is robots running wild, go and watch the DAX, SPI or Hang Seng for a while.
My thoughts are that this is simply an illiquid market doing what illiquid markets do.
If you have any questions about the products or services provided, please send me a Private Message or use the futures.io " Ask Me Anything" thread
The lack of liquidity hurts markets, and if all of that liquidity is provided by robots then yes their absence hurts markets. However, you have to consider why they were turned off in the first place. They started screwing up so they got shut down. And yes they do turn the market making robots off sometimes. I have first hand knowledge of that from working on systems that detected out of control algo's and cut them off.
I believe the sequence of events went something like this.
Quants and longer term models say if treasury yields go up too much it's a problem.
Treasury yields move up because Federal Reserve communicated future policy clearly, and signs of rising inflation appeared.
Larger institutions try to reallocate their funds in a panic.
Moves cause liquidity providers to turn their scalping liquidity providing robots off.
Flash crash and volatility the rest of the week ensues.
And then you have today. What happened today? The treasuries had a trend down day and completely blew through their lows from last week. Yields moved up. Many levels that equities were worried about treasuries hitting were breached....and equities went up!
So now all of the sudden the market no longer cares about all of the things that it was panicking about last week. So we have two conflicting moves. I say that the one that pushed equities up today is the rational one, and all the strategies dumping stocks last week because of rising interest rates are on the wrong side of the trade.
So think of it this way. Automated trading and quant based strategies are a trade. Right now that trade is crowded, and last week's market action was the result. My theory is that if wall street wasn't so invested in this algo trade that the action last week wouldn't have been so volatile.
Sorry that logic is not credible on even a cursory analysis. I didn't say just any aggressive traders: I stated HFT. Most traders aren't HFT. If you look at the futures, faster aggressive HFT traders take the liquidity aggressively which causes passive liquidity providers to put out less liquidity which advantages the faster traders even more at expense of retail traders. This was really easy to see for anyone watching the market, just imagine the exact same system running on a co-located box vs. a retail platform. In less than a second the ES would move multiple points, the fast trader might capture $300-$500 or more in a second while the same strategy on retail would experience instead the loss in slippage. It doesn't have to be an arbitrage. It could be a directional strategy. The point is that in faster markets then the aggressive HFT have a greater relative advantage against other traders using market orders.
Right, it means the retail trader needs to adapt and change the way they trade. It creates new opportunity for other sorts of trading.
What I usually see is that the percentage of LFT-aggress-HFT-passive trades increases in very directional environments, so HFT participants on aggregate actually become more passive and provide more liquidity, which benefits retail traders.
