@s0mmi , or anyone else who trades futures on faster time-frames and cares to weigh in;
I don't have a specific question per se, but something I've wondered about from time to time is how people settle on a maximum clip size. At this point my interest is still academic, but I'd be interested in hearing your 2c.
How many contract in a clip can one get up until it starts to produce an affect on their strategy, assuming they're doing live day-trading, reacting to levels and relatively low TF bars like you're talking about? One of the guys that taught the paradigm I use in years past limited his size to 40 ES. I also know his mentor used to clip 100 ES at times, and he also said he's done 50 ES each with five linked accounts, so 250 total per order. I know they stopped scaling up at that point, but I don't know why.
Assuming proper capitalization, risk control, etc... are psychology and slippage the only real limiting factors for size constraints or at some point do other participants start to notice and react to size once it gets past a certain point? You sound rather successful trading futures; have you personally hit a max clip size or are you still scaling up your strategies? What are the factors that goes into a decision like that? Is clip size in relation to instrument liquidity relatively linear (e.g. 10 NQ contracts will behave about the same as 100 ES contracts) or are there other considerations?
Thanks,
llIHeroic
I have watched the primary cash activity periods of the S&P500 and can definitely tell you that doing 250 ES is absolutely nothing. It does a lot of volume and can absorb that order instantly. You could easily scalp on 250 ES if you knew what you were doing.
When it comes to size, it depends on your strategy, time-frame and market.
Australian 10-year Bond
For example, in the Australian 10-year bond I was at point doing clips of 100. But when it came to exit, if I had collected just 200 on, I would find it extremely hard to get filled in 2016 compared to 2011-2015. My slippage was enormous because it moves in half basis-point, so my slippage for exiting was my entire edge itself. This was because I was day-trading it and basically market-making during its off-session.
Australian SPI Equity Index
Another example is in the Australian SPI equity index overnight, where I was getting 40-lot clips on and collecting to about 80-100 sometimes, but the bids and offers were about 5-15 lot. I had basically maxed the market and its mean reversion rate went down due to more algorithms coming in and willing to take less profit on every move.
The only way for me to get out was either spew market and lose a months worth (at least), or hold into the next days morning auction. This turned out to be beyond dangerous eventually (after being safe for a long time). The daily night equity range in the spread with the U.S. S&P500 might be like 10 ticks, but in the first 15-seconds of the auction it can easily gap 7 ticks away and travel another 5-7. You can literally lose a week to a fortnights worth off nothing.
In contrast to the thinner Australian markets, the S&P500/ES can easily eat up a clip of 100 with no problem. You could use the iceberg feature (hidden order) to get off ~300 contracts somewhere without an issue.
By the way, I find it very suspicious somebody would have linked accounts to trade futures. You have the same margin requirement on each contract if you are retailing so it makes no sense. Alarms should go off in your head when people talk about doing the same strategy but on multiple accounts. This is only useful if you plan on going bankrupt in one account and making it in another to run off with your money. Also, the argument of "I don't want my broker to know my position" makes absolutely no sense. 250 ES is nothing because yesterday it traded 318,000 contracts in the first 30-minutes of Cash.
When it comes to problematic sizing in a market, it is really beyond the scope of most people on this forum. You will be surprised, though, at how fraudulent some of the strategies used when it comes to 'big size traders'. It is not uncommon to always be risking 3-months to 6-months of profits at any time because of the nature of market making and what they do on their size. I will explain a bit more further down.
I can say that for almost everything you will ever glance your eyes on that is reasonable to trade, there will be enough volume and size to trade you out for your time-frame because of the enormous nature of the market.
Thoughts on maximum clip and hitting it in the Australian Market
I definitely hit a maximum clip in the Australian markets in both Bonds and Equities. To summarize the story, over the years what happened is more players entered the game slowly over time, and they are all sh*t c*nts. When I call them a sh*t c*nt what I mean is, they were happy (using algorithms) to take a 55% win rate in most of the products because of their market making strategy implemented over 10 to God knows how many other products and 'theories' of spreads. It started with with less players doing it 5 years ago, and having such a strategy was yielding 80-95% win rate for the risk taken.
Put simply, 3-years ago if an order came in to send a product 10 ticks down you easily faded for 3-5 tick retracement. This has now changed; an order only sends it down 5 ticks now and comes back for a 1-2 tick retracement. This doesn't seem too bad at first, because your profit is a bit more than halved but your risk is also halved. This is the fallacy with algorithmic trading and market making things into oblivion; when these things 'blow out' then it does not matter what the current volatility is like. All that matters is everyone who collected, is now insanely offside and can't afford to lose their year because they've already lost their month.
If it doesn't want to come back that day, all bets are off the table when it comes to the question of "How far can this go?". This is the unfortunate nature of reaching a maximum clip in a market; you will always be collected full-size when you are wrong, but barely be collected when you are right. Therefore, you need to find a strategy with an edge. What would an example of such a strategy and edge be?
A quick example would be trend-following a Yield Curve when inflation has kicked in, and you have strong belief that the Central Bank will continue hiking interest rates and build those into expectations. Selling the short end w/ buying the long end, of whatever Country you are trading (Australia's 3/10yr or U.S. 5yr/10yr). This is the type of strategy that banks put on. A typical bank trade could even take a few days to unwind out of their position because they are so big!
Most individual people who trade big size also have very large swings. They do not have the bank order flow, or insider information, just a 'feel' for the market.
To clarify, having a swing of -$100,000 means nothing by itself. If you know your monthly average return is $400,000 per month for the last quarter, than -$100,000 is not that bad for an average positional trader who doesn't like cutting losses. The problem arises when most algorithms are taking a 55% win rate on market-making order flow, because it forces a swing trader who gets size on to probably now swing -$200,000 to -$300,000. As a function of your month, you used to risk 1/4 of it (a week) in a hairy situation, but now you are risking 1/2 (fortnight) to 3/4 (near a month). In my experience, most traders do not know this is happening but they do feel emotional pain in the gut as to why things aren't working. You pray that this period stops here, but unfortunately the trend of over-saturation has continued in many examples I can list; which means you end up risking an entire month or quarter sometimes because the market has evolved and changed to accomodate for the bottom-feeders in there.
A lot of people will eventually bleed to death, blow up or just burn out mentally because they cannot handle these new numbers. The answer is to drop your ego and size down, or reset your entire trading paradigm, unlearn your current strategies and relearn new ones. This is much easier said than done, trust me.
For all my examples the evidence of this is that there is now 2 times more volume (compared to 2012-2013), but there are now, according to the bids and offers, 5 times as many players willing to sit everywhere and auto-spread.
Considerations for maximum clip
What I have just told you is to give you an example of a Market-Making strategy that has a limit to size if you want to keep things nimble and reasonable for positive EV. Market-Making strategies are the simplest and easiest ways to trade. This is because most people are banking on volatility holding stable, looking at a chart, and auto-collecting somewhere based on a theory they have. Therefore, you are limited by how much size the order (or flow of orders) will be.
This is, in actuality, no different to scalping the ES. If you want to sell at a level, and you see someone buy 1000 at one price and it remains offered, then another 1000, and you now offer, then you are hoping the next batch of order will fill your size. If you offered an enormous amount like 1000, you probably won't get it in the ES if you are right about the next move. But if you offered 50 ES, you'll probably get it.
One more thing
Clip size in relation to liquidity can be transferred if you have a general feel for both markets. For example, I would say having 5 ES on ($62.50/tick), is like having 1 HHI (Chinese H-Share) ($8.50/tick). But it ends there. The strategies used, players, and behavior of both markets will always be different. The best way to see this applied in the real world is to take note of the fact that. At the end of the day, it is never a size excuse to be losing money.
Sure, I might not be able to make $100,000 a month doing what I used to do, but I could still make $20,000-$40,000 by changing the execution style and a few parameters of the time frame.
The only question you have to eventually ask yourself is, is the risk/reward and time required worth the effort or can I spend time and capital better elsewhere?