Why the S&P E-Mini Stinks

[steveh2009 10]

If you're a small trader, the CL is the best contract.

[TheNegotiator 13]

If you're a small trader, the CL is the best contract.

 

Exactly why do you think this is true?

[:::grimReaper::: 10]

ES

26k/400 = 65

 

TF:

1500/30 = 50

 

YM:

1500/25 = 60

 

NQ:

3100/40 = 78

 

I made a few trades on the YM this week and some in December. I feel there is more flexibility in entering/exiting trades. And if one's scalping (profit target ~1pts on ES), they should obviously trade YM, b/c the YM will go 10pts, giving 10 areas to exit the trade, vs. 4 ticks on the ES. In fact I looked at 5min bars from 11/28 (after Thanksgiving) to 12/21 (when volume died out) during RTH, the YM has a range of 16 ticks on average while the ES has an average of 8 ticks.

 

However, as mentioned a few times itt, ES is more technical (S/R) and suits me. Since I'm pretty sure YM and ES are the most closely correlated equity futures, hypothetically you could trade YM looking at the ES, which was sort of what I was doing, but it felt awkward and there's less screen space. I'm looking forward to going solely back to ES.

[Tams 20]

I made a few trades on the YM this week and some in December. I feel there is more flexibility in entering/exiting trades. And if one's scalping (profit target ~1pts on ES), they should obviously trade YM, b/c the YM will go 10pts, giving 10 areas to exit the trade, vs. 4 ticks on the ES. In fact I looked at 5min bars from 11/28 (after Thanksgiving) to 12/21 (when volume died out) during RTH, the YM has a range of 16 ticks on average while the ES has an average of 8 ticks.

 

However, as mentioned a few times itt, ES is more technical (S/R) and suits me. Since I'm pretty sure YM and ES are the most closely correlated equity futures, hypothetically you could trade YM looking at the ES, which was sort of what I was doing, but it felt awkward and there's less screen space. I'm looking forward to going solely back to ES.

 

You would seldom get a 1 tick spread on YM.

a spread of 2 ticks is the norm.

which still gives you a slightly better slippage than ES, but not much.

[:::grimReaper::: 10]

You would seldom get a 1 tick spread on YM.

a spread of 2 ticks is the norm.

which still gives you a slightly better slippage than ES, but not much.

 

True. I believe the spread is 2 55-75% of the time and 1 the rest of the time. But I'm not that considered with a spread of 2, I can place an order inside the spread and be first to be filled.

 

I wish the ticks were 0.20 on the S&P. Small change, but it would it allow better fills and more # of ticks to play with, while likely still maintaining 1 tick spread. Has CME ever made changes like this?

[Predictor 11]

Hopefully not.. a smaller tick would invite more HFT trading. The spread does not equal the liquidity available in a product.

 

True. I believe the spread is 2 55-75% of the time and 1 the rest of the time. But I'm not that considered with a spread of 2, I can place an order inside the spread and be first to be filled.

 

I wish the ticks were 0.20 on the S&P. Small change, but it would it allow better fills and more # of ticks to play with, while likely still maintaining 1 tick spread. Has CME ever made changes like this?

[:::grimReaper::: 10]

Hopefully not.. a smaller tick would invite more HFT trading.

 

I doubt it, I believe HFT have an edge (you're implying) when the spread is several ticks wide. 0.20 ticks are large enough to seal the spread.

 

The spread does not equal the liquidity available in a product.

 

I wasn't saying otherwise. I mentioned "spread of one" because I believe the reason the tick size are so large is to allow the big boys to trade with minimal slippage. As I recently found out, the tick size use to be 0.10 and the contract size us to be $250. Increased the tick size and decreased contract size is consistent with my theory.

 

Btw, do you believe ES is tradable if the tick sizes were 0.50 or 1 point?

[:::grimReaper::: 10]

Does anyone know which matching algorithm CME uses? Here's a list: list:http://ftp.cmegroup.com/globex/introduction/features-and-functionality/elements/matching-algorithms.html

 

And do those who have seats have priority in getting filled?

[GDR 10]

The matching algorithm is chosen by the platform provider, is it not? e.g. NT uses FIFO

[:::grimReaper::: 10]

No need to worry:

http://www.cmegroup.com/globex/files/PriceBanding.pdf

 

The matching algorithm is chosen by the platform provider, is it not? e.g. NT uses FIFO

 

The FIFO in NT is just an accounting mechanism that you can choose to use or not when scaling in/out contracts. The FIFO in the context of matching algorithms means it's first-come-first-serve in getting your limit orders filled.

[GDR 10]

Thanks for the links, these will help a lot in explaining how the different contracts fill.

I know people who lease and from what I can gather they get the same speeds we do, because their data is still fed through the same servers we use, if we were at the same broker as they use. I've seen them test speeds before trading when using their laptops on the road. They are leasing to reduce brokerage costs, but still have many of the same issues we do, and still pay exchange fees.

I've heard many times that some of the HFT's are owned by high level people at the CME and that have co-located servers in Chicago, that helps them get the speeds and fills that they do. I would imagine that some seat owners would also have their data servers, if they owned their own, located at close as possible to the exchange as well. It a fairly complex web of information to grasp, I would guess, and one tries to understand the routing that is used by say a brokerage buying stocks for their corporate account vs a small retail trader trying to execute over a filtered data feed on an ancient platform, delivered to a market maker ( stocks) offered by many of the larger investment firms. This is an extreme example meant to show the range of processing that is available out there, and the minefield a trader stumbles through in learning the business.

Does this makes sense to anyone on this thread?

Thanks for the inputs.

[BlueHorseshoe 10]

I've heard many times that some of the HFT's are owned by high level people at the CME and that have co-located servers in Chicago, that helps them get the speeds and fills that they do. I would imagine that some seat owners would also have their data servers, if they owned their own, located at close as possible to the exchange as well.

Does this makes sense to anyone on this thread?

Thanks for the inputs.

 

Colocation of servers, not just in Chicago, but actually within the exchange buildings, is standard procedure amongst the HFTs. This has nothing to do with any kind of exchange corruption, per se, or 'high level people at the CME', but simply the priorities and resources of the HFT firms.

 

A few encouraging things to bear in mind, however, are the following:

 

HFT firms operate most extensively in individual stocks, where the superior speed of execution has largely usurped traditional market-making. When they do operate within futures markets such as the @ES, it's largely in the mode of arbitrageur, hedging off a position against a basket of underlying stocks (or a similar contract such as the SPY).

 

The HFT firms, as the name suggests, employ 'high frequency' strategies, and seldom hold positions for more than a few seconds. Their trading models are incredibly sophisticated and are very far removed from anything that you (or I) are likely to be using. In this sense, their demands are unlikely to intersect with yours, the retail trader. So although they might want to have an order filled at a particular price that is the same as yours, it's not because they're going through the same trade decision making process as you, but then somehow getting in there faster.

 

Worrying about the actions and effect of high frequency trading is a bit like an athlete about to run a marathon worrying whether certain celular metabolic processes will take place efficiently within his body. HFT occurs on a 'microscopic' level far beneath that of day traders (and that's whether you trade momentum, volatility, reversion to the mean . . . anything), and apart from the fact that it makes contracts highly liquid it can be ignored. To look at it from a different point of view, do you really think that the big trend-following funds, who hold positions for many months or even years, are interested in what a daytrader is doing to affect the market at any particular time, or fretting over the few ticks of slippage that result?

 

I hope this post doesn't sound too confrontational - I'm just trying to offer a helpful perspective (plus, it's my first post on here!).

[TheDude 10]

I doubt it, I believe HFT have an edge (you're implying) when the spread is several ticks wide. 0.20 ticks are large enough to seal the spread.

 

 

 

I wasn't saying otherwise. I mentioned "spread of one" because I believe the reason the tick size are so large is to allow the big boys to trade with minimal slippage. As I recently found out, the tick size use to be 0.10 and the contract size us to be $250. Increased the tick size and decreased contract size is consistent with my theory.

 

Btw, do you believe ES is tradable if the tick sizes were 0.50 or 1 point?

 

Nope, other way round.

 

Buy-side prefer smaller ticks as it allows them to hedge more accurately.

HFT/locals prefer bigger ticks as it allows them to make more money on smaller moves.

 

this is one reason the ES took off - the larger spread/tick in es v big S&P allowed arb traders (locals) to take advantage of inefficiencies/difference in the contracts.

 

look at liffe and the euribor as a prime example - they introduced half ticks on demand of the banks and hedge funds. all of a sudden, the locals ran off to trade eurodollar and crude until liffe reintroduced fat ticks

[GDR 10]

Yes !! Agreed, 100 %

If you stand in the ES pit at the CME you can see the locals trading the spread, watching the S&P quotes on the screen. The big brokerages do the same for their big customers with their runners in the pits working the diff. Many is the screen trader that thought he got his stop run, by what was really a bunch locals working the spread.

[GDR 10]

Colocation of servers, not just in Chicago, but actually within the exchange buildings, is standard procedure amongst the HFTs. This has nothing to do with any kind of exchange corruption, per se, or 'high level people at the CME', but simply the priorities and resources of the HFT firms.

 

 

HFT firms operate most extensively in individual stocks, where the superior speed of execution has largely usurped traditional market-making. When they do operate within futures markets such as the @ES, it's largely in the mode of arbitrageur, hedging off a position against a basket of underlying stocks (or a similar contract such as the SPY).

 

The HFT firms, as the name suggests, employ 'high frequency' strategies, and seldom hold positions for more than a few seconds. Their trading models are incredibly sophisticated and are very far removed from anything that you (or I) are likely to be using. In this sense, their demands are unlikely to intersect with yours, the retail trader. So although they might want to have an order filled at a particular price that is the same as yours, it's not because they're going through the same trade decision making process as you, but then somehow getting in there faster.

 

Worrying about the actions and effect of high frequency trading is a bit like an athlete about to run a marathon worrying whether certain celular metabolic processes will take place efficiently within his body. HFT occurs on a 'microscopic' level far beneath that of day traders (and that's whether you trade momentum, volatility, reversion to the mean . . . anything), and apart from the fact that it makes contracts highly liquid it can be ignored. To look at it from a different point of view, do you really think that the big trend-following funds, who hold positions for many months or even years, are interested in what a daytrader is doing to affect the market at any particular time, or fretting over the few ticks of slippage that result?

 

Thank you for your thoughts, much appreciated ! Here are a few more of mine.

Some of the directors of the CME are owners of those HFT outfits, so I see a bit of a conflict there, which I will try to address. I do not think or mean to imply corruption, simply an advantage that a retail trader should be aware of, especially scalpers.

I completely agree with your remarks about the HFT, hedgers etc not caring about retail traders, and never meant to imply that they did, quite the opposite, they are just doing their thing, and we can inadvertently get taken for a ride in the wrong direction. Anyone with an unfiltered feed can see HFT's operating all day long in the ES contract. You will see the volume spikes that really stand out, and the most obvious thing to me, is the equal number of buys and sells. They occur regularly and without fail. A daytrader trying to make ticks in an around one of these events is going to have trouble getting fills, would they not? I trade off a 5 min, but use a 1 min for entries and exits, and am very cautious about placing an order when expecting another HFT volume spike. I have evolved my trading to working a longer time frame and larger moves in contracts like the ES, going for parts of the 10 pt moves that happen once or twice a day, and the more obvious 3-4 pters that come up.

More dramatic examples are seen in gas and oil trading, where speed is essential for good fills and the server location is a big advantage. Again I'm not implying corruption, just stating that there are advantages that even owning a seat cannot overcome, if you aren't very well financed to acquire the hardware etc to get the fills you think you should get looking at the DOM.

I am very curious how the role of arbitageur could explain the equal number of buys and sells in these volume spikes ? Does anyone know about this? If they are in and out quickly where is the arbitrage in a trade like that?

Thanks for the input!

[:::grimReaper::: 10]

HFT/locals prefer bigger ticks as it allows them to make more money on smaller moves.

 

However, don't you have more "smaller moves" the smaller the ticks are? For example, if ES had 0.01 ticks, price could fluctuate +/- 0.10 several times, whereas our 0.25-tick ES wouldn't budge.

[bobcollett 10]

[quote name=

I trade off a 5 min, but use a 1 min for entries and exits, and am very cautious about placing an order when expecting another HFT volume spike.![/quote]

 

Hi GDR

Great post!!

How do you anticipate a HFT volume spike?

regards

bobc

[GDR 10]

Bobcollett: Run a 1 min chart with volume histo at the bottom, and go back for 5 days or so and watch for the 8K+ spikes in vol. You will start to see a pattern to the number of minutes between trades and the location in the movement of the index. Rarely moves in a large candle , hard charging sequence, often picks quieter times when it looks like nothing is going on. Only happens on good vol days +500k near the open min. It's the "art" part of trading, not hard to learn, takes a little practice. There's an algo running, I've not taken the time to try to figure it out, by the time goofs like me get close, they will change it. It appears to be a volume related algo.

[SunTrader 10]

The majority of traders lose.

 

The majority of traders try to trade like, anticipate or just generally concern themselves with the big money players, i.e. HFT's, Hedgies, etc.

[BlueHorseshoe 10]

 

Some of the directors of the CME are owners of those HFT outfits, so I see a bit of a conflict there, which I will try to address.

 

I am very curious how the role of arbitageur could explain the equal number of buys and sells in these volume spikes ? Does anyone know about this? If they are in and out quickly where is the arbitrage in a trade like that?

Thanks for the input!

 

I believe that, as well as directors of the exchanges, many HFT firms are also operated by former super-successful floor traders (the likes of Tom Baldwin etc). I imagine that they would benefit from the same advantages in terms of influence at the exchanges.

 

I'm afraid that I am not able to answer your question about arbitrage, although I would hazard a guess that though they're in and out incredibly quickly, any small profit made is greatly magnified by the number of contracts traded and frequency with which the strategy is implemented.

 

Your remarks about gas and heating oil are interesting - I'm not familiar with these instruments. Pressumably they're still relatively heavily pit-traded? And pressumably the thin liquidity makes any institutional activity more conspicuous?

[BlueHorseshoe 10]

The majority of traders lose.

 

The majority of traders try to trade like, anticipate or just generally concern themselves with the big money players, i.e. HFT's, Hedgies, etc.

 

I think that there are two main issues with doing so:

 

1) The strategies used by HFTs, hedge funds, floor traders etc are incredibly diverse, so unless you're able to distinguish and follow a particular strategy, then there's going to be no consistency to your methodology. This is one of the issues that I have with VSA - yes, 'smart money' is coming into the market, but you've no idea why - is it because a particular fund is looking to position itself for a trend that it expects to last twelve months (in which case, are you going to hold your position for twelve months as well?), or is a HFT algorithm scalping the market for a single tick (in which case there's little point you acting upon this information hoping for a ten tick gain).

 

2) A lot of funds and, in particular, commercial hedgers, have very deep pockets. A trend following fund that uses a 'stop and reverse' type strategy, such as Dunn Capital, will allow a position to drift against them by a very significant amount. Can you afford to do this with you daytrading account?

 

These are two good reasons, I reckon, for not becoming overly concerned with what the 'big money players' are doing, as Sun Trader suggests.

[mingamonkey 10]

I think that there are two main issues with doing so:

 

1) The strategies used by HFTs, hedge funds, floor traders etc are incredibly diverse, so unless you're able to distinguish and follow a particular strategy, then there's going to be no consistency to your methodology. This is one of the issues that I have with VSA - yes, 'smart money' is coming into the market, but you've no idea why - is it because a particular fund is looking to position itself for a trend that it expects to last twelve months (in which case, are you going to hold your position for twelve months as well?), or is a HFT algorithm scalping the market for a single tick (in which case there's little point you acting upon this information hoping for a ten tick gain).

 

2) A lot of funds and, in particular, commercial hedgers, have very deep pockets. A trend following fund that uses a 'stop and reverse' type strategy, such as Dunn Capital, will allow a position to drift against them by a very significant amount. Can you afford to do this with you daytrading account?

 

These are two good reasons, I reckon, for not becoming overly concerned with what the 'big money players' are doing, as Sun Trader suggests.

 

Being a new trader myself, I was highly interested in how HFT might affect me, but like many of you are mentioning, I'm not really concerning myself with it too much and simply watching the price action with three time periods.

 

Horizontally I trade the 1 min, for the larger point moves I reference a 4 min, but one of the cool things I like to watch is a 100 tick. I call it my peek inside and you can see a lot of info from it. You can actually see the big lot sizes of either bigger pro amateurs or multiple sequential 100+ lots coming in. Every once in a while, you'll see a 1000+ shoot in on the 50 or 100 tick time frame.

 

Yet another way to get more insight - at least on my personal strategy.

[joshdance 10]

This is one of the issues that I have with VSA - yes, 'smart money' is coming into the market, but you've no idea why - is it because a particular fund is looking to position itself for a trend that it expects to last twelve months (in which case, are you going to hold your position for twelve months as well?), or is a HFT algorithm scalping the market for a single tick (in which case there's little point you acting upon this information hoping for a ten tick gain).

 

I find the whole "smart money" quest to be a waste of time. However, when heavy volume trades, it does mean that more people, whether it's a lot of small traders, or a few large traders, or some combination, have taken an interest in a particular price area. True, in all cases a big move does not follow, but at least it does indicate interest.

[BlueHorseshoe 10]

However, when heavy volume trades, it does mean that more people, whether it's a lot of small traders, or a few large traders, or some combination, have taken an interest in a particular price area.

 

This is a very fair point - I didn't mean to dismiss the relevance of volume data altogether!

[phoenix01 10]

Ive been trading ES for a few months now and been have some success. However recently due to volatility coming down the range was become smaller and its hard to get filled 1x my risk and 2x my risk (I usually no more than 2 points a trades, targets +2 and +4). Which wasn't difficult the last couple of months of 2011.

 

So recently i was looking at other markets and the NQ seem to have better results. When risking the same candle ( going long - one tick below the candle, vice versa for short), it seems to reach 1x and 2x risk more of the time.

 

So i was thinking i could execute trades through the NQ but continue my analysis on ES. However the catch is sometimes a trigger on the ES is not the exact same point on the NQ. Sometimes on it will go one tick lower or completely do the opposite and actually move down whilst the ES slowly pushes up. Which will result in losses which would have not have occurred via the ES.

 

Iam sure other trades have looked over this and was wondering what they concluded was the best thing to do? Does execution through different but correlated markets work or should i stick to what I've got and just reduce targets during low volatility (Which is a lot harder than it sounds). At the end of the day the better risk to reward ratio the easier it is to be profitable.