@tigertrader , I knew you couldn't resist the allure of this thread!
From the very brief amount of information I have read so far, and from the info presented on this thread, I get the impression that it's not really a bang-for-the-buck issue, but rather an issue of what one's opinion is that he wants to express in a trade. If you think that the discretionary sector is set to outperform health care, you can express that opinion and not care whether both will rise or fall with the rising or falling tide of the broad market. That opinion can't be expressed by an outright long or short position in either sector. Maybe they will both suck, or both do great, but that's not what you care about in this case.
This sounds like a good point--and this was what I was saying earlier in the thread about using technical analysis for a spread trade--that intraday movements are, IMO, not really about fundamentals or supply and demand. As "proof," open interest in the ES contract rarely fluctuates more than a dozen or three per day; it's clear that the bulk of the volume that is trading intraday is short-term, day-trader/HFT noise that has zero to do with anything long-term fundamental. So your paragraph here really brings about the question of the method of analysis used to determine whether to buy or sell the spread.
I have not calculated any real examples of cross-margin discounts (would appreciate examples if anyone here has some), but I get the impression that one can get a quite large bang for the buck spreading versus outright, largely due to the very reason you mention--that the volatility is lower. I haven't run the numbers yet, but it seems with the margin discount, that one can actually get a similar bang for the buck compared to outrights, due to larger spread positions that can be held overnight.
You knew I was going to ask. Let me just guess a couple:
1) NOB - 10y / 30y
2) high yield or investment grade corp bonds (using JNK or LQD maybe) / 10y or similar US treasuries
3) 10 year Bund / US 10 year
4) Gold / something
5) Crude / something
6) S&P / Nasdaq,Dow?
As an aside, with a spread trade I really like the idea of having a simple loss limit on the trade (in dollars), in the event that I'm just wrong about the direction of the spread, without having to make that be about where the market will trade. You can't really get your stops run with a spread, or so it would seem, at least not in the traditional sense.
The following user says Thank You to josh for this post:
Favorite Futures: Retrospectroscope - for 20/20 Hindsight
Posts: 115 since Aug 2012
Thanks: 68 given,
Just had a quick skim over the thread, so apologies if I am repeating here. Have a look at some Aussie software called PairsTradeFinder.com as it might be what you're looking for.
Co-incidence (>0.9 ideally) is currently regarded as the best measure of a pair's likelihood of profitability on a reversion to the mean trading basis. You are looking for a pair to push out/stretch to at least 2 standard deviations from its mean and then look to trade a reversion until it crosses the 1 SD. Normal practice is to trade equal nominal value ie $1000 each side say, so divide each underlying's price into 1000 to give the position size. Typically stop losses are not used but rather time. One advantage of pairs/spreads trading is that you are not prone to being 'jabbed' out by market maker stop running games.
With Spreads you need to use line on close charts for prices
As for playing pairs/spreads on a directional basis I would suggest studying Wyckoff in order to 'read' the price action. Plenty of good threads on the methoidology here on futures.io (formerly BMT), but if you google 'Wyckoff Schematic' that will get you started. The added bonus is it will divorce you from a reliance on lagging indicators!
Hope that's of some help and good luck.
"Give me control of a nation's money and I care not who makes the laws."
Mayer Amschel Rothschild (1744 -1812)
The following 2 users say Thank You to SisyphusStone for this post:
@josh: If I want to express myself, I'll write a poem, or post a comment on futures.io (formerly BMT). I trade to make money and I execute in a way that will afford me the best chance to make the most amount of money. And, if one wants to do this for a living and not a hobby, then the sooner they implement this strategy, the better.
I wouldn't recommend reading books on the topic. The majority of them are written by authors who make their living from writing books, not trading. They tend to be theoretical and their validations are all done post hoc. I like to gain my insights from a) individuals that know more than me, and b) individuals who have a proven track record executing the methodologies I wish to learn; but most often through my own heuristic process.
Market relationships are constantly changing, the information is complex, and the learning process is implicit. So, your solution is not reading more articles, books, or threads; the value of any piece of information you read is probably zero or worse yet, misleading or irrelevant to current markets. The real solution is to do your own homework and make your own observations.
Nevertheless, here's a tip (remember, it's value is probably zero): Don't rely on single data points. Instead, look for consilience, or a convergence of uncorrelated signals, to provide strong conclusions.
Last edited by tigertrader; September 21st, 2013 at 09:59 AM.
The following 4 users say Thank You to tigertrader for this post:
Hi Josh - yes it's on CME/NYMEX. It's an exchange defined spread - meaning you can trade it as a product, and if you like leg out, or leg in, and trade out in 1 - or any combination there of.
I see your using IQ Feed. So do I. The symbol is qrbh14-qrbj14 (or if you use Ensign qrbh4-qrbj4). Spreads on CME/CBoT have the @ prefix, so corn will be @ck414-@cu14, wheat @wu13-@wu14. Crude spreads would be qclv13-qclu14, Nat Gas qng...... You get the picture I'm sure....
Sorry, no empirical evidence here - just observations and experience. Ive traded loads of crude spreads where the daily outrights have been stuck in a range, but the spreads have been trending like a MF'er.
The bold highlights seem to answer your question in a way? The nearest thing I could perhaps refer to is the MRCI results I provided a link to. Not ideal though - basically 180 spread trades over the course of 2012 making >$100k trading 1 lots v 180 outright trades making a big fat zilch - suggesting a propensity to trend more? interesting that this seems to suggest that the less risky spreads are returning more than the more risky/volatile outrights, implying that less risk does not necessarily equate to less return at all? Since 1990 (4140 spreads and 4140 outright trades) spreads returned $800,000, outrights a cool $1,000,000 (approx) Although the spreads returned $200k less, I'd rather be trading them as the reduced risk and margin would allow us to blow outrights out of the water assuming we were using the same capital base. Here's the link again: http://www.mrci.com/results/2012results/
I agree 100% with what you say with regards to intraday price and daily price. I was trying to say just this a while back - but not so eloquently as you
Last edited by TheDude; September 21st, 2013 at 10:41 AM.
The following 3 users say Thank You to TheDude for this post:
Where in the document do they state that the results are based on 1 lot only? I couldn't find a reference to that.
If it's really based on 1 lot, then it sounds too good to be true. Or, as I'm a believer of "no-free-lunches" I would expect the same strategies to generate a loss in the same magnitude in other years (or smaller losses for several years in a row). But then one has to question how much less risky these strategies really are…
Otherwise Mr. Moore and Mr. Toepke must be VERY rich men!
Unless explicitly stated otherwise, MRCI research always features spreads in a ratio of 1:1.
In such cases where the pricing and size of contracts differ, there is always the question of whether and how to balance it.
For the sake of simplicity, and because the industry refers to and quotes such spreads in terms of nominal price difference, MRCI has chosen 1:1.
Last Updated on Thursday, 03 January 2013 12:07'
They do emphasise these are hypothetical results. I dont think they are using any kind of risk management. The computer spits out optimised dates from past years and they blindly enter and exit accordingly. They always emphasise that the trader should apply his own management methods. These are not trade recommendations, just research. The results are for illustration. Could a trader do better with risk management and trade filters based on fundamentals or TA? Some of the drawdowns on these trades can be pretty big - so in other words, most would find trading like this quite challenging. Also there may be quite a few spreads on at the same time as some last for days, others 4 months+ so capital allocation is a non issue in the illustration. But if we did work out the amounts of initial and variation margin to trade every suggestion, would 100k be such a fantastic amount of return? 100k isnt really a huge amount of money in a year is it? It's kind of low for an average/good trader.
If you do the math, 100k is just over 8.3k on average per months worth of suggestions/trades. With 15 trades per month, that means the expectancy of a trade is approx $553. If the average holding period of each trade is 45 days, then thats just $12.3 per day per trade. Commissions and slippage arent factored in. It also assumes the trade was entered and covered at the daily settlement price - which is kind of tough to know in advance
Last edited by TheDude; September 21st, 2013 at 03:12 PM.
The following user says Thank You to TheDude for this post:
@tigertrader , veering a bit off topic here, but I'd like to briefly explore this.
I think you make an important distinction between expressing an opinion and seeing a trade opportunity. If I simply look at a spread ratio and think, "hey, the NQ has been really outperforming the ES lately, I think I'll sell the ES/NQ spread because it's likely to continue," then this seems to fall into the category of expressing an opinion. Kind of similar to what you said here:
So, you take a trade, not really because you want to express an opinion about what might happen, but because you see an opportunity, and want to exploit or seize that opportunity. I don't know if it would be going too far to say that something has to happen; maybe you could say that there is a catalyst that must occur that makes the opportunity something you want to act on. Either way, you are taking the trade because the opportunity for profit exists, not simply because you have an opinion and want to risk money to see if you are right. I think the distinction is important that you have made, if this is indeed the distinction.
The following 3 users say Thank You to josh for this post: