A swan strikes

General discussions about futures.
Post Reply
Jake Carriker
Site Admin
Site Admin
Posts: 1493
Joined: Fri Sep 12, 2003 10:32 am
Location: Austin, Texas

A swan strikes

Post by Jake Carriker » Fri Sep 03, 2004 2:16 pm

OK…so maybe it is not quite a black swan, but at the very least a dark gray one.

This morning when the employment report came out, some of the bond markets went extremely volatile temporarily. I had stop orders in to enter a long Eurodollar position along with contingent stop loss orders that activate upon execution of the entry order. The entry order is filled near the high of the volatile period. The exit stops are filled near the low. Also, I had stops in the market to protect a long 2 year note position. They were executed near the low during the same period. Price subsequently stabilized at a much higher level than the stop.

OK, there’s the sob story, now for the point. I have analyzed situations like this in my testing. I know that this kind of one day percentage loss is within the realms of what is possible. There have even been decent performance months with similar one day losses. Before I began trading systems I pondered whether some kind of discretionary override could improve performance during these periods. For a number of reasons I decided not to override the system at any time unless a drawdown exceeded my predetermined “bailoutâ€

Nathan
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Sun Jul 13, 2003 3:52 am

im don't think

Post by Nathan » Fri Sep 03, 2004 4:30 pm

I would not call this a black swan event, because the unusual price behavior in the minutes after the announcement was fairly predictable and even expected.

You can incorperate this into your method in many ways. For example, because the crazy price girations were expected, You can hold you orders for a predefined time after the announcement takes place, then place orders that need to be placed as a breakout of that post-announcment range. This is just one possible adjustment, not advice.

This is simply a tactic for dealing with a specific situation. It is a tactical adjustment and in my view not a violation of a strategy. Such things can be incorperated into your method so that you feel you are acting consistantly with your rules.

You have to make sure any tactical adjustment is consitant with the logic of the system.

leonardo
Roundtable Knight
Roundtable Knight
Posts: 100
Joined: Wed Apr 30, 2003 2:27 pm
Location: Minneapolis, MN
Contact:

Post by leonardo » Fri Sep 03, 2004 5:29 pm

Today I had sells in the bond market to execute if the market fell substantially; which it did. Since these were initiations, I treated them like stop-limit orders and got filled (actually filled better than my original sell price) when the market gyrated up after its nasty thrust down.

Like many, I exercise discretion on execution of a portion of my trades. But, I tactically differentiate between my handling of entries and exits.

Entries which are a "taking on" of risk, can and should be "managed"; while existing positions already at risk are subject to immediate action, often with distressing fills. :(

This technique tends to work in my "real world" trading much better than just putting in stop orders and "letting the market buck". But even if you don't finese either side of the trading transaction, you have to get out and in to be true to your method, or you don't have a method.

Leonardo---

MCT
Roundtable Knight
Roundtable Knight
Posts: 102
Joined: Fri May 16, 2003 7:27 pm

Post by MCT » Sat Sep 04, 2004 5:31 am

[quote]Before I began trading systems I pondered whether some kind of discretionary override could improve performance during these periods. For a number of reasons I decided not to override the system at any time unless a drawdown exceeded my predetermined “bailoutâ€

kianti
Roundtable Knight
Roundtable Knight
Posts: 335
Joined: Fri May 02, 2003 6:10 am
Location: Florence - Italy

Re: A swan strikes

Post by kianti » Sat Sep 04, 2004 7:43 am

[quote="Jake Carriker"]My query to the board is this: What do you do, and how does it work? Has anyone found that temporarily withholding orders at “announcementâ€

richard
Roundtable Fellow
Roundtable Fellow
Posts: 94
Joined: Tue May 18, 2004 10:17 pm

Post by richard » Sat Sep 04, 2004 10:34 am

I get out prior to the announcement and then if warranted get back in.

I find that starting around 8:25am eastern US time, there are sometimes wild ups and downs that can take out catastrophic stops I've set, and that have nothing to do with the overall market direction.

Tim Arnold
Site Admin
Site Admin
Posts: 8541
Joined: Tue Apr 06, 2004 1:41 pm
Location: Boston, MA
Contact:

Post by Tim Arnold » Sat Sep 04, 2004 12:45 pm

Hi Jake,

Great topic, as the same thing happened to me on Friday. I was long TU, TY, and US (and GC). So I got knocked out of the TU and US, but my TY stop was filled then busted. First for me. Also, my entry long stop in SF got hit and now I'm big in the hole on that one. Overall a 5% equity drawdown in 5 minutes! Grey Swan indead...

But the way I look at it is that I have have tested the system over the last 20 years, and this type of scenario has happened many times, so this is accounted for and expected as part of the system results. I trade exactly as the system indicates, both entries and exits, in order to trade what I test (and test what I trade).

But there is always room for improvement, so my plan is to see if there is a systematic way to test this theory that we should remove all stops on big news days. My guess is that this could result in a real black swan! But worth checking anyway. Don't know how "news days" are indicated in the historical data, but I'm looking into it.

Good luck and hold the course.

Tim

richard
Roundtable Fellow
Roundtable Fellow
Posts: 94
Joined: Tue May 18, 2004 10:17 pm

Post by richard » Sat Sep 04, 2004 9:55 pm

it does show you how correlated markets are. All the commodities and currencies and precious metals were hit (except the dollar of course.)

richard
Roundtable Fellow
Roundtable Fellow
Posts: 94
Joined: Tue May 18, 2004 10:17 pm

Post by richard » Sun Sep 05, 2004 5:50 pm

BTW, thinking more on it, we had a "price shock" on Friday, which are fairly common, not a "black swan" which is not common but is more common than commonly assumed :)

Forum Mgmnt
Roundtable Knight
Roundtable Knight
Posts: 1842
Joined: Tue Apr 15, 2003 11:02 am
Contact:

Post by Forum Mgmnt » Sun Sep 05, 2004 7:41 pm

Just to put things into perspective.

The first graph is Friday, which I'd call a bad day. It could have been a lot worse, and you better count on something like this or worse happening to you if you want to stay alive in this business.

I was short 1200 contracts of Eurodollars and 600 TBills for Richard Dennis during the spike for the second graph with a 30% personal interest in the profits. This is my worst Black Swan. I went from up 120% to down 5% overnight. For Eurodollars this was a $2,400 per contract overnight loss as it opened near the high.

The entire range of the top graph is only $700 per contract. The range of the bottom graph is three times greater and this occurred without any trading.

As a rule, if I have any sizeable position on, I would never leave stops with the market unless I knew the floor broker personally. If the market is screaming, you are better off waiting till the panic settles, if you place a market order you'll end up getting filled at the end of the panic and frenzy anyway. What I do is wait till it stops and then wait for the rebound, the minute the rebound stops, I get out with a limit order or a market order depending on my size.

Sometimes you get lucky and the rebound continues up through your original stop without turning around. If this happens, I just reinstate my original stop. This happens more often than you might think.

The thing to consider is that the price movement after a major run will be much more orderly since the stop orders have already been hit. So if it does go back down a second time, it will be more orderly and you'll generally end up with a decent fill.

Strategically, you need to obey your stops, but tactically the minute your price gets hit, you are really trying to get out at the best price possible. There's nothing about the stop price getting hit that says you need to execute a market order immediately, you are trying to get out but you are also trying to do so at the best price possible.

Rich spent at least a good hour on this in the original Turtle class.

That strategy probably saved me 15% on that dark October 87 day as I got filled 300 or 400 points off the high and slightly better than the open, saving me about 15% to 20% of my account.

- Forum Mgmnt
Attachments
BlackSwan.jpg
BlackSwan.jpg (50.95 KiB) Viewed 10120 times
PriceShock.jpg
PriceShock.jpg (28.83 KiB) Viewed 10117 times
Last edited by Forum Mgmnt on Tue Sep 07, 2004 11:24 am, edited 1 time in total.

Jake Carriker
Site Admin
Site Admin
Posts: 1493
Joined: Fri Sep 12, 2003 10:32 am
Location: Austin, Texas

Post by Jake Carriker » Tue Sep 07, 2004 11:13 am

I am back from the weekend, and I have had a chance to peruse the replies to my original post as well as some very thoughtful private messages I recieved.

Thanks for your input and keep it coming if you have more. I am doing some serious thinking and testing regarding "managing" entries and exits in a manner that is less prone than simple stops sitting in the market to exacerbate volatility.

There were a couple of key comments that are helping me to clarify my thinking on the matter:

Leonardo wrote:
Entries which are a "taking on" of risk, can and should be "managed"
c.f. wrote:
Strategically, you need to obey your stops, but tactically the minute your price gets hit, you are really trying to get out at the best price possible.
Now it is up to me to determine some criteria for testing these ideas at least somewhat systematically and possibly develop a procedure for consistent implementation. There are several questions to answer such as: "Under what circumstances do I "manage" entries and exits?", "What is the procedure for doing so?", "How much more 'box watching' will an alternative entry exit method require of me than my current method does?" There are several other items of concern as well, but you get the drift.

If I am to maintain the integrity of my systematic trading ideas, yet enter and exit the market in a "smarter" way I need to develop some rules that are clear and simple enough to implement consistently, yet effective enough to be worth the trouble.

Thanks for spurring some additional thought on the topic.

Best,
Jake

leonardo
Roundtable Knight
Roundtable Knight
Posts: 100
Joined: Wed Apr 30, 2003 2:27 pm
Location: Minneapolis, MN
Contact:

Post by leonardo » Tue Sep 07, 2004 2:10 pm

Utility theory raises its head again.

As has been mentioned elsewhere in the forum, the rewards one receives ultimately comes in the form of a trade-off.

Someone may be able to improve the reward vs. drawdown ratio with skillful handling, but at a time and opportunity cost to whatever else they could profitably be doing instead. (Watching a quote machine and doing the "handling" is terribly time consuming. I keep trying to figure out ways to cut that time down, as it mostly feels like a waste until you have the 4 alarm fire!!!)

If the account size is $10 Million plus; one may very well be willing to (or be forced to by utility concerns) a great amount of handling to limit risks and maximize their rewards.

But if someone is trading only one or two contracts per market, they might just as well put in orders with their broker in the morning and go on with whatever else they find profitable for their time.

If the principles on which the traded system is based and the structure of the market traded have not changed, then a big loss is just that.
-----------------------------------------------------------------

Thanks Forum Mgmnt, for the Eurodollar example of the chaos in October 1987.

Brings back memories. :shock:

Fortunately, in the previous Bond night session, I covered my last few shorts and went long a few contracts. After the huge open, I sold OTM calls against my Bonds and didn't trade bonds again for a month. I figured it was time for me calm down with the markets.

The devastation to many of my trading friends was enormous.

Leonardo----

Ghostrider
Roundtable Fellow
Roundtable Fellow
Posts: 83
Joined: Thu Jul 31, 2003 11:46 am
Location: USA

whipped

Post by Ghostrider » Tue Oct 26, 2004 12:33 pm

Hi Jake,
Someone says in this string, "I don't call this a black swan day.", and I agree.
I think the issue here is NOT how do you build rules around funny-mental release dates, it's about following your system when you get whipped... Kind of sounds like .......... Hotseat! What's the feeling? You know the drill.
Are you giving Fred wiggle room with these new 'rules'?

All the best and see you in November.

:P

Jake Carriker
Site Admin
Site Admin
Posts: 1493
Joined: Fri Sep 12, 2003 10:32 am
Location: Austin, Texas

Post by Jake Carriker » Tue Oct 26, 2004 12:45 pm

Right on, GR. I know the drill. I end up taking this issue into TTP at the next opportunity, and I don't change my strategy a bit.

Sometimes a loss is just a loss. I deterimine that my trading strategy does not benefit from changing the system based on this event, and I trade through to today without missing a signal. Seems like things are just fine.

Thanks for your support!

Jake

Ghostrider
Roundtable Fellow
Roundtable Fellow
Posts: 83
Joined: Thu Jul 31, 2003 11:46 am
Location: USA

Good Job!

Post by Ghostrider » Tue Oct 26, 2004 12:49 pm

YES!!

:P

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Tue Oct 26, 2004 5:59 pm

I particularly like c.f.'s and Leonardo's replies.

On a discretionary basis I trade news events buying retracements in the news with tight stops on a very short timeframe so I see much value in c.f.'s stop and seek best exit strategy. I buy as he exits and sell not to long afterwards. Whether c.f.'s strategy is a good one depends on how far out of the most recent consolidation his stop is of course but I expect it would normally be far enough.

On a systems basis I would say to Leonardo that withholding orders from the market before major news events doesnt have to be counter to your system. As long as its included in your system. So, perhaps you have a rule that says "no extries on Greenspeach days."

John

Post Reply