Wilmott Forums - Split-Strike Conversion Strategy

Forum Navigation:

login

FORUMS > General Forum

Topic Title: Split-Strike Conversion Strategy
Created On Sat Mar 08, 08 08:55 PM

Topic View:

View thread in raw text format

JohnnyQ
Junior Member

Posts: 5
Joined: Nov 2002

Sat Mar 08, 08 08:55 PM

User is offline

Hi,

There was a previous tread on Wilmott describing a Split-Strike Conversion Strategy.

In essence, the strategy goes long a subset of securities from a liquid equity index (say, 35-50 stocks out of the S&P 100 Index). Then there is a collar overlay strategy on the index itself (buy out-of-the-money S&P 100 put and sell out-of-the-money S&P 100 call). The strikes of the options can vary over time but typically they are not more than 5% out-of-the-money and are rolled over frequently (1-3 month duration). The notional amounts of the collar are equivalent to the notional amount of the underlying long equity portfolio.

In analyzing the historical returns the strategy has virtually no drawdowns .... and very consistent performance (7-9% per annum with less than 2% volatility). The strategy is implemented by a proprietary trading group at Madoff Securities (a large securities market maker as well). Without actively timing the market (ie, short term long/short bias) I am wondering how you can get that kind of risk/return out of the strategy? If you are activley timing the market, surely you will have some losing bets .... If you are passive, even though the moneyness of the collar can vary, it seems to me that the volatility skew works against you and you will be forced to give up more upside for less downside protection ....

So, I am at a loss ... Can anyone shed more light on this or has anyone heard of the this Split-Strike Conversion Strategy before?

Thanks,

JohnnyQ

-------------------------
JohnnyQ

khoofoo
Junior Member

Posts: 1
Joined: Sep 2008

Fri Dec 12, 08 06:59 PM

User is offline

perhaps the news that came out today regarding Madoff Securities might help clear this up for you?

nice work spotting something that was too good to be true!

fars1d3s
Senior Member

Posts: 452
Joined: Aug 2004

Fri Dec 12, 08 07:23 PM

User is offline

Here is my attempt, since I myself don't employ this strategy. Please do not blame me if you actually employ my tactics and it doesn't return a satisfactory performance:

To trade these out-of-the-money options, they usually have a "window" of a few days, where they take their time to complete their positioning. It's timing technically, but realistically it's not considered "timing" because the positioning must be disciplined and done within a "strict guideline". Also, they can legally engage in selective market-making with any counterparty within this "window of opportunity". As a practical matter, these 35-50 stocks are not necessarily set in stone .... they can switch in and out of some of them from time to time, or vary the relative weighting from time to time, etc ... Call it "timing" if you wish, but I bet it's called "discretionary adjustments" if you will.

So, the actual result is they're short calls at 2-3 different strikes bunched up in the general area of, say, 5% OTM. Same with their long puts. Do this every month, for 12 months a year, while collecting dividends on the 35-50 stocks, should yield a positive return every year.

PaperCut
Senior Member

Posts: 1610
Joined: May 2004

Sat Dec 13, 08 05:18 AM

User is offline

Well, the "split-strike conversion" is a basic option strategy. I believe it's covered in all the usual beginning option trading books.

As it turns out, there are a lot of strategies in general ( not just option strategies) which will show terrific results when back tested.

There is nothing wrong with the strategy, and there is nothing wrong about back testing. But imagine you are trading and you have a manager standing over your desk. Imagine he asks you a simple question: how does this trade go wrong? If you answer him, "according to my calculations, it can't go wrong," then you are in trouble. If you say, "according to my calculations, it's most likely I will make money. However, if {set of conditions} should occur, then it will lose money."

This would be regarded as an acceptable answer.

Madoff had a big operation based around the faulty logic as delineated above. As it turns out, it wasn't quite OK...

-------------------------
"A man should be learned in several sciences, and should have a reasonable, philosophical and in some measure a mathematical head, to be a complete and excellent poet."

timbiggam
Junior Member

Posts: 1
Joined: Jul 2002

Sat Dec 13, 08 03:50 PM

User is offline

The split strike conversion strategy is essentially a synthetic long call vertical spread...difficult to make money on those in big down months. Based on that, anyone looking at the returns should have been aware that they were a little too good.

daveangel
Senior Member

Posts: 4978
Joined: Oct 2003

Sat Dec 13, 08 04:27 PM

User is online

only if you have 100 delta hedges - it can be turned into anything you want given the appropriate hedge

-------------------------
Knowledge comes, but wisdom lingers.

jfuqua
Senior Member

Posts: 1144
Joined: Jul 2002

Wed May 13, 09 04:34 PM

User is offline

Mr. Madoff's Amazing Returns: An Analysis of the Split-Strike Conversion Strategy

Carole Bernard
University of Waterloo

Phelim P. Boyle
Wilfrid Laurier University - School of Business & Economics; University of Waterloo

April 2, 2009

Abstract:
It is now known that the very impressive investment returns generated by Bernie Madoff were based on a sophisticated Ponzi scheme. Madoff claimed to use a split-strike conversion strategy. This strategy consists of a long equity position plus a long put and a short call. In this paper we examine Madoff's returns and compare his investment performance with what could have been obtained using the split-strike conversion strategy based on the historical data. We also analyze the split-strike strategy in general and derive expressions for the expected return, standard deviation, Sharpe ratio and correlation with the market of this strategy. We find that the Madoff's returns lie well outside their theoretical bounds and should have raised suspicions about Madoff's performance.

Keywords: Madoff, split-strike conversion strategy, performance measurement
Paper at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1371320

LTrain
Senior Member

Posts: 821
Joined: Jun 2004

Thu May 14, 09 06:52 PM

User is offline

The paper is interesting, but it lacks a lot of practical and real world inputs. (a common defect in academic models)

For example, if I could see enough deal flow, I could absolutely recreate Madoff's steady returns. Madoff's results obviously WERE too good to be true, but this is hindsight. How much non-model-able information could Madoff see???

Another way to ask the question:
Why did Jim Simons and other top quants at Rentec believe Madoff's story?? More than anyone, Renaissance was in the position to know what was possible and what was not.

acastaldo
Senior Member

Posts: 349
Joined: Oct 2002

Sat May 16, 09 06:08 AM

User is offline

Quote
Originally posted by: LTrain Why did Jim Simons and other top quants at Rentec believe Madoff's story?? More than anyone, Renaissance was in the position to know what was possible and what was not.

You are a bit unfair to the Ren guys. They did figure out that something was wrong:

As the WSJ reported (quoted here):

Quote
In 2004, the Stony Brook Foundation pulled $3.5 million of its roughly $8 million from Madoff's firm. Simons did urge Stony Brook to pull out all of its money, but other members of the foundation's board disagreed.

"It was at Jim's insistence that we took the position down," says Mr. Frey. It isn't clear exactly what bothered Simons, but internal SEC documents show emails between Renaissance employees around the same time Simons expressed worries about Mr. Madoff. "We at Renaissance have totally independent evidence that Madoff's executions are highly unusual," one employee wrote.

By unusual executions they presumably meant that they could not find evidence of the trades reported by M@doff's statements in standard sources (That is a very strong statement; there are always some roundabout ways to make trades that might not show up, so you can't be 100% sure that he is not trading).

View thread in raw text format

FORUMS > General Forum

Forum Navigation:

© All material, including contents and design, copyright Wilmott Electronic Media Limited - FuseTalk 4.01 © 1999-2010 FuseTalk Inc.