Risk Management. Presented by Lawrence G. McMillan The Option Strategist. Continuing Webinar Series

Size: px

Start display at page:

Download "Risk Management. Presented by Lawrence G. McMillan The Option Strategist. Continuing Webinar Series"

Bernice Wiggins
5 years ago
Views:

1 Risk Management Presented by Lawrence G. McMillan The Option Strategist Continuing Webinar Series

2 McMillan Analysis Corp. A Derivatives Firm Recommendations (newsletters) Money Management Option Education including Mentoring

3 Special offers on most of our newsletters, software, and educational products Visit:

4 Today s Topics Position Sizing The Concept Of Expected Return Stops and Partial Profits Using Greeks to Manage Risk Portfolio Protection

5 Determining Trade Size Fixed Percent of Assets The Kelly System Optimal f Probability of Ruin

6 What NOT To Do Double up to catch up Martingale: = +1

7 A Better Approach Increase size as you profit Fibonacci-style: 10, 15, 25, 40, 65, 105, 170, 275, 7 wins, 1 loss: +$155 vs +$60 Return to initial size ($10) when you lose

8 How Many Options Should I Buy? Fixed Percentage Risk Management Risk a fixed percent of your account on each trade (3%, e.g.) Automatically increases when you win and decreases when you lose Example: Account size = $100,000 You plan to risk 5 points on a stock trade Therefore, buy 600 shares of stock (3% risk)

9 How Many Options Should I Buy? You could figure your risk = premium, but that s unrealistic. Option costs 10 points ($1000) So buy 3, if your account size is $100,000 (3% risk)

10 How Many Options Should I Buy? Using the model to estimate risk. Oct 100 call costs 10 today ($1000). What would it be worth if XYZ fell to 95 in a week? Black-Scholes model says: In 1 week, if XYZ = 95, Oct 100 call = 7 Therefore, risk = 3 points ($300) so you can buy 10 calls, not 3!

11 Another Approach: The Kelly Criterion J. L. Kelly, Bell Labs, 1954 Original Formula: Amount to bet = (W + L) * p L Where W = amount you could win L = amount you could lose p = probability of winning If result is negative, don t bet

12 Applying Kelly: Sports Betting In sports betting L = 1.1 * W (you lay $110 to win $100) Amount to bet = (W + 1.1W) * p 1.1W = W *( 2.1p 1.1) Will be negative if p <=.52 Assume W = 1 (your risk bankroll) Amount to bet = 2.1p 1.1

13 Kelly: Sports Example Amount to bet = 2.1p 1.1 Example, assume p = 60% (i.e., you can predict sports winners with 60% accuracy*) Then amount to bet = 2.1 * =.16 which means risk 16% of your bankroll on the next bet. *: if you can do that, you don t need this seminar

14 Kelly: More Complex Situations Applying (W + L) * p L Where p = probability of a winning trade PDR System: avg win $2359; avg loss $1952 And p = 56.7% So L =.83W Amount to bet = (1.83p 0.83) Amount to risk = 21%

15 The Probability Of Ruin Define ruin: down 80%? If you risk 50% of your capital, and lose it all on each trade, then: One loss: 50% left Two losses: 25% left Three losses: 12.5% left = ruined!

16 The Probability Of Ruin Define ruin: down 80%? How many losses in a row would ruin you? (1 r) n = 0.2, where r = % of assets you risk on each trade If r = 3%, n = 52 If r = 50%, n = 2.3

17 More on the Probability of Ruin Kelly is designed to optimize your results, not protect your capital! Professional blackjack players rejected Kelly because the probability of their bankroll shrinking to any percentage was that percentage itself. 50% chance of losing 50% 20% chance of losing 80% 10% chance of losing 90% Was too much risk for them

18 Kelly for Traders Our investment is not equal to our risk Our win amount is not fixed, either We don t make sequential investments Not all of our trades have equal probabilities of profit (unless we only trade one system) Reference: Google papers by Edward Thorp

19 Optimal f f = Kelly formula According to Ralph Vince, in a complex trading system, the f that provides the maximum return on our money can only be determined by iteration. We cannot average our wins and losses from trading and obtain the true optimal f using the Kelly formula. Your Bet unit = (System s Largest loss) / f The full concept of understanding Optimal f requires reading the book.

20 The Concept of Expected Return A logical way to analyze, compare, enter and exit strategies

21 What is Expected Return? The return one could expect to make on a position over a large number of trials. Assumes the distribution of possible stock prices can be defined; also assumes implied volatilities of an unexpired option can be estimated as well. Highly dependent on volatility estimate.

22 Expected Return Example A Call Bull Spread XYZ: 52 Oct 50 call: 7 Oct 60 call: 4 Assume the stock must be at one of the following prices: Stock Price Probability < 50 45% 52 8% 54 7% 56 6% 58 4% >60 30% Total: 100%

23 Calculating The Expected Profit Now, add in the profit picture of the strategy: Stock Price Prob Profit Expected Profit < 50 45% -$300 -$ % -$100 -$ % +$100 +$ % +$300 +$ % +$500 +$ 20 >60 30% +$700 +$210 Total: 100% +$112

24 How Will This Spread Do? Expected Return = $112 / $300 = 37.3% Annualized Exp Return = 37.3% x 4 = 112% But the only point you actually would make $112 is if stock is at at expiration. Chance of that is < In any one case, you could make as much as $700 or lose as much as $300

25 What is Expected Return NOT? It is not a black box, magic formula It is not a guarantee of profits It is not a shortcut to the diligence needed to trade options profitably

26 What Does It Mean? On average, if you invest in positions with high expected return, you should approach that return eventually The Casino Analogy Erroneous Assumptions - Distribution not lognormal - Bad volatility estimate - Event Risk

27 Trading Decisions Based on Expected Return In the bull spread example, Suppose XYZ moves from 52 to 55 quickly And you have a profit of $120. That s your expected profit. Should you take it? If you do, your annualized return increases!

28 Expected Return on The Strategy Zone and Option Work Bench Our statistics on The Strategy Zone include expected return for Calendar spreads Covered Call Writes Naked Put Sales

29 Position Sizing Using Expected Return Kelly is useful because it optimizes results while increasing our trading size when winning and reducing it when losing It is a rational way of investing in a consistent manner, placing the most money on the most likely profitable trades We have expected return as a guide We can also often determine the probability of winning or losing on a trade In complex situations, Kelly effectively reduces to: Amount of account to risk = p win p loss

30 Applying Kelly In some cases, we can compute p win and p loss but that might not relate well to expected return (naked writing, for example) Better to use expected return, e p win = (1 + e) / (2 + e) P loss = 1- p win

31 EXAMPLE: CEPH Calendar $75,000 account CEPH: Buy Aug 45 call Sell May 45 call for 1.20 DB e = 17.2% (not annualized) Kelly Method: p win = (1+e)/(2+e) = 54% p win -p loss = 8% So, risk 8%, or $6,000 => 50 spreads

32 Portfolio Considerations Can t just use the Kelly or Optimal f number for each position: What if Kelly says invest 12%, but you have more than 8 positions you want to establish? Two approaches: 1) Use Kelly on remaining equity in account, or 2) Rebalance all positions (daily)

33 Recommended Reading Fortune s Formula The Untold Story of the Scientific Betting System that Beat The Casinos and Wall Street By William Poundstone Against The Gods The Remarkable Story of Risk By Peter L. Bernstein The Handbook of Portfolio Mathematics: Formulas for Optimal Allocation & Leverage by Ralph Vince

34 Risk Management Once the Trade Has Been Made Stops Partial Profits Rolling Portfolio Greeks

35 Speculative Positions Why hedge? 1) Lock in some gains 2) Reduce risk Any time you adjust or hedge a speculative position, you harm your potential profits. Most harmful: partial profits Modestly harmful: rolling As a result, hedging should not be overdone. I don t like: selling out-of-moneys against long options, hedging calls with puts and vice versa, or other methods of converting long option to another strategy.

36 Type of Stops 20-day SMA 10% EMA Chandelier Parabolic Trailing Stops Closing stops recommended Should be applied to all unhedged trades

37 Profit-taking Methods 1) Partial Profit: Sell a portion (quarter or a third) of your position Or 2) Roll the position: Sell the options you own (when they are fairly deep in-the-money) And buy a strike that is near-the-money

38 I feel I should be doing something ATK: 17 points as shown 43 points if no partial profits had been taken Profit-Taking: Allows you to book some gains

39 Partial Profits: Completely eliminates any further gains from the portion that you sold Everyone worries about a gain turning into a loss Works best if stock (or market) is not trending A huge mistake in a trending situation

40 Rolling (Up) Is moderately harmful. WHY? Your position: Long 10 XYZ Jan 50 calls with XYZ = 60 Roll: Buy 10 XYZ Jan 60 calls Sell 10 XYZ Jan 50 calls This is a bear spread, but you are bullish! But at least you retain your upside quantity

41 Speculative Summary In a trending market, the best gains come from merely holding until your trailing stop is elected. Hedging/Adjusting in trending markets is harmful It behooves you to know if the underlying is prone to trending or not. Many commodities and commodity-related stocks trend. Stocks that depend on fundamentals for propulsion generally are not trending stocks.

42 Adjusting Hedged Positions You should always have a stop whether you are Speculator Naked Option Seller Maybe even in a limited risk spread So it is not mandatory to adjust; you could just sit back and let the stops take care of things

43 Hedged Strategies Whether or not to adjust a hedged position, really depends on the strategy and your attitude. For example: Long Straddle Long XYZ July 50 $5 Later, XYZ = 55. Scalper would sell some stock, looking to buy it back on a pullback. Trender would sell the puts, and use a trailing stop on the long calls.

44 Simple Adjustment to Hedged Strategies Adjust stop as time passes to protect unrealized profits: Assume you Sold XYZ Straddle for 12 Now XYZ is near strike and straddle is selling for 6 Stop yourself out if either side goes 6 points in-the-money Similar to trailing stop.

45 Simple Adjustment to Hedged Strategies Or if relationship of profit to stock price is difficult to discern, then adjust your stop in dollars Assume you have July-April 60 calendar spread It is marking up $200 with XYZ near 60, But there is more to be made if XYZ stays near 60. Stop yourself out if your profit drops to $150, say. Similar to trailing stop.

46 Position Greeks Calculate the Deltas (Gammas, Thetas, etc.) for all options in a position and add them together. XYZ: 45 Price Delta Gamma Long 5 XYZ July 50c Long 5 XYZ July 50p Short 3 XYZ July 40p Delta: (+5x+26) + (+5x-74) + (-3x-19) = -183 (ESP) Gamma: 5x4.9 +5x4.9 3x4.1 = Long gamma means you want stock to accelerate; Short delta means you have a downside bias; Hedge: buy ~200 common => now neutral

47 To Scalp or Not? Whether or not to adjust a hedged position, really depends on the strategy. If you are a gamma scalper or delta neutral trader, then you would adjust when others might not. In a long gamma position, all adjustments are for the purpose of taking profits; In a short gamma position, all adjustments are for the purpose of preventing losses

48 Gamma Determines Attitude Position Gamma: Long Short Adjustment Whipsaws Good Bad Straight-line/gap Move Good Bad Dull market Bad Good Stop needed No Yes

49 Neutralizing Several Greeks You can always neutralize 2 or 3 of the Greeks, moving exposure to the area you desire. Example: you want to establish a call ratio spread, but you would like to neutralize your market risk and leave only the volatility risk.

50 2 Equations in 2 Unknowns Position Delta Gamma Theta Vega IV XYZ Jul 50c % XYZ Jul 55c % To Neutralize Gamma (delta comes later), while making $100 per point drop in Vega: 4.7x + 3.5y = 0 9.6x + 8.4y = -100 Rounding, x = 60; y = -80 (y / x = 1.33) Delta = 60x45 80x29 = 380; so short ~400 shares Theta = +38 (you make $38 per day in time decay)

51 Staying Neutral Later, XYZ = 52 Position Delta Gamma Theta Vega IV L 60 XYZ Jul 50c % S 80 XYZ Jul 55c % S 400 XYZ Position Greeks: To Neutralize Gamma and Delta: Buy 15 July 55c: increases Gamma by 15 x 4.5 = 67.5 Increases Delta by 41x15 = 615, so short ~800 more XYZ Leaves Theta +38; Vega 60 (to get back to 100, mult by 1.67)

52 Portfolio Greeks Can t just add all position greeks together Because different stocks have different volatilities For example, Long 200 deltas in RIMM + Long 400 deltas in VZ Does not make long 600 deltas Rather you must reduce all deltas to a common factor (EFP), such as SPX or QQQ.

53 Equivalent Market Position Assume market volatility = 15% ($SPX, say) ESP Price Volatility Adj.Factor Adj. Dlrs L 800 RIMM % 3.0 $85,200 L 2000 VZ % 1.0 $71,000 $156,200 So this portfolio is long $156,200 as market equivalent So, if SPY = 128, then to fully hedge the risk (delta) of this portfolio, you would sell $156,200 / 128 = 1220 SPY

54 What Is Portfolio Protection? The use of listed derivatives to reduce or eliminate the risk of a general market correction Most investors view protection as insurance or disaster protection, rather than a complete hedge of their entire portfolio.

55 Protecting A Stock Portfolio Buy $SPX Puts (Traditional) Or Buy $VIX calls (Modern)

56 Macro Protection Using broad-based index ($SPX or $VIX) options Most efficient in terms of cost and effort Problem: Tracking error Micro Protection Hedge each individual stock with its options No tracking error Can be extremely time-consuming

57 Macro Protection Using broad-based index options $SPX or $VIX Or other index that represents your stock portfolio Most efficient in terms of cost and effort Problem: Tracking error

58 Macro Strategies Buy Broad-Based Index Puts Buy Index Put Spreads Sell Index Calls Use Index Collars Sell Index Futures Volatility Products

59 Buy Broad-Based Index Puts Most Popular Approach; not necessarily the best approach Typically buy 5% to 15% Out of Money (OOM) The OOM portion is the deductible With care, can keep cost down to 2% - 3% of NAV Disadvantage: protection not dynamic if market rallies Advantages: fixed cost; upside profits unencumbered

60 Buy $SPX Puts For Protection

61 Buy Index Put Spreads Buy bear spreads as a hedge Drawback: cap on protection Advantage: lowers cost of protection In effect, your insurance benefits only engage between the strikes of your spread a poor approach if you actually need protection

62 Sell Index Calls Drawbacks: doesn t provide disaster insurance; and may limit upside profits Advantage: sale of a wasting asset

63 Index Collars Buy OOM puts and sell OOM calls Calls defray cost of the puts (sometimes completely) Limits both risk and reward Distance between strikes is hurt by high dividend and/or low volatility Best when using LEAPS options (can spread strikes out quite far with LEAPS): June 2007: $SPX: 1517 Feb 2011: $SPX 1340 Dec ( 09) 1450 put: Dec( 13) 1200p: 120 Dec ( 09) 1700 call: Dec( 13) 1400c: 125 Nowhere nearly as attractive (vol, div, rates)

64 Volatility Products $VIX Futures on $VIX Options on $VIX

65 History of $VXO

66 Hedging With Volatility Products Since $VIX moves opposite to $SPX, Buy VIX futures Or Buy VIX calls Advantages: stock upside unencumbered; cost limited to time value premium. Disadvantages: $VIX derivatives may not track well with $VIX itself.

67 $VIX Calls are a better hedge than $SPX puts Better than buying $SPX puts: protection is dynamic If you buy $SPX puts and market rises, your protection is virtually worthless If you buy $VIX calls and market rises, your protection is still viable

68 $VIX Hedging Is Cheap According to a well-known study, a portfolio consisting of 90% $SPX and 10% $VIX outperforms $SPX in both up and down markets. Later studies suggest 20% for hedging. So you only need to hedge 10%-20% of your portfolio s EFP, assuming it performs in line with the broad market. Buy out of the money $VIX calls, one or two months out, and keep rolling them. Strike ~= futures price + 7

69 How Many $VIX Calls To Buy? Hedge 20% of your NAV Assume EFP = $100,000 on 9/16/2008 VIX Oct 30 call = 4 You buy $20,000/(30x100) = 7 calls Stock market down 26% ($SPX 1210 to ~900) NAV: -$26,000 loss Oct $VIX Futures: +125%; $VIX +131%; Oct 30 call = 27 7 calls x $2300 profit = +$ 16,100 gain You lose -$9,300 = your deductible (9.3%)

70 Buy $VIX Calls For Protection

71 $VIX Option Protective Collar Buy VIX calls for protection And Sell VIX puts to help fund the calls Problem: Put premiums small

72 Summary Risk Management is Important: Position Sizing Use Trailing Stops For Speculative Positions For simple hedged positions, use dollar stops For complex positions, hedge with Greeks For portfolios, hedge with $VIX

73 Special offers on most of our newsletters, software, and educational products Visit:

74 CD Seminar Series Regularly $499 Show Special $199

75 Mentoring McMillan Analysis Corp. Offers One-on-one Mentoring Structured to your skill level Conducted on the internet or in person Call or visit:

76 Thank you for attending! Contact information: Phone: Fax: web site:

Learn To Trade Stock Options

Learn To Trade Stock Options Written by: Jason Ramus www.daytradingfearless.com Copyright: 2017 Table of contents: WHAT TO EXPECT FROM THIS MANUAL WHAT IS AN OPTION BASICS OF HOW AN OPTION WORKS RECOMMENDED