Thursday, May 31, 2012

How Much is Too Much? (Position Sizing)

30 people in a room, everyone gets the same exact trades in the same exact sequence. How many different outcomes will result?
   Years ago I did a lot of work with Van Tharp, one of the original Market Wizards from Jack Schwagers books. Van had game where he had 100 marbles in a bag, each representing a trade. Each marble (trade) was denoted with a return that was a multiple of the risk (know as “R”) taken on that trade, +1R, -2R, + 5R, etc.  60% of the marbles (trades) were winners. The unknown was the amount of the gain or loss from each trade. Participants were each allocated $100,000 and instructed to bet a percentage of their equity on each marble pull. The range of results was astounding. Some made huge multiples of their equity some made a solid return, some had small losses and some blew out. In all there were 30 different results
   How could this be? How could some lose money when 60% of the trades were winners? How could someone blow out? And how could some double or triple their equity while others had small gains or losses? They were all taking the same trades, akin to everyone buying a stock at exactly the same price and time. Well in Vans bag of marbles was a 10R loser, the equivalent of being short a stock that gets a takeover bid, but there also were 2 10R winners and 1 20R winner, like being on the right side of a takeover, along with trades in between. The market is the same; you don’t know what is going to happen on any individual trade. This may be the big one, good or bad! But how could the results vary so widely? Well the answer to the question is position sizing. That alone is the determining factor in how the results of the 30 traders differed. Did you buy 300 shares or 3000 or short 1000. Does your position size increase dramatically after a few losers because “you’re due”?  Well here’s some numbers for you. Even in a 60% system, after 1000 trades it is likely that a run of 7 or 8 losses will have occurred. In fact, 5 losses in a row are very likely during just 100 trades on a 60% winning system. Increasing your risk after a few losers is called the “Gamblers Fallacy” and it has caused the ruin of many. If you flip a coin and it comes up heads 10 times in a row, what are the odds for the next flip? 50%, that’s it. The coin doesn’t know that 10 heads have come up in a row. It just knows that it has an even chance of being heads or tails on that flip. In Vans game that day 5 people blew out and lost everything, and 60% of the trades were winners. Clearly they were risking TOO MUCH! More than likely after a run of losers in an attempt to “get their money back” . This game was played after a day of tutelage had been given on the risks of improper position sizing. And people still had the temptation to increase size beyond reason. This all happened without the pressure of it being real money. It was just a game. The pressure to increase size after 4 losses in a row is much greater. The temptation to rationalize more risk at that point is intense. The thought that this trade may be the 10R loser seems to get lost, but it just might be. You don’t know. You think you know, but you don’t.
So what is the solution? The solution is a money management plan that allows you to take advantage of the nuances of your trading system. Don’t have a system, a rigid rules based plan? Just wing it on gut or feel? Do you  sit in on a trading room and follow along with the trades of the whoever runs the room?  That’s ok; you can still employ risk management and trade safer.  The initial $ risk on every trade should be the same. And it should be a small percentage of capital as you just don’t know whether the next trade is the big one, the 10R or more loser. The most important part of winning is staying in the game.
I’ll do a future post about different types of money management schemes, but for now a simple, relatively safe methodology is the Percent Risk Model. It risks 1%-2% of your equity on any trade. Many may say “That’s way to small, I need to make more money than I will with a trade that small”. Maybe, but the market doesn’t really care what you “need” to make. It is going to do what it is going to do, just like it did before you started trading, and like it will do after you are done. And if you risk a lot more than this you will be done at some point.  Ed Seykota, another Market Wizard, said that risking more than 3% of equity is gun slinging, and he was right, with the gun pointing directly at your own head.
If you risk more than a small amount of capital, at some point
1)      You ignore your stop because you don’t want to lose that much money on that trade.
2)      When you ignore your stop you add to your losing trade in an attempt to get back to breakeven.
3)      When you add to your loser you add again in an attempt to get back to a smaller loss.
4)      When you add again you add one more time because it can’t go down forever and your wife is going to kill you.
5)      When your wife is going to kill you you puke it all out at what you later realize it the low of the move.
6)      When you puke it out at the low you put Ed Seykota’s gun to your head.
7)      Don’t put Ed Seykota’s gun to your head. Get Direct Cable.
8)      Ignore the last part of #7, I couldn't help it, I love those commercials.
Here’s how not to get caught in that vortex: Assuming an account at $100,000 a trader willing to risk 1% of equity and looking to place a trade in AAPL would 1st determine his stop. So if he decided to buy a breakout over 574 and use the low of the morning at 572 as his stop, he would be risking $200 per 100 shares. 1% of equity returns $1000 which equates to an allowable position size for this trade of 500 shares. Maximum loss will be $1000 plus slippage & commissions. Stops should be raised as price rises.
   Do this for a while and chart your results. Attach an R value to each trade. A loss of exactly $1000 would be –1R. Include your actual slippage in the computations as it can be significant. Assess yourself, try to keep all of your losses at 1R or lower. Allow your profitable trades to run.  And don’t put Ed Seykota’s gun to your head!

Important Levels for SPY & AAPL

Cloud Top138.32Cloud Top607.21
Cloud Bottom138.03R3602.10
62% Fib137.35Kijun-Sen598.53
50 Day MA137.2762% Fib597.08
50% Fib136.86R2588.67
R3136.7150 Day MA588.38
100 Day MA135.86R1583.92
Kijun-Sen135.6150% Fib582.70
R2134.51Yest Hi579.99
38% Fib134.38ConfluenceYest Close579.17
20 Day MA134.25Pivot575.24
Yest Hi133.69S1570.49
R1133.1438% Fib568.33
Pivot132.31Yest Low566.56
Yest Close131.7620 Day MA563.15
Yest Low131.49Cloud Bottom556.85
200 Day MA128.53100 Day MA539.51
S4125.71200 Day MA465.02

Wednesday, May 30, 2012

Are you the “King of the Penny” ?

How many times have you done your market analysis, entered your trade, set your stops and your targets, and then watched as your stock moves exactly as planned.  How good does that feel? You bought a support level as it was being retested, the Dollar  was weakening and market internals showed underlying strength. The selling dried up and volume came in as the stock bounced hard off that support level and closed strong at 36.00. You held it overnight looking for follow through. The next morning, the stock is looking  up 2 points just under strong resistance at 38.10. At 9:25 EST, satisfied with your gains up there  you offer out your stock at 38.00. It opens at 37.88 and trades up to 37.95, fluctuates around 37.90 and then trades up again , 37.94, .96, .98, 38.00. YES! 38.00, 38.00, 38.00. Ok, where’s my audio signaling my fill? 37.95 Damn…I didn’t get it off, 37.98, .99, 38.00 again. Ok, Ok……. WTF? And then 37.90, 37.85, 37.80, 37.70, Shit Consumer Confidence just hit, didn’t realize it was 10:00AM already. Sold at 37.50, ahhh, missed it. Damn, 37.00, 36.50, 36.20. F$Ck, Sh!t, P&ss, is heard as a cup of coffee gets knocked across the room! The whole f#cking  profit gone. Just for a damn penny. Again!! That’s me, King of the f&cking penny.
    Has this ever happen to you? Well I have news for you. It ain’t just you. You’re not The King of The Penny!  Many have laid claim to that title. It’s happened to every single one of us. And it will happen to all of us again too, and again, and again. Unless we decide to use market orders 100% of the time, but then there’s the whole rebate thing and…well that’s another post. But how would you like to reduce the # of times this happens? That’s reduce…not eliminate. Eliminate would be impossible. But if you could reduce it by a significant percentage that would save a lot of stress, not to mention $ ….& coffee.
   A little history, back in the days before electronics hit Wall street, a specialist on the exchange floors would keep a book of all of the public’s open orders for each of his stocks. It was simply a piece of cardboard about 3x5, the size of the order tickets that brokers would leave with the specialist clerks when they weren’t executable immediately, orders that were ”Away from the market”. The clerks would file them face up on either side of the cardboard, sells on one side, buys on the other, continuing down and up in price order on each side. Orders with the same price were placed by time of entry at that price with the oldest showing first. On the top shown order at a particular price the clerk would write in the upper right corner of the top ticket, the total amount that was bid for or offered at that price. This allowed the specialist, when asked “ How’s Crystal ” (COR, Crystal Oil), to know exactly how much stock was bid for and offered at each particular price. He’d reach into his rack, pull out COR’s cardboard, and quickly give a quote, “ 37 5/8 – 7/8 4000 by 2700” he could answer. (Having a wrong quantity at a particular price level resulted in a public flogging for the offending clerk).  “I got another 6,000 to buy Ray, said Buddy from Asiel. So Ray replies, you bought 2700 at 37 7/8, 1000 Merrill & 1700 Hutton, & 3300 at 38.00, Shearson”. How’s it now” asked Buddy?  37 ¾ – 8.00, 2000 by 14,700 replied Ray the specialist.
   So what can we earn here? Even after buying 3300 at 38.00 there was still 14,700 left at 38.00 also known as” the figure” as whole #’s were called. There was almost always lot of stock offered or bid at “the figure”. Human beings like round whole #’s. There were always many more orders, for much more stock at whole #’s than any other price increment, and more at .50 than there were at .25 or .75, and more at point .25 & .75 than there were .125, .375, .625 & .875. And you can extrapolate that to out stocks and options that traded in sixteenths too. Smart traders on the floor knew that if they weren’t standing in the crowd, and entitled to a “match” (in those days the book would get the majority of the trade but brokers standing in the crowd could be entitled to some of the sale), they may have been better off offering their stock one trading increment below the whole #, or bidding one increment above in the case of a buy order.  A bit painful as increments were wide back then, 1/8th or a 1/16th, but better than missing your trade. This was known as “Teenieing the bid or offer” Today, with penny wide increments, it makes absolutely no sense to place an order at the figure, the whole #, 38.00 as the cost to teenie that large grouping of orders is only .01, a small price to pay to Avoid the logjam just overhead. Human nature is still the same, people still like whole #’s, and to a lesser extent halves, (.50). It is not as prevalent as it was in the past as algo’s aren’t governed by human biases and they account for so much of our volume, but it still is very prevalent and the cost to a trader is miniscule compared to the old days, .01 verses .0625 or .125. Take a look at the chart below. It is Facebooks opening day. The stock opens at 42.05, prints at 42 in the 1st minute (if you were bidding 42.01 you got filled, bidding 42.00, maybe, maybe not). In the same minute it rallies up to 45 (if you were offering 44.99 you got filled, offering 45.00, maybe, maybe not). It sells straight down 7 points over the next  2 ½ hours to 38.00 (if you were bidding 38.01 you got filled, bidding 38.00, maybe, maybe not), rallies to 40.50, and drops to 39.73, rallies to 42.00 and drops to 40.25, rallies to 42.00 and drops to 40.00, rallies to 40.99 and then drops to 38.00. Eight instances in FB’s 1st day alone where placing your bids and offers .01 above or below a round # would’ve resulted in a definite fill verses the very real possibility of not getting one at all. In all of these cases the stock moved enough away from price to have caused (if you did not get filled on your original offer)one to trade at  a less advantageous price  or no fill at all if one were not nimble.

Now this obviously does not mean I am advocating trading at these prices as a system, lol. This only pertains to situations where a trader is intent on trading at a price near a round # anyway. Also keep in mind, that many times the stock will trade through the figure and keep going, and you won't realize any benbefit in thse instances, in fact you will cost yourself a penny. But just imaging that you were offering FB at 45.00 and it didn't sell, and you didn't hit a bid immediately. How many pennies would you have cost yourself? So, be the TEENIEer, not the TEENIEed! And save a lot of coffee!

Important Levels for SPY & AAPL

R4141.21 Cloud Top607.21 
R3138.45 62% Fib597.08 
Cloud Top138.27 R4596.60 
Cloud Bottom137.93 50 Day MA588.79 
50 Day MA137.45 R3587.91 
62% Fib137.35 50% Fib582.70 
50% Fib136.86 R2579.22 
100 Day MA135.83 R1575.74 
R2135.69 Yest Hi574.00 
Kijun-Sen135.61 Kijun-Sen573.33 
20 Day MA134.70 Yest Close572.27 
38% Fib134.38 38% Fib568.33 
Yest Hi133.93 S1567.05 
Yest Close133.70 Yest Low565.31 
Pivot132.93 20 Day MA563.30 
Tenken-Sen132.05 S2561.84 
S1131.94 Cloud Bottom554.29 
Yest Low131.17 S3553.15 
S2130.17 Tenken-Sen552.58 
200 Day MA128.47 S4544.46 
S3127.41 100 Day MA537.89 
S4124.65 200 Day MA464.01 

Tuesday, May 29, 2012

Important Levels for SPY & AAPL

Cloud Bottom137.93 Cloud Top607.85 
50 Day MA137.58 62% Fib597.08 
62% Fib137.35 50 Day MA589.05 
50% Fib136.86 R4584.34 
100 Day MA135.77 50% Fib582.70 
Kijun-Sen135.61 R3576.96 
R4135.45 Kijun-Sen573.33 
20 Day MA135.01 R2569.58 
R3134.38 38% Fib568.33 
38% Fib134.38 R1565.94 
R2133.31 Yest Hi565.85 
Yest Hi132.85 20 Day MA562.88 
R1132.71 Yest Close562.29Confluence
Pivot132.24 Pivot562.20 
Tenken-Sen132.18 S1558.56 
Yest Close132.10 Yest Low558.47 
Yest Low131.78 S2554.82 
S1131.64 Tenken-Sen552.58 
S2131.17 Cloud Bottom551.10 
S3130.10 S3547.44 
S4129.03 S4540.06 
200 Day MA128.38 100 Day MA536.31 
Cloud Top38.27 200 Day MA463.02