StockTwits for Learning Investors: Week #6

February 15th, 2009 | investing stocktwits trading

At the suggestion of @sorenmacbeth, I changed the tag line on these posts to reflect the real purpose—how an otherwise inexperienced retail investor can use StockTwits to make money. (Besides, I’ve maxed out the Google juice for the phrase “idiot retail investor”; maybe that’ll come in handy in the next retail investor bubble.)

This week’s trading didn’t turn out all that well, so I’ll cover that quickly and then dive into two other topics that I’ve been pondering: how I trade, and the whether averaging down works for me or not.

This Week In Failball

Going into this week I was way short, so I tried to get long financials prior to the stimulus hijinks. Unfortunately, that didn’t work out; my position in FAS was creamed (down about 30%), and my BAC calls are still underwater by about 30% even after averaging down. I also averaged down on both Best Buy (BBY) and Palm, two trades I expect to pay off well. I tried salvaging things on the financials by selling some premium, but was a day too late to get off some call premium on FAS and FAZ.

I’m still working on ideas for next week. My hold portfolio is all long (except for about 25% cash) so a bear rally next week doesn’t hurt too much, although there’s no denying that I was on the wrong side of a lot of financials this week.

The pain is no fun to talk about right now; I’m in denial. Let’s get to the good stuff.

How I Trade

I’ve mentioned before that the demands of my real job make it nearly impossible to day trade of follow the market on a daily basis. I tend to swing trade over the course of a few days or a few weeks, and sometimes over the course of a month or two. I focus on equity options, with a mix of equity positions. Over the past 5 months I’ve been primarily on the short side, although I have played AAPL, INTC, AMZN and DGP long trades. If I broke down the trades by dollar volume, it would probably be around 70% options (with about 15% index options), 15% equities and 15% ETFs.

On average, I’d estimate I’ve been putting in about 7-10 hours a week focused on the market and managing my trading portfolio. I typically start the night before by checking the stock futures on Bloomberg to help get a sense for what’s going on the next day. Next, I check after hours quotes on the stocks I’m following, as well as any new commentary that’s turned up on my StockTwits portfolio page. After that, I place my initial trades for the next day. Since the first 30 minutes and last 30 minutes of the trading day tend to be the time that market makers screw over the part-timers like me, I generally place my trades slightly outside the money, trying to snipe overreaction in sentiment one way or another. While this sometimes keeps me out of profitable trades, it also keeps me from some of the morning shenanigans. For the most part, this has worked in my favor.

The next morning about 90 minutes before market open, I take about 10 minutes to check the futures again and the pre-market quotes. About one in every ten or fifteen trades, I’ll adjust my bids/asks based on changes in conditions. That’s where I stand most of the day (on days I work) with occasional checks and adjustments over lunch or as time and schedule permits, occasionally checking my Twitter feed for any updates.

And the next night, it starts over again; hit my Twitter network, then the StockTwits main stream, then check in on the main blogs I follow for trading purposes:

There are about a dozen other blogs I kind of half follow (or catch up with on the weekends); but since I maybe have 30-45 minutes a night to review and get trades in, I can’t follow everything I want to (or should). Really, the crest of my Twitter network acts as the filter. Over the last 6 months I’ve come to see this as both a blessing and a curse. A blessing because I find good stuff I wouldn’t find otherwise; a curse because I tend to rely more on the network than on primary sources. And that costs.

So far, I haven’t set stops on my trades. Quite a few of my winning trades have been down 40% before turning around for gains. However, I think I can improve my return slightly with a couple of adjustments to this practice described below. The main reason I haven’t been setting stops is because of the extreme volatility; it seems like three times out of four, the trade will recover quite a bit from where I might typically set a stop. Because I can’t trade during the day, I almost always set an ask price outside the spread before the day begins and take it. I shoot for something in the range of a 10-20% gain, adjusting up and letting stocks run a bit if I have conviction. To a fault, I average down when I think my thesis is right but my timing is wrong. I talk about this a little more below.

And that’s it. I was lucky—really lucky— when I started in September in that I was making trades that represented about two to three units on AAPL and ran off a great streak that doubled my bankroll of sweep cash in short order. Believe it or not, I hit a streak of 15 winning trades right out of the gate (although some took a month or so to finally be closed out.) This came from a combination of tuning into AAPL’s volatility, going against Best Buy (BBY) per Lindzon’s blog posts and tweets, and trading a couple of other stocks I was already familiar with from my portfolio. I’ve come back down to earth since then, but I’m still doing okay, and tuning how I work.

I expect that a year from now I’ll have this workflow honed even further. Early on, I didn’t really have a notion of position size. Lacking discipline, I put too much at risk and lucked out in that it paid off. I didn’t pay much attention to beta-weighting or any kind of traditional balancing. If anything, I was successful because I took my trading account extremely short during an epic downturn primarily to hedge against my hold portfolio.

Workflow is important when you only have limited time. Since getting into digital photography several years back, and processing thousands of images, I’ve come to realize that when you’re learning, you go slowly and take many unnecessary steps and make unnecessary optimizations that just take time. With changing light conditions outdoors, you jack around with white balance on every shot, and adjust the unsharp mask on every shot. When learning, processing a series of 200 shots picking and finishing winners might have taken me 4 hours to complete. Yet after you’ve done a few thousand shots, you figure out what really matters and you can get through the same 200 shots in about 1/3rd to 1/4th of the time. I expect at some point, I’ll reach this with analysis. One key element will be always tuning the social filter.

Averaging Down

Last Sunday, fortune8 reminded me that averaging down is generally a bad idea unless you have a strong feel for an idea. Since I’ve started trading more actively over the past five months, I have enough data (113 trades) to see if that applies to me.

In the last 5 months, I’ve taken “initial positions” in securities 69 times, and averaged down on an initial position one or more times for 44 trades. Of those 43 trades, 21 turned out to be profitable, 12 turned out to lose, and 11 trades are still open. The net gain per trade on these trades is 9.69%. When I average down, 67% of the time I make money.

By way of comparison, overall I’ve made 113 trades with 66 winners, 26 losers and 21 trades still open with a net gain per trade of 7.86%. (For initial positions, I’m 45-14-10, with an net gain per trade of 7.38% with 76% wins.) Overall, that’s picking right 71.7% of the time.

Overall, I’m up a little over 6 “units” of my current average trade size. Until recently, I haven’t been that consistent on position size, in some cases going as high as 4 units, with most of the average-down trades on the order of 1/4th or 1/2 a unit.

Those numbers are eye-opening to me, even though I have no frame of reference. I think I saw that some systems assume that they pick right as little as 35% of the time, willing to take more losses to let bigger winners run (e.g., typical winner might be 2.5x the size of a single loss.) In theory, with such a system, you’d end up winning about 17% per trade over a course of trades.

The numbers I’m showing are about half of that in terms of return. And what comes to mind when I see that?

Poker. The classic amateur poker player is the one who’s the happiest guy at the table, pulling in the majority of the pots, winning almost every hand, frustrating other players as he sees each deal through to the river, grinding out low-return hands. And does that guy ever really win? No. The solid poker player lets the amateur harvest all those low return pots, content to sit back and play tight until it’s time to play loose and take the big pot.

Can I change my style and let my winners run? Is averaging down throwing money into a lost hand when you’re already beat?

At this point, I don’t think so. The data shows that for the most part, the primary difference between “initial” position trades and the “average down” trades is that the initial position trades tend to close out more quickly, and the average down trades tend to get more run. But in terms of success and return, there’s not a significant difference. In other words, for the most part, they should be thought of as independent events, even though psychologically, I tend to clump them all together, trying to “save the trade” as a single unit (e.g., referring to my “cost basis” rather than treating those as separate trades).

And maybe that’s the lesson: there’s no such thing as averaging down. Each trade is truly an independent event for a swing trader. Cost basis is interesting only for tax purposes. In trading, it’s simply a psychological construct that may inhibit appropriate decision making.

As I mentioned above, I go without stops and allow a lot of play when I have time. Based on the data above, I could probably improve my results slightly by setting a stop at something as crazy as 40%. Doing so would a) maintain my overall “initial position” winning percentage, b) cut my overall return slightly, but c) keep me “smaller” (as in too small to fail) by keeping me out of some of the long-running, large position trades. And in doing so, this might also allow me to get back to larger initial position sizes, which could make up for some of the loss in terms of actual trading units.

Focus on the Feel

The thing is: it’s not simple. Part of the reason I have so few trades is because I have such limited time. When I have a “feel” for a stock (e.g., AAPL and INTC), I can dial in and trade it with great success. Averaging down is a form of that, what I call “focus on the feel”. The areas where I’ve been killed have come from stocks outside the “feel” and in places where I shouldn’t have been. USO calls in the fall. The financials pretty much anytime. So cutting out the more profitable (but lower success rate) average down trades would be a bad thing if done on stocks for which I have a feel. But I could safely do so on stocks that are outside my comfort zone.

And this is all just a long way of partially agreeing with fortune8’s comment:

AveragingDown