This Man Has Studied Program Trading For 23 Years
By Dave Goodboy
In this weekend's Big Saturday Interview, we are pleased to have with us program trading expert, Hank Camp of HL Camp and Company. Hank has studied program trading for the past 23 years. In this interview, he will give you an in-depth understanding of what program trading is, how it impacts the markets and how it can potentially improve your trade selection process. This interview was conducted by Dave Goodboy.
Before we begin, it is important to mention that one of the keys to getting the most of out of Hank's knowledge is for you to understand the term "PREM." That term is used throughout the interview and we thought it would help you if we defined it at the outset:
The "premium" (PREM) or "spread" is the difference between the most active S&P 500 Stock Index Futures Contract minus the actual S&P 500 Stock Index (cash). That difference, which usually ranges between $5.00 to $-5.00, and slowly decays or rises as we reach the S&P 500 Futures Contract expiration, is what program trading is based on. When the PREM difference rises to a certain execution level, "buy" programs kick in.
Now, let's get started!
What Exactly Is Program Trading And How Does It Impact Traders?
Dave: Thank you for joining me today, Hank.
Hank: Thank you for inviting me.
Dave: First, can you please tell us exactly what program trading is and how it moves the markets?
Hank: That's a good question and there are really different answers and definitions depending on who you ask. Probably the best known definition is from the New York Stock Exchange. The NYSE says that any time a member firm executes a trade in 15 or more stocks simultaneously that are worth more than a million dollars, this "simultaneous trade" is to be defined as a program trade. Also that NYSE member firm that did that simultaneous trade must report that trade to the exchange. That is how the NYSE knows that program trading is now over half of the volume.
For us, however, a program trade is one that encompasses the PREM with a predetermined execution level either as a buy trigger or a sell trigger. Whether we trade only one stock, or 15 or more stocks, during the time that PREM execution level is hit, is not as important to us as the actual PREM execution level being hit in the first place. That pattern of PREM execution levels hitting on certain days and at certain times is the key to program trading for us and our clients. And it is not just for plain vanilla index arbitrage, either. The PREM is the key for almost all program trading.
Others define program trading as the purchase or sale of a large number of stocks contained in or comprising a portfolio. Sometimes you will hear clerks on the NYSE say that a sell program or a buy program is being done on the floor. But this type of portfolio change, typically done by mutual funds, is not as important for us or our clients. If the PREM is not moving, then we do not consider it a true program trade.
Dave: How did program trading first come about? What is its origin?
Hank: We knew that index arbitrage, a type of program trading, would occur even before the S&P Futures contract ever started trading in 1982. The concept of a futures contract trading above or at a premium to the actual price of the underlying cash commodity was well known at that time. We knew that the S&P futures would trade at a premium to the underlying stocks in the S&P cash index, and that sometimes that premium or PREM could get way out of line...so much so that we could sell the futures contract and buy the stocks in that index as the same time and make a profit on the difference in prices. Once the S&P Futures contract began trading, we began plotting the PREM in real time and could instantly see whenever that PREM did move a lot and got way out of line. When it did, we could instantly see the results of that buy or sell program as the Dow Jones moved quite a few points in only 5 or 6 minutes. We have continued to track the PREM tick by tick since 1982.
Dave: What are the regulations on program trading? Things changed, didn't they, after the 1987 crash?
Hank: Yes. After the crash of 1987, the NYSE established collars on program trading to slow down market declines. Those collars kick in whenever the Dow drops a certain number of points or percent. You may have seen this collar on the screen on CNBC as "Curbs In". The curbs force member firms to shut off computer entry of orders. The exact level that the curbs kick in is changed each quarter by the exchange. Currently the NYSE applies program trading curbs whenever the Dow Jones Industrial Average moves 210 points higher, or 210 points lower than the previous day's closing price. That will change again at the end of the month. We update the new collars on the web site each quarter.
Dave: It sounds has though program trading has grown in its impact on the market. How much of the volume on the NYSE is part of program trading?
Hank: During the past year, on average, a little over 55% of the total NYSE volume has been program trading. Some days that percentage jumps to over 60% and 70% of the total volume. So it is a lot. Obviously the more you know about program trading, the more you know about that volume and how it relates to your own stock and futures trades.
Dave: What about the Nasdaq? Does program trading take place there as well? If so, what would estimate the percentage of total volume?
Hank: Yes, program trading takes place on NASDAQ too. Stocks like INTC and MSFT are in the Dow and therefore are a part of program trades based around Dow stocks. Stocks in the NDX 100 are used every day too. But unlike the NYSE, NASDAQ does not produce a weekly report that details how much volume on NASDAQ is program trading. I would guess that in certain stocks like INTC and MSFT, program trading volume would be similar to the NYSE program trading volume levels.
Dave: We hear day traders talk a lot about Fair Value. Can you elaborate on this concept and how it relates to program trading?
Hank: Fair Value is a term used to describe the relationship between something like an index futures contract and that contract's underlying cash index. The major component in fair value is the interest rate charged for buying stocks in that index today and then carrying them until some date into the future. That "cost of carry" is really what Fair Value is all about. And once you know that value, then you can design computer programs to take advantage of any discrepancy between two markets such as index futures prices and the underlying stock prices in that index. That type of program trading is called index arbitrage and accounts for about 9% of all program trading.
Dave: Is there a formula to determine fair value?
Hank: Yes. The actual formula for Fair Value is very simple. It is basically the value of S&P 500 Index, or any other index you use, plus the interest you pay your broker or banker to buy all of the stocks in that index, minus all of the dividend checks you receive from the stocks in it. That is a simplification but you see the point. Fair Value is basically the net cost to carry stocks until expiration.
Dave: Why not just rely on the fair value listed on CNBC?
Hank: Some traders indeed do for the NYSE opening. CNBC shows where the Spoos are trading before the NYSE opens and how that price relates to today's Fair Value. For example, CNBC may say that the Spoos are trading up a dollar or up 1.50 above fair value. That in theory will cause the Dow Jones to be up immediately right after the opening. Of course the Spoos can drop that same 1.50 in a heartbeat on their opening, dropping down to fair value once again, and negating any early move in the Dow. Program traders know this well and rarely rely on Fair Value alone for trading.
Dave: Does program trading utilize Fair Value to determine buy and sell levels?
Hank: Yes. You must know Fair Value for today in order to determine at what price away from Fair Value you would break even on a pure index arbitrage trade. That is great in theory, but in the real world the biggest problem for us is not even figured into the Fair Value formula. And that huge problem is slippage, which can be substantial, both in the spoos and the stocks. Only by knowing your percent of slippage in the past, along with your break even levels for index arbs, can you come up with reliable PREM execution levels for buy and sell programs in the future. We do that every day and post those buy and sell levels on the web site. That information is free for your readers. Fair Value is just one part of the puzzle to figure out and currently is not the most important part. There is an academic debate that Fair Value changes as the price of the S&P 500 (or any other index) changes through out the day. But that debate is not very useful for trading in real time.
How PREM Can Be Used To Make Short-Term Predictions
Dave: Clearly PREM plays a prominent role in giving traders an edge. Can you now go deeper into how it's used and how it can help?
Hank: Now that you understand Fair Value and how it is calculated, you need to know that it is somewhat irrelevant in the real world of trading. What is most important is related to Fair Value, however, and that is the PREM. The PREM is the single most important indicator for professional traders. It is the difference between the current futures price and the underlying cash index price. The current formula is PREM = SP5Z - SPX for example for the S&P 500 PREM. The E-mini PREM is EPREM = ES5Z - SPX. There are other PREMs for the other indexes too. All PREMs must be calculated constantly tick by tick and not just as a snapshot at some predetermined time interval.
The PREM is a leading indicator and predicts well what the markets will do in the near future. If you know what the PREM is doing and going to do, then you know what the markets will do soon after. After 30 years of research in the markets, we have never found a more reliable or accurate indicator of future price moves than the EPREM and PREM.
Dave: I know different data feeds use different symbols for the PREM. What are the 3 main symbols and their corresponding service provider?
Hank: All PREMs are calculated by the individual data vendors. Unlike stock quotes, the PREM is not calculated or fed to the data vendor by any of the exchanges. Most data vendors either do not do PREM calculations at all,or do a horrible job with the calculations. Only a handful of data vendors do the PREMs correctly and accurately.
With DTN, all of the PREM's are good. The PREM is SP-PREM.Z. The E-mini PREM is EM-PREM.Z. They also have the Nasdaq PREM. The symbol is ND-PREM.Z and NQ-PREM.Z for the mini contract. They recently began calculation for the Russell 2000 PREM. The symbol is RL-PREM.Z and RM-PREM.Z for the mini contract.
eSignal does a good job too, especially with the E-mini PREM. Their ticker symbol for PREM is PREM A0 (zero not O) and the E-mini PREM is EPREM A0. Most traders that have eSignal do not know that. If they look at the PREM at all, they look at $PREM. That symbol is not correct. It is a snapshot every 6 seconds and therefore is usually wrong. The same is true for $EPREM. It is almost never correct, since the ES contract is so active. eSignal does not have the Nasdaq or Russell PREM.
Comstock does a good job. Their symbol is PREM.X and EPREM.X. They do not have the other PREMs. Quote.com buys their data from Comstock. TradeStation used to buy their data from Comstock too, but now calculates all of their own PREMs. Their symbols are $SPINX and $ESINX, $NDIQX and $NQIQX, and $RLIUX and $ERSIUX.
Dave: In watching PREM, do you watch specific program trading firms to see what they are doing? How does that information give traders an extra edge?
Hank: Number one is UBS, usually by a wide margin, followed by Lehman, Morgan Stanley, First Boston, Goldman, and Merrill, but not always in that same order. Other important firms are Bank of America, Deutsche, JP Morgan, and Nomura. There are other large program trading firms like RBC, that do a lot of index arbitrage, but normally they do not trade for their own accounts.
We have found in our research that it pays to watch these top 10 firms closely on Level III. Especially when the EPREM and PREM hits one our buy/sell triggers. Since they are trading with their own money, they usually do a superior job with their trading, and following them usually pays off very well for us and our clients.
Dave: I would imagine that it helps to watch individual stocks as well, right?
Hank: Yes. Most program traders stick with the stocks in the Dow Jones Industrial Average and the S&P100 Index for sure. A lot of them add some of the stocks in the S&P 500 index and the NDX 100 index. Pure index arbitrage firms may use all of the stocks in the index they are working, but that is not the majority of program trading being done each day. Many firms only use 60 to 75 stocks a lot of the time. So if you know exactly what those firms are doing with those 75 stocks, you have a edge. That is what we do every day with our research.
Our computers capture every large block trade in every stock in the Dow Jones Industrial Average (INDU), S&P100 Index (OEX), and NASDAQ 100 Index (NDX) every trading day. By capturing these trades, the times and sales, whenever our program trading premium execution levels are hit today, we are able to determine exactly which stocks in each index are being bought the most during buy programs; and, what stocks are being sold the most during sell programs. We then pass that information along to our clients who are trading those stocks too.
Dave: Are these stocks traded in a basket or group?
Hank: Yes. Most firms can execute either one or two or all 60 or 75 at pretty much the same time.
Dave: Is it possible to determine beforehand whether program trading will be on the buy or sell side?
Hank: Yes. That is what we do every day with our research and that is the exact information that we provide for our clients. We can usually determine not only whether we will have program buying or program selling tomorrow, but also at what times of the day those programs will be run. By knowing in advance which way and at what times program trading will hit the market, our clients have an edge that most traders do not have. In addition, we also know the exact odds of success of any trades that we will make based on those buy or sell programs hitting.
Dave: How about the strength of the program. In other words, predetermining how much of an effect the program will have on the market. Is this possible and how is it done?
Hank: Yes and no. We can determine which way and at what time programs will be run tomorrow. But knowing exactly how hard or how many programs will be run at that same time is not possible. However, once that time frame is upon us, we can judge how hard programs will hit by watching the EPREM and PREM. The bigger the move through our program buying or program selling triggers, the bigger the move of the Dow Jones and the S&P 500 Index. Those PREM execution levels or triggers are on the web site every day.
Dave: What is the methodology that you use to determine if a move is actually program-driven or is just the natural ebb and flow of the market?
Hank: We track every trade in the S&P 500 Futures, the S&P 500 Stock Index, and the PREM every second of every market day. We have since Wednesday, April 21, 1982. Every time that movement in the PREM creates a "buy" program or a "sell" program, we track the corresponding up or down move in the Dow Jones Industrial Average, the 30 stocks in that average, the S&P 100 Stock Index, and the 100 stocks in that index, to see if buy or sell programs actually moved the Dow 25 points or more, and exactly what stocks in the averages moved the most. If so, our computers log that exact time of day and keep a file of those times for us automatically. If the Dow does not move, we then ignore that PREM movement no matter what the media says. Many times, novice journalists blame program trading for a move in the Dow, when in fact, programs were not present nor did they affect the Dow or the other markets at all.
How To Apply Hank's Research To Help Your Own Trading
Dave: You mention a common manipulation that takes place in the last hour of trading on the Thursday before Friday’s option expiration and during the first 30 minutes of trade on expiration day on your website. This actually occurred last Thursday and Friday (September 15-16, 2005). What often happens during these times?
Hank: What happened this expiration is what we term a "reverse manipulation". It is more rare than the normal expiration manipulation but still hits a couple of times a year usually. Large trading firms artificially drive the markets down in the last hour of trading on the Thursday before expiration by selling a lot of S&P Futures. They do that in order to load up on OEX Calls for the next day's opening. After they finish buying all of the calls they need, then they buy a ton of OEX stocks going into the close on Thursday to drive up the OEX index price back up. Then they wait for the ripple effect to hit on opening on the Friday of expiration.
Dave: What are the steps in the process? How do these firms accomplish this?
Hank: First you start buying "in the money" puts on the OEX on the CBOE as quietly as you can around 2:10 Chicago Time. It helps if the spoos are rallying so you can slowly pick them up real cheap and on some Thursdays about then they do that. But overall this is only about 25% of the time.
Second, you start selling short big blocks of the top OEX stocks on the NYSE around 2:20. Don't use the third market. That's GE, XOM, PFE, C, WMT, AIG, BAC, JNJ, IBM, JPM, and IBM. Make sure the dollar amount of stocks that you are buying closely matches the total number of option contracts that you picked up earlier. Also, it helps to throw in a few big orders for GD, UTX, GS, and FDX too, since they are big stocks and can move fast. The crowd on the NYSE really wakes up and takes notice when you use them too. Your move will drop the OEX over 1 point. If you have Level II, you can start hitting the bids on MSFT, INTC, CSCO, and DELL at 2:41 too. That wakes up the Level III crowd. Then you wait about 6 or 7 minutes for the ripple effect to hit. Hopefully your move will panic the long day traders, making them sell out before the close, because it can trap them with margin calls in about 11 minutes if they don't. They almost always panic and sell out with market orders. At exactly the same time that the specialists are dropping their bids and trying to get out of the way. These long day traders selling out will drop the OEX over 1 point. And then a selling panic three to five minutes before the close makes the specialists extremely nervous and they drop their bids even more and that drops the OEX another point or more for a total of over three points in the last hour. And then the OEX usually goes out near or on the low of the day, with the tape running a little late.
Third, you buy "in the money" calls on the OEX from 3:09 into close. Since the OEX went out on or near the low of the day, you'll usually pick most of them up hitting the lowest offer of the day. Make sure the total number that you pick up matches the dollar amount of stocks that you are short.
Fourth, starting at 3:16 you send notices to all of the clearing firms on all of the PUTS that you hold to exercise them that night, thereby locking in your profit on all of your puts.
Fifth, on Friday morning at 8:31 you start buying all of the stocks you sold short on Thursday afternoon. This usually starts a panic and every one start buying at the market. And the day traders start buying too with no offers in sight. This usually pushes the OEX up about three points.
Sixth, you sell all of your CALLS, now even deeper in the money on the CBOE and go flat in all of your accounts, stock and options.
For a good example of this manipulation, look at a 5-minute chart of the OEX last Thursday and Friday morning, and you will see the move down on Thursday afternoon in the last hour and the corresponding move up on the opening and first 30 minutes of trading on Friday morning.
Dave: I find your research into days of the week activity to be fascinating. Let’s go over each day, starting with Monday and talk about what you have discovered.
Hank: Wow. Dave, that would take us all day. I wish trading was as simple as just what day of the week it is. But it is not that simple. For example, on Mondays the S&Ps close positive a little over 56% of the time. But on a Monday in option expiration week, they close positive 63% of the time. And if that particular Monday in option expiration week is what we have determined from our proprietary research to be a certain pattern day that we color code as green, then the S&Ps close positive over 67% of the time. And the best trade on that particular Monday is to be long with over 72% odds of success. So you see that even if you know about Mondays, every Monday is different. Whether it is in expiration week, or the week after it, or the week of the unemployment report, for example.
A lot of traders think that Fridays usually rally. But the S&Ps only close positive on Fridays a little over 50% of the time. The opposite is true for the unemployment report Fridays. They close negative a little over 50% of the time. But when you look at only unemployment report Fridays that our research has identified to be a pattern that we color code as red, then those unemployment report Fridays close positive over 60% of the time. Plus the odds of a long trade working on that particular Friday are over 75% of the time!
So you see, we could talk about this for days. Just about anything you want to know about for tomorrow, we already know. Whether it is the day of the week, date of the month, expiration week or the week after, a Fed day like today, or with Greenspan talking somewhere, or any other event, we already know about it and what that day has done in the past. So we know exactly what we will do with our trading tomorrow and we know exactly what the odds of success will be in advance. For us, that is program trading.
Dave: Your research discovered something very odd about Friday the 13th. What was this?
Hank: At first glance, you might think that Friday the 13th would be a bad day for trading. But we found the exact opposite, with that Friday closing positive most of the time. Since 1982 when the S&P contract started trading, the Dow Jones has closed positive 64% of the time, and the S&Ps have closed positive 67.5% of the time. More recently since 1990, Friday the 13th trading days have seen the markets close positive 70% of the time. We also noticed that since 1990, 68% of the daily highs in the spoos were made between 11:45 and 3:15. That information gives us a pre-programmed trade that we do on every Friday the 13th.
Dave: What’s your take on why this is? Seems kind of mystical.
Hank: Over the past 30 years we have found that whatever seems obvious, is usually obviously wrong. This is certainly true about trading on Friday the 13th. What seems like should be a negative day or a bad day for trading based on superstition, is really a very good day to be long stocks and futures into lunch for sure, and then sometimes even on into the closes.
Dave: As you know, I recently interviewed Larry Williams, creator of the Williams %R indicator. I know you have found this indicator to be fairly accurate with its signals. Please elaborate a little on your findings.
Hank: Williams %R was one of the very first indicators that we programmed into our computers back in the 1970s. We also inverted and reset the scale from William's original work to make a reading of 100 overbought, and a reading of 0 oversold. Over the years we have found it to be very accurate for trading S&Ps and stocks when set to a relatively short time frame. Whether we are using 2, 5, or 15 minute charts, or even daily or weekly charts, we usually set the time interval to four, as in four bars. Very simply, we buy whenever %R hits 100 and sell short whenever it hits 0. Most Windows-based software will not reach those extreme levels very often, if at all. So we still run %R on our own software in DOS, with satellite feeds for that very reason. Just to pick up 100's and 0's. We have found that simple formula to work very well. It is especially accurate when 2 or more multiple time frames give the exact same 100 or 0 reading.
Dave: Has your extensive research discovered anything else of interest that you can share with my readers?
Hank: Over the past 30 years we have programmed just about every technical indicator, time frame, event, or trading theory; on just about every commodity contract, and on most of the active stocks trading on the NYSE or NASDAQ. In all of those years, we have only found one or two things that can accurately predict what the markets will do in the future and give us an edge with our trading. So much so that now all of our research is based on those two things first and foremost. One is the PREM. And also the EPREM. The EPREM now leads the PREM most of the time and the PREM leads the S&Ps. And of course the S&Ps drive the markets and most stocks. If your readers study the EPREM and PREM well, they will be well rewarded for their work. We have not found any indicator more reliable and accurate. If you know what the PREM is doing now, has done in the immediate past, and have a good idea of what it is going to do in the immediate future, then you know for sure what the markets are going to do. That is a huge edge that most traders do not have. Most traders do not have a clue what the PREM is in the first place. And even if they do know a little bit about it, they usually do not have a data vendor that gives them accurate or reliable information. So with the right data feeds and software, that knowledge gives your readers a huge advantage, once they study the PREM and understand it extremely well. But that study is not easy. It takes training, time, and experience to succeed.
Dave: We are almost out of time, thanks for joining me today. Where can my readers find more information about your research?
Hank: My pleasure, Dave. Our site is www.programtrading.com