Learn How to Trade with Falcon

The McClellan Summation Index is a breadth indicator derived from taking a cumulative summation of the McClellan Oscillator, which is a comparison of advancing issues vs. declining issues. In a healthy bull market the advancers vs. decliners should also be expanding in favor of the advancers, which will cause the McClellan Summation Index to rise.

If the market advances but the McClellan Summation Index declines then the advance is based on a weakening base of support. Also, if the market makes a higher high but the McClellan Summation Index makes a lower high then the second high was made with much weaker buying strength and is a sign of a potential pullback. This sort of divergence is common before a market pullback. The McClellan Summation Index is not just another twist on the price action, like so many other indicators, it is measuring market participation by the best stocks in the market (NYSE listed) which gives you excellent confirmation and advance warning of a change indirection.

It also works in reverse for market declines. See the examples below:

The top chart ends on 8/15/2013 and was made with MultiCharts. The bottom chart depicts the market bottom on 10/4/2011 and was made with TradeStation. These charts use daily data. The reaction time of weekly data is too slow and intra-day data does not provide reliable information.

The top chart ends on 8/15/2013 and was made with MultiCharts. The bottom chart depicts the market bottom on 10/4/2011 and was made with TradeStation. These charts use daily data. The reaction time of weekly data is too slow and intra-day data does not provide reliable information.

The McClellan Summation Index is not part of every stock chart package but it is available on StockCharts.com ($NYSI) and you can contact us at sales@tradingcomputers.com to request our custom code for MultiCharts and TradeStation (It is not a standard indicator on either one of those either).

Using only the McClellan Summation Index, I was able to have enough confidence at the recent double top high to buy a lot of put options on SPY at almost exactly the top.

You can follow some of my market observations at MarketPirates.net free of charge (Link opens new tab).

Successful swing trading requires you to consider and balance many elements to decide which trade to put on and when to get out. With 70% of stocks following the direction of the major indexes each day you also need to consider the direction of the overall market when trading individual stocks.

In this process you will probably look and listen to the opinion of experts. When you consider expert opinion you should listen for ideas based on good logic that resonates strongly with you because experts often disagree with each other. When considering the opinion of experts or other market news you need to avoid confirmation bias, which is a tendency of people to favor information that confirms their beliefs. You need to also be able to spot someone who is putting out poorly constructed opinions.

Confirmation bias often sees ambiguous evidence as supporting an existing position. You may even form an early opinion based on some preliminary information that resonates with your own market theories and then ignore later evidence that conflicts with your views (Known as Belief Perseverance and Irrational Primacy Effect –greater reliance on information encountered early in a series). Confirmation bias often occurs because people weigh the costs of being wrong, rather than investigating in a neutral, scientific way. These are human tendencies that are natural but must be avoided in trading to prevent unnecessary losses.

Confirmation bias can even maintain or strengthen beliefs in the face of contrary evidence as people seek out additional sources that agree with their views even as evidence mounts to the contrary. In trading I see this when traders fail to follow proper exit criteria and give various reasons why they are still in a bad trade. As a relatively disinterested observer looking at their trade and the evidence the trader is presenting, their judgment seems obviously flawed. These situations usually end badly.

An experiment gave 45 subjects a complex rule-discovery task involving moving objects simulated by a computer. Objects on the computer screen followed specific laws, which the subjects had to figure out and they could test their hypotheses to see if it worked. Despite making many attempts over a ten-hour session, none of the subjects solved the problem or even came close to solving the problem. They typically sought to confirm rather than falsify their hypotheses, and had a strong tendency to disregard disconfirmation and consider alternatives. Even after seeing evidence that objectively refuted their working hypothesis, they were reluctant to try something else. Some subjects returned time and time again to the same wrong hypothesis, seemingly unable or unwilling to give it up. Ideas that were proven to be wrong were only abandoned 30% of the time. The intent of the study was to assess the Instructional Effect: Half of the subjects were instructed in proper hypothesis-testing which would have made them successful, but these instructions were almost entirely ignored and had no effect.

In another experiment the subjects displayed a surprising and strong tendency to seek diagnostically worthless information to confirm their beliefs. Those with a strong math background tried to use math to explain phenomenon that did not lend itself to mathematical analysis. They then strengthened their conclusion based on that irrelevant information. I have seen swing traders use ideas irrelevant to swing trading to justify a bad trade.

Traders with a strong science or engineering background tend to over apply math in trading decisions. When you over apply math you run the risk of falling into the False Positive Paradox, which is a statistical result where false positive tests are more probable than true positive tests. Even tests that have a very low chance of giving a false positive in an individual case will give more false than true positives overall (more losing trade than winning trades). This effect is more likely when the overall positive sample size is small (The number of winning trades in the sample is small). Also, combining ideas using math can produce Simpson's Paradox in which a trend that appears in different groups of data disappears or even reverses when these groups are combined.

My experience is that a trading idea proven by math is only valid if it has a sample size of at least 1,000 trades. I see math people constantly putting out trading ideas based on far smaller sample sizes. Several years back an article was published noting that eight out of eight times when the weather was unusually warm in Africa that the Coffee crop in South American was unusually good. This was theorized to be caused by warmer African ocean currents or perhaps the Jet Stream bringing good coffee weather to South America. However, after the eighth correlation it was never correlated again.

A key element in avoiding confirmation bias is to deliberately try to see if you can effectively refute your opinion. Disconfirmation is very important and has led to the development of Red Teams* in business and government to check an idea before it is implemented. IBM, SAIC and CIA have long used Red Teams to improve decision making. The 2003 Defense Science Review Board recommended Red Teams in security organizations to help prevent the shortcomings that led up to the 9/11 attacks. You need to periodically turn yourself into a Red Team to check your ideas if you want to make good decisions.

The logic of entering a trade often involves little more than a hunch but I would encourage you to have at least a confirming chart pattern and one other good reason before putting on a trade. Before you enter the trade and become emotionally involved with it you need to set clear profit and exit criteria. Once you are in the trade then you follow your rules without exception. Rules can only be changed for future trades, not the current trades and then only after careful consideration and Red Teaming the logic. When deciding on the best profit and exit criterion you should use support and resistance levels (more on this later).

*Red Team vs. Blue Team is a military simulation convention where the Blue Team represents the friendly forces and the Red Team represents the enemy forces.

If you max out your risk on every trade then you will lose your account some day, it is inevitable. The amount you should risk on every trade can be mathematically determined using game theory to give you the highest possible return.

For example: If your winning percentage is 60% and your risk/reward is 1.5:1 then your maximum per trade risk should be 20%. At 20% you will get the maximum return on your money once losses are factored in. If you risk more than that, the losses will set you back too far and you will waste too much money regaining your losses. If you risk less, your return will be less.

The Kelly strategy will make your returns more volatile when compared to a constant dollar approach. It was invented by Dr. J. L. Kelly of Bell Laboratories in 1956: … the maximum rate that earnings may be made over the long term of a series of gambles when the probabilities of winning and losing are well known.

Most traders would find the volatility of a maximum Kelly percentage too emotionally stressful so a compromise approach will make you sleep better at night: fractional Kelly. Most people who use fractional Kelly agree that 20-25% of Kelly is best. In the above example a stop loss of 4% of your entire account balance would be 20% of Kelly. Keep in mind that Kelly is only valid when “...the probabilities of winning and losing are well known” so play it conservative until you have good data. You should also note that probability of ruin is not zero if you go with 100% Kelly.

If you are new to trading or for some other reason you don’t have an accurate trading history then you should keep your trades small until you have established your track record. If your track record is poor then paper trade until it gets better.

Most of the larger hedge funds use fractional Kelly for managing risk. There is even an app for it called “Kelly Bet Basic” and “Kelly Bet Adv”($5) for Android that I highly recommend because it employs advanced methods to determine the best fractional Kelly amount to almost completely avoid you losing your account to a bad run of trades. A simpler alternative is available for free at albionresearch.com (opens new tab).

If you are holding your trades overnight and the investment you are trading does not trade overnight then the risk is much higher because the investment might gap down when it opens if there is bad news. I know a guy who put his wedding money in one trade and the next morning the stock was cut in half by bad news about accounting irregularities and the CFO resigning. His fiancé was pretty unhappy.

As a general rule, your stop loss should be no more than the fractional Kelly amount and the total trade size is not more than the full Kelly amount. Let’s look at an example: If your winning percentage is 60% and your risk/reward is 1.5:1 then the formula would be .2 = 1.5*(0.6-0.4)/1.5 or 20%. I like to use a fractional value of 20%-25%. So if you have a $5,000 account with 60% winners and a 1.5 risk/reward ratio then your Kelly value would be $1,000. Multiply $1,000 by the fractional value of 20-25% then you get $200-250. Therefore $200-250 is your maximum stop loss for the entire trade. Do not enter a trade unless you can use a $200-250 stop loss on the entire trade, you have a profit target of $300-375, and your winning record is about 60%.

Floor Trader Pivots are a set of intraday support and resistance levels that represent points of resistance and areas where price reversals are more likely to occur. It was originally developed by floor traders (“local”) in the trading pits of the equity futures exchanges in Chicago.

The pivot levels are calculated using the prior day’s High, Low, Open, and Close. The center level is called the “Daily Pivot” and is calculated as follows: P = (H + L + C) / 3. The Daily Pivot is the weakest of the levels and generally represents the dividing line between a bullish day and a bearish day. The rest of them can be referred to as "levels" or "pivot levels".

Below the Daily Pivot level are the support levels. The main support levels are S1, S2 and S3 (see the chart above). Likewise there are three major resistance levels above the Daily Pivot known as R1, R2 and R3:

• R3 = R1 + (H - L)
• R2 = P + (H - L)
• R1 = 2 × P – L
• S1 = 2 × P - H
• S2 = P - (H - L)
• S3 = S1 - (H - L)

The pivot levels attract price to them and when price gets there we get a reaction. When the price pauses at a pivot level it will generally pivot away by reversing back where it came from or accelerate as it passes through it. If it passes through it then it will likely go to the next pivot level. If it gets close to a level but is reluctant to touch it then it is very likely to reverse back.

There are some observations that we can make based on these pivot levels: Do not put on a short trade when the price is above R2. Do not put on a long trade when the price is below S2. The first time the market touches S1 is bullish if the price velocity is slowing and it is in the first 30 minutes that the market is open. After 30 minutes you can still trade the first touch for up to an hour after the market open if it touches S1 gently.

Only putting figure tags around smaller images for centering, larger images fit the full width

If you have an unusually wide range day then the next day will have much wider spacing on the pivot levels, which will render them much less effective. Wide range days are frequently followed by days that are much narrower in range so you may only get an interaction with just one pivot level -and that may not amount to any useful trading opportunities.

Conversely if you have an unusually narrow range day then the next day will have much narrower spacing on the pivot levels. However, narrow days are often followed by more narrow days and the narrow levels that are produced are still tradable, but they are less strong.

You can follow some of my market observations at MarketPirates.net, free of charge. (link opens new tab)