Tuesday mornings $SPX gap down just closed after reaching Monday's close of 2730.13. Hesitating there initially.

Hang with me on this one Matt. When you say "We filled yesterday mornings (Tues) gap from the previous day's (Mon) close." And what was Monday's close?   It was 2730.13. Check it. You have drawn the top of your purple gap box at around 2727, which is incorrect.  Fact is today's high of 2727.76 is still over 2 points short of Monday's close of the $SPX. Raise up the top of your purple gap box to Monday's close at 2730.13 and let's see if she closes tomorrow.

Matt, if you draw the gap on a daily chart for S&P500 index, you will see my point.

I am talking about the gap from Monday's close (5/14) to yesterday's (5/15) open. The tuesday morning gap originated fron Monday's close, the top of Monday's last 5-min candle, which was 2730.13. You drew the top of the gap to the OPEN of Monday's last 5 minute candle, not the close. Yesterday morning there was no trading from Monday's close of 2730.13 down to yesterday's open at 2718.59.  My data comes from the $INX  5-min chart on Tradestation.

The gap was not filled. The top of the gap is at the Monday 5/14 close, which is 2730.13. The S&P500 index high today was 2727.33, almost $3 short. Your gap is drawn to the opening price of the last 5minute candle.

Here is the note from stockcharts when you try to report a problem:

Monday, 10/17 @ 9:40am - SPY, FAZ, FAS and other ETFs are not updating for ExtraRT members this morning. We are checking into the reasons for this right now. It is not necessary to report this problem.

Trader8,

I tend to disagree. One thing I appreciate about Matt's work is that he is working hard to improve both GDX and SPY systems and is pretty quick to impliment something he believes is better. He always gives us an explanation as to why it is better and backtests. Not sure why I would want him to slow down those efforts.  I trade them both and have been happy with all the improvements, which I see as real improvements. So I don't mind when the system I'm trading is improved. These are both awesome mechanical systems.

Doug Milliken

What I have learned from this (ongoing) trade is that in the future it is up to me to make assessments of the major trends before automatically making any SPY System entry signal. I sorta did that this time and decided to take multi-entry because of percieved risk of trend shift. Normally I would take single entry.

SPY System results would obviously be excellent if starting trading March 9, 2009, or in any similar multi-year uptrend. The danger is that the last long of such a long rally could possibly be a big loser if a trend shift occurs during the trade, as it did  this time.

So in critiquing my own performance, I have to ask myself why did I even take a multi-entry SPY trade in the face of 1) Mar 9 rally completed EW count in April  2) huge H&S staring me in the face on S&P daily chart  3) Significant divergence on weekly chart.

Yet in spite of what I know about these warning signs, I didn't take enough caution.

My Point: Taking every SPY buy/sell signal based on 9:1 success rate is not always prudent. Self, don't check your brain at the door. Be cautious any time a long trend could be ready to reverse.

SPY System is mechanical. It does it's job. It looks like most trades are 90% "safe". But most is not all and I don't have to take every trade. In other words, I could profit fine from the SPY system just like it is with no "improvements" as long as I have a rational assessment of the probability of a long term trend shift.

Doug Milliken

Matt,

Your spreadsheet shows worst ever drawdown -10.0% ended up a 19.7% winner. 2nd worst 8.7% drawdown ended up 7.1% winner. 3rd worst was 7.8% drawdown but final profit was cut off from your screen capture. What was it?

Thnnks,

Doug Milliken

Marketguy, that's why I opted for the multi-entry this time around even though I am a strong single entry kinda guy. Doug

I see some signs that lend "comfort" to a buy right now. Namely we have 200 DMA support immediately below, and in fact have bounced off it so far this morning. Also, the VIX closed above it's upper bollinger band yesterday then gapped up this morning ! Of course it could go up some more, dragging the upper band up with it. But when this happens, there is usually a rally starting "soon". When VIX closes back inside BB and an indicator (such as Stoch(50) on the VIX) reverses from overbought and closes below 80we should have at least a short rally. Have a look at what SPY did after these historic occurences. Lends comfort to this trade. In spite of what is going on with the elephanys and the asses, I can see some cause for short term bullishness. Matt's latest trade could be another surgical precision entry yet.

In any case, it is what it is. I'm OK with that, every time.

Doug

Welcome to GDX trading, Roger

Marketguy,

If you do this analysis on any other chart service other than Stockcharts.com your S/R lines you have drawn are quite a bit different. This is because Stockcharts.com is one of the very few who manipulate past data down to adjust for dividends and cap gain distributions every time they are declared. This applies to all ETFs and stocks on Stockcharts. This is why I only use Stockcharts for Indices when it comes to long term charts. Never for ETFs or stocks. Their charts historical data is a moving target !!! Just my opinion that Stockcharts is useless for longterm ETF analysis unless you are long term investing and therefore care about total return.

For instance, the 2007 high for IWM shown on your stockchart is 81.50. Yahoo is 84.90 on the same date. That is 4% stockcharts has reduced the 2007 high so far. Every time IWM declares distributions, past data ticks down some more. It is a moving target! I'd prefer to check S/R on a chart that never moves (except for splits).

http://support.stockcharts.com/entries/20634-historical-price-data-is-adjusted-for-splits-dividends-and-distributions

Doug

Gram,

Say you sold something during trading hours today. Only those funds take 3 days to "settle". You are free to do what you want with any other funds that are already settled.

For example, I sold my SSO per the SPY sell signal at the open this morning, Wednesday. Let's call Wednesday trading day zero. Those funds are not settled until my broker does their settlements during the after hours between trading day 2 (Friday) and trading day 3 (Monday). That means technically I should wait until Monday's open, the first time my funds will be settled during regular trading hours,  to consider using these funds for a purchase. However, I will use them as early as the close of trading day 2, Friday. When I hit the "Buy" button, I get a warning that I am purchasing with unsettled funds (assuming I don't have other settled funds available) and do I want to proceed anyway. I go ahead and over ride the warning which just means I can't change my mind and sell what I just bought in the after hours between trading day 2 and 3, which I never do anyway.

So you can jump the gun by one day if you purchase with unsettled funds at the close of trading day 2. That's the way it works on my platform anyway.

If you want you can buy with unsettled funds on day 1, but if you do you can't sell until the open of trading day 3. I personally would never do that.

Others have posted on this forum that Sink or Swim has some type of margin IRA (?) that makes sold funds available immediately.

Doug

I find that while SPY does not coorelate real close with 1/10 of SPX, the % change from yesterday's close is usually much closer, usually within a few hundredths of a %. So I made a little spreadsheet calculator where

the inputs are:

1. yesterday's SPX close

2. yesterday's SPY close and

3. Matt' SPY signal target.

The outputs are:

1. the % change from yesterday's SPY close to Matt's SPY target and

2. the SPX that gives the equivalent % change

The only resaon I use an SPX equivalent approximate target is that sometimes I am away from my computer for an extended period (I am retired and usually home all day - up until market close), reading a book or doing something downstairs where there is no computer and I just put the TV on CNBC (muted of course ) and every once in awhile I glance at the screen which shows me where SPX is and I have an idea if we are getting close to a target or if I should stop what I am doing and get back to my computer charts.

I agree wholeheartedly. With 9:1 win ratio in backtesting, I am thinking of devoting a significant portion of my retirement accounts to SPY System and using a much smaller portion for my own swing trading, which has not approached SPY System backtest results. As long as average annual returns in SPY System remain better than my own trading, I will hold or increase the amount devoted to SPY System. If I never approach SPY results, that is fine. I just have to be continuously trying to improve my own proficiency at some level.

It is the 9:1 consistentcy that gives me the confidence to follow the signals with minimal emotion. That reduction in emotion frees me up to focus on improving my own trading competancy on a smaller trading account. So it is a sort of a relief that frees my mind up to focus on other constructive educational endeavors while I am making serious money.

I took a few courses at one of those trading schools and they preached that you only need a 50:50 win ratio to be a successful trader, as long as your losses are small and you let your winners run. At no time did any instructor imply that a 9:1 win ratio was a realistic goal.

So thanks Matt. Your points about the possibility of missing a larger move are clearly understood and accepted as the nature of the logic of the system. I would never be disatisfied with anything that has a 9:1 win ratio.

Bought into weakness, sold into strength. Very nice trade Matt.

Randy,

I'm not sure why anyone cares what you are doing outside the SPY system, which is the subject of this blog.

I never traded after hours before. I got two of my IRAs in SSO before the close but missed the deadline on the two smaller ones. So I called my broker (Scottrade) and he told me how to do it online. So I was able to buy SSO about 10 minutes after the close - at 52.67  - 4 cents below the closing price (52.71).

Now I know how to do it and next time we have a closing cliffhanger where I am not sure whether to pull the trigger at 12:59 PDT, I won't sweat it. If it remains unclear, just let the market close, observe the closing price and. if appropriate, buy it a minute later after hours.

My commission was same as during hours trade - $7.

Doug Milliken

Matt had no need to comment further. His 4:10 instructions are crystal clear and complete. If you are long anything, sell on tomorrows open. If the crystal clear instructions for multi-pointers was to buy an additional 20% on today's close and sell all on tomorrow's open, why does Matt have to comment that the trade is overnight? Isn't that patently obvious? He showed us awhile ago that more market movement happens after hours than during hours.