Thursday, March 22, 2007

Thursday, March 22nd Morning With The Doc

9:48 AM - OK, my hope is that the research at the start of this AM, combined with a read of preopening markets and the moves in sentiment in the AM helped you see that this AM has been a consolidation of yesterday's gains. We've seen net hitting of bids in ES, a negative distribution in Adjusted TICK, and a break below the overnight lows in the major indices. Still, with all that, declining stocks only lead advancers by about 250 issues. That, so far, is consistent with a pullback in a bull move, not an outright bear leg. In my own trading, because I view the main edge to be a move over the next few days (follow through of upside momentum) my main focus is waiting for selling to dry up to find a good place to go long. That will require more than the TICK bounces we've seen to this point. Have a great rest of your day!

9:40 AM - After the failure of the TICK to sustain its breakout and upward distribution, we're now seeing a resumption of selling. Note how staying with the distribution of TICK helps keep you on the right side of the market. TICK broke its early AM lows prior to the downside move in the futures. Wrap up in a few.

9:32 AM - Until we see a TICK distribution maintain an upward shift--note how we're well below that average TICK of 250--it's premature to jump in aggressively on the buy side.

9:27 AM - Need to see selling dry up above the early AM lows to hit the buy side hard. Otherwise, it will be back to selling TICK bounces.

9:15 AM - I continue to view this as range bound action in ES, with today's highs and lows framing the range. Note how buying that first pullback after a TICK breakout provides a nice continuation trade.

9:11 AM - Note upside breakout in NYSE TICK; looking at those overnight lows as important support now and leaning toward buying pullbacks in TICK that keep us above the AM lows.

9:08 AM CT - Watching to see if the market holds here, as ES has come down toward its overnight lows.

8:59 AM CT - We've been getting consistently more volume at the bid than the offer in ES, as shown in Market Delta. Under those conditions, you can wait for the lifting of offers and then fade those bounces if you're a very short-term trader. Note how ER2 and NQ have been leading ES...important tell.

8:49 AM CT - Remember LEI at 9:00 AM CT.

8:46 AM CT - Note that we've broken preopening lows in ER2 and NQ. Watching for spillover into ES.

8:44 AM CT - TICK and advancers/declines remain positive, so even though we had an early pullback, I'm not selling into those. Remember that TICK reflects sentiment and is highly influenced by institutional trade. As long as that TICK distribution is positive--and especially as long as it's above its 20-day average (which is about 250), I'm not likely to sell the market. So far, especially in NQ and ER2, the response to buying sentiment has been poor--a classic sign of inefficiency, which is not characteristic of markets that are uptrending.

8:35 AM CT - Advancers lead declines and TICK has stayed above zero in early trade. What we're seeing is inability of buyers to lift market so far; not a surplus of sellers. I'm keeping an eye on ER2 weakness. If that is leading the downside, I'm more likely to lean short on ES. Volume moderately high in first 5 min. No big institutional push.

8:33 AM CT - Note that we've already retraced the preopening gain in NQ. Market very much following the scenario mentioned earlier of testing high, failing the test, and perhaps establishing range bound trade.

8:18 AM CT - So far, so good. We've gotten our test of the previous day's high and now we've pulled back into yesterday's range. My main idea for the early morning is to watch for early buying in the NYSE TICK and see if we get a situation where the buying can't push us to new price highs. If that occurs, I'll be willing to sell the market and take a short-term position for a move back into the thick of the preopening range. My larger idea and main trading priority for the day is to see if we can get some weakness and then get to a point where the market holds up well on subsequent selling. At that point, I'd be a buyer to ride a possible move higher over the next few days, in keeping with the research posted today. Note that interest rates have already bumped up over the level from the post-Fed announcement reaction. Dollar/Yen has also retraced its post-Fed move. I don't think the Fed announcement really changed things in a macro sense--another factor leading me to believe we could see a digesting of yesterday's gains today. Back after the open.

7:57 AM CT - So here are three steps I take to prepare for the day's trade: 1) I conduct my studies to see where there might be a directional edge. The better the studies look, the more aggressive I'm likely to be in trading. My studies take a look at what is unique and distinctive about the recent market. I then examine occasions when these distinctive features have occurred in the past and what the market has done in response. So, for instance, the broad and strong upside momentum is most distinctive about the current market. My research, posted recently, found no edge the following day, but a solid upside edge over the next few days. That helps me frame my strategy for the upcoming trade. 2) I look at the larger market context. I want to put the current market into a framework that helps me understand what's happening at a longer time frame. We had a sharp decline on housing mortgage woes and concern over the Yen carry trade; that led to unusual bearishness in options sentiment. That, along with the improving market internals, told me that market history was not on the side of the bears. With the recent rally, we're now in a position to be testing the bull market highs. As noted in the entry for today's Weblog, we're already seeing new highs among 43 S&P 500 stocks and a couple of market sectors. That suggests to me that, while we may very well see near-term consolidation of market strength, the general direction is likely to be to the upside. 3) I examine the most recent market ranges. This helps me gauge where we have support and resistance and where we're most likely to have breakout moves. My last post a few minutes ago outlined some of those ranges. More broadly, I look at the previous day's trade as a range, since 85+% of all days are *not* inside days. That means that we'll either take out the previous day's high or low. In a strong market, we should be able to take out the prior highs. Indeed, as I've been writing, that's exactly what's happened. The best trades will occur when your view of the short-term ranges coincides with your research and your view of the longer-range context. My preference is to be selective with trades and wait for all three of these factors to line up, rather than to trade actively with a dubious edge. Back before the open.

7:41 AM CT - OK, we had some buying come into the market and take us to a new high for the premarket session. All of that fits well with the notion of testing yesterday's highs at 1449.75 in ES. Note that we have near-term support at 1442.75 - 1443. If we can hold above that support on any AM session selling, that would have me looking to test those highs. A lot of what I do in the early AM is play out "what-if" scenarios in my head to prepare myself for the morning trade. One of those scenarios is early AM selling holding above the overnight lows, leading us to break out of the overnight highs and test yesterday's lows. Another scenario is a failed test of yesterday's highs, leading us to return to the middle of the overnight range. Still another scenario is unexpected selling taking us below the overnight lows on negative NYSE TICK readings. That would normally have me waiting for the first bounce, gauging the market's ability to rebound, and then entering the downside if that ability to rebound is deficient. The scenario that's worked well the last few days--breakout to the upside on solid volume and TICK, followed by a pullback that you buy--is a lesser probability in my view, but I like to consider all angles, play them out in my head, and then be prepared when market action tells me what's happening. Back in a few with some context.

7:20 AM CT - Good morning! We have Initial Claims at 7:30 AM CT and Leading Economic Indicators at 9:00 AM CT. Let's see if those reports can nudge us out of a preopening range that has been pretty narrow so far. We saw strength in Asia and Europe on the heels of the post-Fed announcement rally, but the follow-through so far in the U.S. has been muted. This is consistent with the idea from the previous post, that high momentum rises tend to pause before resuming their upward course. Accordingly, I'm not planning a great deal of trading for my personal account today. My main goal is to buy on weakness for a holding period of up to four days, in accord with the posted research. Note that volatility in the premarket tends to correlate well with volatility in the morning session, so I'll be watching to gauge response to the Claims number. At this point, I'm looking at the pullback low following the surge yesterday afternoon as the bottom of a range and yesterday's high as the range high. Normally, in a market with solid upside momentum, we should be able to test that high. My research didn't find any upside edge on the day, however, which leads me to believe that any such tests of the high might be part of a consolidation, not another sustained upswing. Once again, I'll be tracking volume and buying/selling sentiment (Market Delta, NYSE TICK) to assess the likelihood of such a directional swing. Back shortly with some background for the day.

12 comments:

Tim said...

Thanks so much for your comments this morning. I have always watched TICK on 5min bars. What regressions settings do you use for 5min? Also I am assuming that when you discuss tick distribution, you are referring to the amount of time Tick remains below the middle line in the regression channel. Is that correct? Thanks again for all of your help!

Tim

Jeff said...

a quick technical question: why are you noting the 20-day average, and not some other period?
Eye-opening commentary, BTW.

Brett Steenbarger, Ph.D. said...

Hi Tim,

I look at how much time the TICK spends below its 20-day average level, which is the midline in the channel. The distribution of TICK often begins a shift with TICK breakout highs or lows.

Brett

Brett Steenbarger, Ph.D. said...

Hi Jeff,

I find the 20 day average to be representative of recent trade and to provide a worthwhile reference point re: current TICK values. It might be interesting to experiment with other values. Thanx--

Brett

Tim said...

Brett,

Thanks for your response. What would you use for regression settings on 5 min bars of tick?

Tim

Brett Steenbarger, Ph.D. said...

Hi Tim,

I don't use 5 min bars typically, but I'm not sure that would change anything. I'd still use the average 20 day TICK value as a benchmark--

Brett

bhong4664 said...

Hi Dr Brett
Let me share something I learned yesterday on a webcast by Tom Sosnoff of Think or Swim (TOS).

Based on his many years of experience as a pit trader and fund manager, He observed this pattern:

After a big blow-off move UP or DOWN (as we had yesterday), there tends to be a day of consolidation. During this consolidation, there can be violent intra-day moves as the Bears try to reassert their influence and Bull who come late to the table try to get in on the game.

The day following, if the Bears (in this case) are unable to take the market back down, there may be a end-of-day rally, signalling their capitulation.

It is this end-of-day rally that signals that the subsequent day would be a good countertrend day (to go short, in this case) as both sides will have exhausted their ammo. The scenario would, of course, be reversed following a big sell-off day.

Hope this is helpfull and, of course, I'm sure that this will stimulate your intellectual curiosity.
As always, I thoroughly enjoy your research and analysis.

Tim said...

Brett,

One more question...sorry! In the information you put up for us to read prior to today's session, you indicated that you use the regression channels in esignal on the 1 min tick chart. I thought you said the # of bars setting was 405 and the std dev was 2. I am confused about the 20 day average you are referencing. Please explain if you don't mind. Thanks! Tim

Brett Steenbarger, Ph.D. said...

Hi Bruce,

Thanks for that idea. My hunch is that the breadth/momentum of the initial move would play a huge role in whether the consolidation day would be followed by reversal vs. continuation. As noted in the recent research post, very strong momentum rises tend to be followed by further strength several days out.

Brett

Brett Steenbarger, Ph.D. said...

Hi Tim,

My bad for the confusion. My primary point of reference for NYSE TICK is the 20 day average TICK value, which is used to calculate the Adjusted TICK. If the Adjusted TICK is negative on the day, I look to hit downside pivot targets and vice versa.

The regression channel and 1 min TICK values is a way of monitoring average TICK over the past day to see if sentiment is sloping up, down, or flat. That is a shorter term view that helps identify distribution shifts in TICK.

Brett

AnaTrader said...

Hi Brett

Quote
Note that we have near-term support at 1442.75 - 1443. If we can hold above that support on any AM session selling, that would have me looking to test those highs
Unquote

Bearing in mind, it would likely be a rangebound day, but seeing that the support levels hold, I went long at 1444, for an intraday trade.

Although it swiftly moved up a couple of points soon after, it was truly a dead market for about two hours, and it being bedtime in Singapore, I decided to exit only to see the market finally moving up to 1448 level.....

Trading a rangebound market is a patient game and best not entered into, I learned from this and a few other instances in the past.

Brett Steenbarger, Ph.D. said...

Hi Anatrader,

I also trade less frequently and aggressively in narrow, range bound markets. The best trades, I find, are near the range extremes, where you can either get a decent breakout move or a reversion to the mean of that range--

Brett