If you click on the link below, it is Block's training class notes provided by Craig and Julia.

You can browse thru it, and try out some of the exercises that appeal to you. New Block's developments get to the training classes very quickly.

Another good way of learning, is to watch some videos from the blocks site, working your way backwards from the most recent.

David. A few more points on Darvas.

He liked bull markets with a good selection of  rapidly rising stocks.

He was interested in maximum growth. "This being the case, it seemed only logical to choose those stocks that had already showed what they could do by actually doing it."
He wanted "stocks not merely rising in price, but actually in their highest boxes ever."

His potential buys had to have at least doubled within the last year.

He was very aware of  the "rhythm of a stocks advance" eg1  "….and sold at 85.5 when it failed to move up through the new box as quickly as I thought it should."
Eg2 … the stock suddenly seemed to darvas to have "lost its will to rise …. It almost seemed ready to tumble."

He doesn't say it specifically but its implied that he had 2 different stop loss positions.
1) When he does a small pilot buy into a new stock that has just broken out of its highest box. He does this with a buy on stop, at a price just fractionally above the high of  the  old box, and then places a stop, just fractionally under this high. This way he only loses about 1 or 2  percent, including commission.

2) If  the stock keeps running up, into higher boxes then he will move the stop upwards, to just under the bottom of  these higher boxes.

He was prepared to be stopped out a few times, and buy back in, if the stock was otherwise behaving itself. After all he is only losing 2% of a small position each time.

He was very aware of keeping commissions down, as they were very high in his day. "I buy the most expensive stock I can afford, and I buy it in round lots."

Bustermu
Part of Sam Allende's post on 9 march 2008 9:58:24 AM

"The logic behind this is the following. The probability is very high, that after the initial two 9 day periods which are down, there will not be two more 9 day periods which are down (subscribers can see this easily; if using percent true, just put in C>C1 and C1 ... there will be more times when there are not at least four consecutive down periods, than when there are).

Blocks3 analysis continued.

We now move to the 9 day bar charts that end 7th march 2008.

Perhaps I didn't explain myself clearly enough previously.

1 bar closing up ( a 9 day bar, bar#1 ) followed by 2 down closing bars (bars 2&3) has 30,139 occurrences in all Worden's S&P 500 stock data back to the 1980's. This is the pool of all possible investments.

Buy price is 2% less than the closing price of the 2nd down bar (bar3 )

No. of achievable investments, at this buy price, either at the open of the 4th bar (day1 ) or during the range of the 4th bar (days 1 to 9 ) = 19,340

Selling took place either at the OPEN of the 5th bar (day 10 ), if this OPEN was >= buy price, or otherwise at the OPEN of the 6th bar ( day19 )

$1000 invested on each of the 19,340 stocks leads to a GAIN of $211,879.

 %gain = 1.1% approx.  These are truly amazing results. 

WARNING
It must be stressed that the chance of a coding error is very high, particularly on the new blocks platform, and my beginner's understanding of it. In a few days, when I get more time, I'll endeavour to check my figures.

For what its worth, I tried breaking the 500 S&P stocks up into groups of 100 stocks at a time. Either there is the same systematic error present in all of this, or Sam's system is amazingly robust.

Groups  Possibles Actuals  Gain  %Gain
1st 100  6105  3969  47.28122 1.191%
2nd 100  5827  3662  41.49598 1.133%
3rd 100  6145  4028  45.35066 1.126%
4th 100  6026  3864  38.541  0.997%
5th 100  6124  3882  39.5094  1.018%

Sam

Some more stats. If you want to wait for 1 more down bar (another 9 days ) before investing

3 or more down bars ( 9 days remember ) in a row   = 13,363 possible investments, out of all the S&P 500 stocks.

Your worry then is 5 or more down bars in a row. This will happen 2,399 times, ie about 18%

Sam

Some statistics, using blocks3 and all the S&P 500 stocks, and 9 days per bar, charts.

2 or more down bars ( 9 days remember ) in a row   = 30,468  These are the ones you would be betting on. If the 3rd bar is up you should win. If the 3rd bar is down but the 4th is up you should be back near even.

4 or more down bars in a row  = 5,611  These are the losing bets out of the 30,468. Bets lost = 18.5% approx.

This is for the 9 day bar charts ending on 14 march 2008. The charts that end on 17th march, 18th march etc will be different, but I would expect similar statistics. 

Sam

You've explained everything very thoroughly and very patiently, I thank you for that.

Correct me if I'm wrong, but I imagine that you have the luxury of nearly always using limit orders, except perhaps, for the final sale at the open of the 19th day, where you are bailing out regardless. And as you state, these limit orders mostly work to your advantage. The initial buy is a limit that assures you of at least 2% MOS, but on down moves you frequently improve this. Your sell at open at day 10 is a limit, to at least exceed your buying price, or you have a second chance to make a profit at day 19. 

Sam

Thanks for an interesting series of posts and explanations. You mention that you like to buy about 12 stocks per day. I assume you have 9 rolling, 9 day watchlists active from which to choose your 12 stocks.

Also would you tend to buy less than 12 if your total number of open positions was building up.

In practice have you had large clusters of losses, or even groups of big losses, or are the losses well mixed as you would hope.