daSARa system

augubhai

Well-Known Member
3rd consecutive loss in ORB. Average returns down to +8.77 points/day (pre-costs).

While on Wednesday, it took 90 minutes to hit my SL, in the last 2 days I am out within 2-3 minutes. After that what do I do? I have already finished 2 rounds of shopping errands today. How do I justify myself being a full time trader?

High time I got an alternative system that needs more chart time from me. Last year, I used just take up any system that I felt like and trade the whole day, but now I want to trade only after some amount of backtesting, and that is holding me back...
 

Tlahuicole

Well-Known Member
Augubhai today your original dasara system would have worked fine as today was a trending day, is it right? Did you check charts with your dasara system in mind?

Sad to see ORB is irritating you.
 

augubhai

Well-Known Member
Was trying to find a positional system, and landed up with another intraday system. Like Columbus found America or... http://youtu.be/HVbxnE2w2PA

Since I am scared of overnight gaps, was testing this system for long options. Then found that it worked better with 30 min futures, and even better with 15 min futures.

Here are the monthly backtest results for 15 min NF, positional:
Code:
	Trades	Gross pts	Cost=2 pts	Cost=3 pts	Cost=4 pts	Cost=5 pts	Cost=6 pts
Aug	49	1689.65	1591.65	1542.65	1493.65	1444.65	1395.65
Sep	42	393.05	309.05	267.05	225.05	183.05	141.05
Oct	56	391.80	279.80	223.80	167.80	111.80	55.80
Nov	45	218.20	128.20	83.20	38.20	-6.80	-51.80
Dec	40	-294.00	-374.00	-414.00	-454.00	-494.00	-534.00
Jan	30	553.55	493.55	463.55	433.55	403.55	373.55
							
Total	262	2952.25	2428.25	2166.25	1904.25	1642.25	1380.25

*Current open position in backtest is 3x short
The results are mainly dependent of overnite gaps. Taking cost = 4 points/trade, +1542 points in August is great :thumb:, but is mainly due a couple of overnite gaps in favour. Similarly, -414 points in december is bad :(, but is mainly due to overnite gaps against positions. And with the kind of (bad) money management that I do, -400 points will sink the account.

So, then tried intraday. The results:
Code:
	Trades	Gross pts	Cost=2 pts	Cost=3 pts	Cost=4 pts	Cost=5 pts	Cost=6 pts
Aug	49	1401.80	1303.80	1254.80	1205.80	1156.80	1107.80
Sep	42	317.70	233.70	191.70	149.70	107.70	65.70
Oct	54	341.20	233.20	179.20	125.20	71.20	17.20
Nov	42	289.85	205.85	163.85	121.85	79.85	37.85
Dec	39	-46.65	-124.65	-163.65	-202.65	-241.65	-280.65
Jan	32	592.90	528.90	496.90	464.90	432.90	400.90
							
	258	2896.80	2380.80	2122.80	1864.80	1606.80	1348.80
So now gains in August and losses in December both reduced + no overnite risk. No need to be at the terminal the next to day just because of overnite positions. The returns have smoothed somewhat compared to the positional version. -163 points in December is still bad, but an improvement over -414 points.

Backtest is covered just a few months, but I am actually thinking of trading this version from tomorrow.

The method: Quite simple. Enter/add on high volume bars. That's all!!

People say that market movements are controlled by market makers, and fool other people like me by creating different price patterns. The premise of this method is to catch them when they expose themselves. High volume bars cannot be created just for fooling people. These are bars where the market player/s have made a commitment. So this method only trades the high volume bars, and ignores the rest. This backtest took trades on all high volume bars - whether wide range or narrow range - just to get more entries. It is up to you to define what a high volume bar is.

in theory, this system should work better on certain stocks that are firmly controlled by market makers, but i haven't backtested. so until I backtest, I will go with index. Lower margins on index futures help.
 

augubhai

Well-Known Member
Thought of analyzing the December results to fix whatever possible. The statistics are surprising.

(The results are actually for the series, not for the month)

Here are the daily results of the Intraday version:
Code:
07/30/14	138.30
07/31/14	-14.00
08/01/14	-9.40
08/02/14	-3.70
08/05/14	4.10
08/06/14	347.85
08/07/14	-25.60
08/12/14	1.35
08/13/14	40.15
08/14/14	-29.90
08/16/14	628.75
08/19/14	-37.05
08/20/14	-21.45
08/21/14	410.20
08/22/14	-19.90
08/26/14	-37.65
08/27/14	40.35
08/28/14	5.25
08/29/14	-15.85
08/30/14	4.35
09/03/14	290.75
09/04/14	22.25
09/06/14	-4.65
09/10/14	-17.80
09/12/14	-46.05
09/13/14	-28.50
09/16/14	166.85
09/18/14	34.65
09/19/14	-24.05
09/20/14	-53.00
09/24/14	3.60
09/25/14	-36.05
09/26/14	5.35
09/30/14	48.15
10/01/14	7.05
10/03/14	48.45
10/04/14	-83.10
10/07/14	-16.55
10/08/14	-28.85
10/09/14	305.10
10/10/14	-21.70
10/11/14	-17.10
10/14/14	6.50
10/15/14	80.20
10/17/14	-39.45
10/18/14	43.00
10/22/14	-12.75
10/23/14	-29.95
10/24/14	-0.35
10/25/14	-2.25
10/28/14	-21.75
10/29/14	54.95
10/31/14	21.60
11/01/14	-5.15
11/05/14	47.80
11/06/14	-17.15
11/07/14	78.95
11/11/14	-18.90
11/12/14	50.65
11/13/14	-15.20
11/14/14	-12.20
11/18/14	18.80
11/20/14	75.75
11/21/14	134.10
11/22/14	-8.50
11/25/14	-2.25
11/26/14	-17.30
11/27/14	-7.85
11/28/14	-11.70
11/29/14	1.45
12/02/14	-17.80
12/03/14	4.55
12/04/14	-3.70
12/05/14	2.90
12/06/14	-3.10
12/09/14	-19.40
12/10/14	-7.65
12/11/14	26.45
12/12/14	44.90
12/13/14	-23.00
12/16/14	-24.45
12/17/14	14.90
12/18/14	-10.60
12/19/14	-12.00
12/20/14	0.30
12/23/14	-8.45
12/24/14	-9.20
12/26/14	-2.75
12/27/14	-0.80
12/30/14	4.40
12/31/14	-7.75
01/02/14	493.25
01/03/14	-20.00
01/06/14	-16.95
01/07/14	10.95
01/08/14	-12.65
01/09/14	-14.30
10/10/14	156.75
There are no extreme loss days in December that can be fixed. The highest loss in December is -24.45.
Days when loss greater than 25:
August = 4, September = 4, October = 4, November = 0, December = 0, January = 0
:confused:

The key to the result is in the profit days... Day's when the profit was a century or half century:
August = 4 centuries, 0 half centuries
September = 2 centuries, 0 half centuries
October = 1 centuries, 2 half centuries
November = 1 century, 3 half centuries
December = 0 centuries, 0 half centuries
January = 2 centuries, 0 half centuries

Sum of 50+ score days vs. other days:
August = +1525.10 vs. -123.30
September = +457.60 vs. -139.90
October = +440.25 vs. -99.05
November = +339.45 vs. -49.60
December = 0.00 vs. -46.65
January = +650.00 vs. -57.10

Moral of the story:
A century a month, can kind of save this system. To be in play, this system needs some big scores. This is not about avoiding big losses, this one is about getting the big scores.

Has the character of the markets changed in the last few months? If that kind of a change has this big an impact on the results, then I should be less confident about this method. At this point, I am not sure if this analysis should make me more confident or less confident about the system.

Many good moves were missed while waiting for a good volume bar. Looking at the December charts - Dec 20 could easily have been a 100+ (nice ORB day), an early entry on Dec 18 would have got good points. other than these days, it was a flattish month on 15 min intraday charts. Should I junk the volume bar logic for something else?? :confused:
 

augubhai

Well-Known Member
So, in the last 2 days, I surprised myself by getting some analysis right. (I think it helped that there was no money on the line)

Since I do not have OI on my charts, I thought I will record the timestamp of the comments on the charts. So, here are the comments and charts...


Comment 1:
Trying to analyze....

Big build up in 6200 puts and 6300 calls unwinding, so bias is up. Break above yesterday's high should be sudden, because there must be a lot of stops above it - need to observe if that BO sustains. Above 6300 is a big fluid zone.
Comment 2:
If it breaks out in the next 30 min above 6270, might be a quick ride to near 6300. Maybe a good idea to place Buy Stop at 6269 and tight SL after entry.

Just my paper gamble...
Comment 3:
Ok, I am claiming the bragging rights :D
Comment 4:
Today's initial bias: down. Puts unwinding at 6300
Comment 5:
Here's what I am thinking. New Puts are now being written with vigor at 6300, but they are still 1/3 of Calls added at 6300. The build-up on both sides is slow, but if it continues, the key is to watch the Calls side. If there is a sudden spike in OI of 6300 Calls, then NF will likely have a BOF of HOD, for the quick move down to 6250. Anyway, don't expect 6300 to be broken with the current build-up.
Comment 6:
6250 has reasonable support, will break only if momentum appears, which is not there at present. I am still thinking of a reversal into range. And if it goes touches HOD, the BOF scenario is very likely... Only paper trading this scenario...
Comment 7:
Bias at 6300 is now hopelessly down. increase in Calls to Puts at 6300 is 46:1. Unless the ratio reduces on a pullback into 1st hour range, touch of HOD is unlikely. 6250 still supporting.
Comment 8:
I might be wrong, but I think it is a minor accumulation at 6250.
Comment 9:
If it was accumulation at 6250, it was a very slow and bad one. Volumes traded below 6255 were just about 12% of the day's volume. So one more dip down to accumulate more?
Comment 10:
Ok, now we arrived back at the low of the 1st hour range, but Change in OI of 6300 Call is 801K, and for 6300 Puts it is 39K. Not good enough for a test of HOD. I don't expect the Change in Calls OI to come down drastically today, but if around 6280, the Change in Puts OI rises to 120-150K or higher, expect the BOF of HOD within 20 minutes.
Comment 11:
Change in OI of 6300 Puts is almost 0 now. No more upside possible. Only need to see if 6250 supports.

 

Similar threads