(10 votes, average: 3.9 out of 5)
Loading ...Jump Off
If you haven’t already you should go read Wood’s series on the Double 7s right now. Come back later when you’re up to speed.
Secondly, this post has nothing to do with his series except for the first paragraph. It got caught in my mental net and I got to thinking about the 2day MA. At some point in the last year, I believe, Wood had posted a piece on the 2 DMA but through working with an inference engine (closed source unfortunately) I stumbled onto another system I will now share.
Rules
- Buy Rule = At the close, if the 2DMA for today will be lower than the 2DMA for yesterday, buy one share of SPY at the close.
- Sell Rule = At the close, if the 2DMA for today will be higher than the 2DMA for yesterday, sell one share of SPY at the close.
Results
- Testing period: Jan 4, 2007 – June 2, 2009
- Number of Trades: 104
- Average Trade: $.54/share
- Average Winning Trade: $1.88/share
- Average Losing Trade: -$2.76/share
- Win Percentage: 71%
- Total gain: $56.56/share
- 2 Tailed P Value: 0.0001
There’s not much to say to the above as it should be very clear for anyone that has been following the format of my spreadsheet trading systems. This may well be the ne plus ultra system for someone who is a hypermaniacal swing trader and who wants to leverage up via a number of means.
For the usage of most of my development philosophy this one is a bit too hyper but, in the spirit of giving someone enough of a howitzer to damage their visage I give you the FAS-Deuce.
Essentially, it uses the above trading system to buy, instead of one share of the SPY, one share of FAS.
With one caveat – one should use the ChiScore to time when to use the system. In essence, use this system for FAS only when the ChiScore is 7 or above (as noted in my other ChiScore related posts).
FAS-insanity Results
- Testing period: Nov 19, 2008 – June 2, 2009
- Number of Trades:6
- Average Trade: $.34/share
- Average Winning Trade: $.89/share
- Average Losing Trade: -$.75/share
- Win Percentage: 67%
- Total gain: $2.06/share
- 2 Tailed P Value: Not enough trades to calculate
In Summation
TNA, or SSO would probably be a better way to leverage gains for Deucey SPY however I’m quite impressesd with the accuracy of the system to pick the bounce day or days and couldn’t ask for a system with more accuracy.
This is a system under development so using a spreadsheet based system created by a pseudoanonymous blogger who resides outside the United States means that you take full responsibility for your own actions. I continue to reside in Halifax, Nova Scotia however that may be a developing situation in coming weeks.
Theme Song
Cuervo, Damian noted where I may have made a small error. I’ll know when I get home tonight. It won’t change the equity curve, but it might be that the system actually buys the lowest close of the last 3 days and sells the highest close of the last 12 days (reverse for short setup).
My code is using 2 and 11, but due to the need for an offset, it may actually be using 3 and 12.
AGain, I’ll update later.
Thanks Wood, I’m off to start looking at first derivatives (aka moving averages) differently.
ne plus ultra, ey?
Might have to get it on some of that then – wonder if it would translate well to SPY calls, which could theoretically prevent big losses (since the debit is all you risk) while still allowing the big winners…
Cuervo, it is 3 and 12.
The system looks back over the past 2 and 11 days, and then executes on the 3rd or 12th.
Thanks for the update Wood.
One question – how far back does your 3/12 system lose money on 60% or more of the trades?
Cuervo, that doesn’t make sense to me. Lose money on 60% or more of the trades? I’m sure you are asking something very simple, and I’m over complicating it.
How far back do you have to go before the system loses money on 60 percent of it’s trades?
Example, the system has been winning 7 out of the last 10 trades but if you go back to 1996 it wins 4 out of the last 10 trades.
I don’t have a clue, and I’m not aware of a canned report that would tell me…and I’m in too much of a good mood to do any REAL work tonight…lol…you want me to send you an excel file with the trade-by-trades?
Actually your update cleared my questions – I am referring to the grid table of monthly returns.
This system can be used to daytrade – though I don’t recommend it.
I real-timed the trade in PPT and on marketguru.com
Picked up a nice 4.8% trade
^^^
Uh, this was used to time FAS today.
I’m down with the deucey.
Went long some SPY calls at the close.
I see it’s up to 94.45 in AH.
Question, could you post (or just email me) how you use options for this system? My fundamentals of options are pretty basic but I’m curious.
I’m still figuring out how ‘fancy’ I want to get as far as a ’system’.
Right now, it’s just buying a front month ATM call when the buy signal hits.
Theoretically (and disregarding effects of volatility, decay, etc.), options should allow me to capture all the same gains as buying stock while avoiding the ‘blow-ups’ since my risk is predetermined and max loss is capped at the net debit paid.
ATM = at the money?
The other weirdness was that close. At the open I made a calculation as to what the buy signal should be and now it’s appearing as a close above 94.58 on the SPY will trigger based on a 94.16 entry on SPY.
Of course, if you’re daytrading with it you could just as well have exited about ten minutes ago.
On the way is a post using Deucey to time all time Fly favourite TNA.
I went looking for the day long series on TNA to no avail.
Yeah, ATM = at the money.
ITM = in the money.
The drawback of the ATM options is that if the SPY finishes flat or slightly up, it most likely won’t capture all the gains due to time & volatility decay. Such as right now, the SPY is slightly up, but the ATM options are unchanged or slightly down.
The further ITM you go, the less you experience these effects…however, you also increase your risk to a potential homo-hammer due to the larger upfront costs.
Well there we go, booked a small win as I’m out of the SPY 93. calls at $2.55 ($2.42 entry). Right about 5% gain, which is miniscule, really, but I’ll take it on a flat/doji day.
1 question…does the system ‘double down’ on consecutive closes below the 2dma? If so, does it still only sell 1 at-a-time, or does it sell all at the first available opportunity…
It doesn’t double down per se however, it can be used to day trade (sorta) as I’ve been experimenting with:
1. At any point if there is a buy signal then buy share and sell when sell signal crosses.
THEN, use the price you sold at as the last of the two prices for the set to calculate the next entry.
The other way to play it is to correlate – if the SPY signal hits then go with USO, TNA, and SPY.
I tried that with FAS, USO, TNA and got singed because of the goofy counter-correlation with regards to the banks.
Hope that makes sense.
Re 5% – doing that on a consistent basis is better than big gains randomly in my book.