The Weekend Dip Myth

[Stephen Gornick]

Buy after the weekend dip on Tuesday and sell before it on Friday/Saturday.

Using volume-weighted adjusted price.

Buy Tues Jul 17 ($8.34) Sell Friday Jul 20 ($8.20).  Lose 2% (after fees)
Buy Tues Jul 24 ($8.63) Sell Friday Jul 27 ($8.89)   Gain 2.5% (after fees)
Buy Tues Jul 31 ($9.21) Sell Friday Aug 3 ($10.73)  Gain 16.1% (after fees)
Buy Tues Aug 7 ($10.85) Sell Friday Aug 10 ($11.41) Gain 4.7% (after fees)

All total, about a 25% gain.

versus

Buy Tues Jul 17 ($8.34) And Sit On Them.  If sold right now, the current price ($11.45) gives a Gain 36.8% (after fees).

[ruski]

Using volume-weighted adjusted price.

Buy Tues Jul 17 ($8.34) Sell Friday Jul 20 ($8.20).  Lose 2% (after fees)
Buy Tues Jul 24 ($8.63) Sell Friday Jul 27 ($8.89)   Gain 2.5% (after fees)
Buy Tues Jul 31 ($9.21) Sell Friday Aug 3 ($10.73)  Gain 16.1% (after fees)
Buy Tues Aug 7 ($10.85) Sell Friday Aug 10 ($11.41) Gain 4.7% (after fees)

All total, about a 25% gain.

versus

Buy Tues Jul 17 ($8.34) And Sit On Them.  If sold right now, the current price ($11.45) gives a Gain 36.8% (after fees).

Buy and hold is such an underrated strategy, isn't it?

edit:
Probably sounding like a broken record about false walls by now, but I'm betting the sell back at $12 will be coming down too. $12.67 looks real enough. Here's what the depth chart should really look like:

[Stephen Gornick]

For scorekeeping purposes, this is the first weekend since mid-May that The Weekend Dip Indicator has flashed GREEN.

Not only does it show that we are in a 7-day downtrend versus the previous 7-day period (as-of Wednesday evening in the west, Thursday a.m. in the east) but this is a 3-day weekend in the U.S., so there's an extra day to dip.

 - http://bitcoincharts.com/charts/mtgoxUSD#rg10zigDailyzczsg2012-08-16zeg2012-08-31ztgSzm1g10zm2g25

So, at this moment the BTC/USD is about $10.70.   Applied properly over the past 18 hours maybe the level was a few cents higher, but $10.70 is a good spot to mark.  Let's see if the pattern repeats where selling now and waiting through Saturday afternoon (west) or Sunday am (east) with cash ready to buy those coins back at better prices ends up being a profitably strategy.

Banks in the U.S. are closed Monday so even if there is a rise on Sunday (west) it might still pay to wait a bit before committing all funds back in.  The sellers will be selling on Monday too and they don't need the banks to be open, but the (bigger) buyers who wire in new funds do!

Because BitInstant accepts cash over the weekend (available in U.S., Brazil and Russia), it is possible that a sharp dip will have a sharp recovery as well, but there's only so much in reserves at BitInstant -- probably not enough alone to worry about.  Let the sellers come to you.

Good luck, fellow Weekend Dippers!

Disclaimer: There are no guarantees.  You may lose money following this strategy.

[Stephen Gornick]

For scorekeeping purposes, this is the first weekend since mid-May that The Weekend Dip Indicator has flashed GREEN.

Not only does it show that we are in a 7-day downtrend versus the previous 7-day period (as-of Wednesday evening in the west, Thursday a.m. in the east) but this is a 3-day weekend in the U.S., so there's an extra day to dip.

Score one more for the Weekend Dip strategy!   (again, this is not something that occurs every weekend, but specifically when a downtrend exists and follows the specific criteria defining when the GREEN light should be flashing on or not.)  It has been profitable for like seven out of the eight times it has occurred so far in 2012.

After selling at $10.70 on Thursday when the weekend dip indicator flashed GREEN, there have been plenty of chances this weekend to get back in at under $10.

Banks in the U.S. are closed Monday so even if there is a rise on Sunday (west) it might still pay to wait a bit before committing all funds back in.  The sellers will be selling on Monday too and they don't need the banks to be open, but the (bigger) buyers who wire in new funds do!

The exchange rate is over $10 now, and I'm locking in my gains here ($10.10) just to guarantee a roughly 5% gain (after subtracting trading fees).  While getting back in without waiting until Monday (bank holiday in the U.S.) might cause me to miss an even lower re-entry point, I'm looking at the absolutely pathetic trading volume and don't see sellers as being anxious to sell at these levels.  So I'm going to take a smaller, certain gain here rather than risk possibly having to chase a rally when I was instead hoping for another dip.

Those following the weekend dip strategy and who happened to have had perfect timing when buying their coins back at the weekend bottom (so far) of $9.75 saw an 8% gain (after subtracting for trading fees).

[proudhon]

Well played.  I keep toying with the idea of getting back into trading, but I worry that I'll be sloppy because I was able to so easily collect so many coins in the fall from $32 to $2.  In a sense, that sloppiness is why I lost 150-170 coins in Bitcoinica.  It didn't feel like I was risking anything, but I don't want to fall into that trap again, which is why I moved everything offline for a while.

[adamstgBit]

Well played.  I keep toying with the idea of getting back into trading, but I worry that I'll be sloppy because I was able to so easily collect so many coins in the fall from $32 to $2.  In a sense, that sloppiness is why I lost 150-170 coins in Bitcoinica.  It didn't feel like I was risking anything, but I don't want to fall into that trap again, which is why I moved everything offline for a while.

buy sell hold

good move

[proudhon]

Well played.  I keep toying with the idea of getting back into trading, but I worry that I'll be sloppy because I was able to so easily collect so many coins in the fall from $32 to $2.  In a sense, that sloppiness is why I lost 150-170 coins in Bitcoinica.  It didn't feel like I was risking anything, but I don't want to fall into that trap again, which is why I moved everything offline for a while.

buy sell hold

good move

In the days of the run-up to $15.xx I was convinced a crash was coming and wanted to move everything to MtGox to take advantage of it, but (1) all my coins were offline without any digital copies, and (2) I was out of town and had no way to access my paper wallets.  That's the trade-off I consciously made, of course.  And there's no way to know that if I hadn't done that I wouldn't have lost a bunch in trading throughout the year.  But my certainty of the impending crash this go around has me feeling a measure of regret in getting so protective.

[adamstgBit]

Well played.  I keep toying with the idea of getting back into trading, but I worry that I'll be sloppy because I was able to so easily collect so many coins in the fall from $32 to $2.  In a sense, that sloppiness is why I lost 150-170 coins in Bitcoinica.  It didn't feel like I was risking anything, but I don't want to fall into that trap again, which is why I moved everything offline for a while.

buy sell hold

good move

In the days of the run-up to $15.xx I was convinced a crash was coming and wanted to move everything to MtGox to take advantage of it, but (1) all my coins were offline without any digital copies, and (2) I was out of town and had no way to access my paper wallets.  That's the trade-off I consciously made, of course.  And there's no way to know that if I hadn't done that I wouldn't have lost a bunch in trading throughout the year.  But my certainty of the impending crash this go around has me feeling a measure of regret in getting so protective.

ya same here

but i did have a bunch on hand, which i sold and i'm buying back now

i thought about redeeming my paper wallets too

[Stephen Gornick]

I'm locking in my gains here ($10.10) just to guarantee a roughly 5% gain (after subtracting trading fees).

On IRC I was asked how I can refer to this weekend dip trade as a gain when really what I did was close a prior position when selling last Thursday and then opening a new position today.

I had explained that if you are still of the mindset that your wealth is measured in dollars (or whatever form of fiat you generally use), then that would be correct -- the weekend dip trade closes a previous trade and opens a new one.  If instead you think of bitcoins as being your base currency, then this weekend's dip trade resulted in a gain on an investment held just over the weekend.

But I'm debating on whether or not to phrase the trade using the right terms.  Instead of saying I was selling coins on Thursday what I should have said was that I was buying dollars.

Then today, I sold those dollars and as a result of that trade I got back about 5% more bitcoins versus the amount I had spent on Thursday.  Thus using the term "5% gain on the weekend dip trade' is the correct way to phrase it.

It gets confusing switching back and forth on what is referred to as the base currency though.

[molecular]

I'm locking in my gains here ($10.10) just to guarantee a roughly 5% gain (after subtracting trading fees).

On IRC I was asked how I can refer to this weekend dip trade as a gain when really what I did was close a prior position when selling last Thursday and then opening a new position today.

I had explained that if you are still of the mindset that your wealth is measured in dollars (or whatever form of fiat you generally use), then that would be correct -- the weekend dip trade closes a previous trade and opens a new one.  If instead you think of bitcoins as being your base currency, then this weekend's dip trade resulted in a gain on an investment.

But I'm debating on whether or not to phrase the trade using the right terms.  Instead of saying I was selling coins on Thursday what I should have said was that I was buying dollars.

Then today, I sold those dollars and as a result of that trade I got back about 5% more bitcoins versus the amount I had spent on Thursday.  Thus using the term "5% gain on the weekend dip trade' is the correct way to phrase it.

It gets confusing switching back and forth on what is used to refer to as the base currency though.

+1. You got it right. BTC is the default, having USD is a position.

[glub0x]

Using volume-weighted adjusted price.

Buy Tues Jul 17 ($8.34) Sell Friday Jul 20 ($8.20).  Lose 2% (after fees)
Buy Tues Jul 24 ($8.63) Sell Friday Jul 27 ($8.89)   Gain 2.5% (after fees)
Buy Tues Jul 31 ($9.21) Sell Friday Aug 3 ($10.73)  Gain 16.1% (after fees)
Buy Tues Aug 7 ($10.85) Sell Friday Aug 10 ($11.41) Gain 4.7% (after fees)

All total, about a 25% gain.

versus

Buy Tues Jul 17 ($8.34) And Sit On Them.  If sold right now, the current price ($11.45) gives a Gain 36.8% (after fees).

I must say that in the first case you gain +25% of your capital, when in the second case you gain +36% on the value of thoose bitcoins.
Wich is not the same.
The best is to think in term of bitcoin capital
Case 1 : +25%
Case 2 : +0%

In the case 1 you would have make around +60% in USD wich is pretty nice

EDIT But i'm not so sure if this post was a joke or not ...

EDIT 2 my mistake ...

[Stephen Gornick]

Weekend dip indicator is GREEN:

Will the weekend dip strategy pay off a second week in a row?  

The high for the previous seven days was yesterday's  ... $11-ish print.   The high in the previous seven day period was above $12.  So we are in a week-to-week downtrend, and the weekend dip strategy says that when this occurs, doing a SELL on Wednesday evening (west) / Thursday morning (east) has historically given the opportunity to buy those coins back at a cheaper rate over the weekend (check the chart starting on Saturday, and if the prices start rising to buy back in -- otherwise can wait until early Monday even to buy back.).

The concern for this weekend might be that lots of USDs were withdrawn from BitFloor today.  If those USDs clear tomorrow and then get deposited elsewhere for credit to another exchange, they might represent pending purchases that were held-up due to the BitFloor security breach.  Personally, I doubt that is going to be a significant amount of funds for that purpose, so ... GREEN light, sell!    The current price is $11, but I'm a little late on posting this, the proper amount should have been at a level a little lower, about $10.80, so that's the amount I'll use in determining if the weekend dip strategy succeeded yet again at generating profit.

[anu]

Weekend dip indicator is GREEN:

Will the weekend dip strategy pay off a second week in a row?  

The high for the previous seven days was yesterday's  ... $11-ish print.   The high in the previous seven day period was above $12.  So we are in a week-to-week downtrend, and the weekend dip strategy says that when this occurs, doing a SELL on Wednesday evening (west) / Thursday morning (east) has historically given the opportunity to buy those coins back at a cheaper rate over the weekend (check the chart starting on Saturday, and if the prices start rising to buy back in -- otherwise can wait until early Monday even to buy back.).

Are you trying to make it a self fulfilling prophecy?

[Stephen Gornick]

Are you trying to make it a self fulfilling prophecy?

Well, game theory shows that never works.  Nope, just surprised at how accurate that signal historically has been.   Maybe it is like rolling the same number five times in a row -- doesn't mean roll number six is any more likely or any more unlikely to be the same number.

But it isn't like some obscure concept.  It is a pattern that occurs what appears to be simply because bank wires don't clear on weekend days.

[dree12]

Are you trying to make it a self fulfilling prophecy?

Well, game theory shows that never works.  Nope, just surprised at how accurate that signal historically has been.   Maybe it is like rolling the same number five times in a row -- doesn't mean roll number six is any more likely or any more unlikely to be the same number.

But it isn't like some obscure concept.  It is a pattern that occurs what appears to be simply because bank wires don't clear on weekend days.

I have an alternate theory: it's a pattern that occurs because too many people think it is a myth. That explains why the weekend dip tends to delay to Sun-Mon.

[Stephen Gornick]

Weekend dip indicator is GREEN:

Will the weekend dip strategy pay off a second week in a row?  

[...]
about $10.80, so that's the amount I'll use in determining if the weekend dip strategy succeeded yet again at generating profit.

Doh!  

Well, that was a disappointment (to those holding dollars over the weekend) -- as the expected dip didn't dip!

My reference trade on Thursday followed by the reverse trade just a bit ago ended up seeing a nearly 3.5% loss as a result.

These patterns work until they don't.

Now there's no denying one thing -- a whole lot of people just learned of Bitcoin (or heard of it before but this is now the third time perhaps they hear it) due to the Romney tax blackmailing and as a result that might have resulted in either new buyers and/or sellers simply being less willing to sell seeing the web traffic.  Or it may have had little effect and no dip would have occurred anyway.  Who knows.

Here's the Google trends 30-day for Bitcoin.  This isn't just a spike and quick drop, ...  this has been high since the Bitfloor debit card news, then the rally to $15 and then the pirate mess, and has stayed high since.



 - http://www.google.com/trends/?q=Bitcoin&ctab=0&geo=all&date=mtd&sort=0

[Edit: The link showing the first burst of media attention in late August:

 - http://bitcointalk.org/index.php?topic=102713.0 ]

[lucif]

I think weekend dip effect might be connected to manipulation strategies when some parties want to slope weekly candle to desired figure when candle is about to close.

[elux]

Bump! 

[Spekulatius]

I think weekend dip effect might be connected to manipulation strategies when some parties want to slope weekly candle to desired figure when candle is about to close.

Maybe the weekend dip has a fair place in other securities, but doesnt apply for bitcoin. It would have to be for some difference, that is not present in the bitcoin market that causes the weekend dip in other markets. E.g. the 24/7 trading of bitcoin opposed to 24/5 trading in traditional forex.

[Stephen Gornick]

Maybe the weekend dip has a fair place in other securities, but doesnt apply for bitcoin.

Though this weekend didn't have a Weekend Dip GREEN light, going short into this weekend would have given a couple percent or four gain (after subtracting exchange fees), if the timing were perfect:



 - http://bitcoincharts.com/charts/mtgoxUSD#ig2-hourzczsg2012-09-21zeg2012-09-23ztgSzm1g10zm2g25


But when there's no green light, this doesn't always play out.  There would have been nearly a 5% loss trying the weekend dip trade last week (heading into the Bitcoin 2012 conference weekend):

 - http://bitcoincharts.com/charts/mtgoxUSD#ig2-hourzczsg2012-09-14zeg2012-09-16ztgSzm1g10zm2g25


So there isn't that great of a pattern except during downtrends -- which is the only time when the GREEN light will go on, and has resulted in a profit well more than 50% of the time when it does go green.  The profits in following the strategy (only when GREEN is on) have more than offset the losses from weekends where it was on but gave an incorrect signal.

Maybe this weekend dip is just coincidence, who knows.