Markets still falling like whoa

[Bozo_Casanova 26437]

15 minutes ago, Superhero said:

My tax bill is going to suck this year.

That's because you made money. Pigs get slaughtered. 

[Incredulity 11655]

2 minutes ago, Bozo_Casanova said:

That's because you made money. Pigs get slaughtered. 

Pigs get fat, hogs.....

[T’Boo Ted Marshall 2732]

That's right.  Hogs go home and bang the prom queen.  

[Sbbruin 21829]

How does Gold and other precious metals purchases work? You don't just purchase positions, but physical metal, correct? So do you get a safe deposit box? That's what my dad did years ago, not sure if that still held true.

[Rusty Shackelford 414]

S&P 500 closed at the 200 DMA again today.  The market should go up from here in the near-term.

[Trey3216 15345]

38 minutes ago, Sbbruin said:

How does Gold and other precious metals purchases work? You don't just purchase positions, but physical metal, correct? So do you get a safe deposit box? That's what my dad did years ago, not sure if that still held true.

You can buy gold and silver etf's.  You can also buy bullion.  I store some of mine in my gun safe.  

[jimmyjazz 45892]

The "risk" on the ETF's is that you could lose your entire investment should the shit hit the fan.

I believe the only accepted definitions of the shit hitting the fan are (a) zombie invasion (b) US wiped out by giant hurricane or (c) military coup.  Personally, I find an ETF like GLD far easier to deal with than physical metal.  (Not investment advice.)

[Tailgate 3021]

As long as you capture the S&P and have a good balance of bonds depending on your situation...stay put. Even strong companies you believe in.

No one can time the market...but history tells us it will always come back and grow.

Been there in 2007 and 2008 and advisors and family were like “sell” and I was like....”no.”

It’s paid off in spades.

[bernorange 8359]

14 hours ago, Sbbruin said:

How does Gold and other precious metals purchases work? You don't just purchase positions, but physical metal, correct? So do you get a safe deposit box? That's what my dad did years ago, not sure if that still held true.

You can buy physical gold that you can hold in your hand.  Storage and security will be your concern.

You can buy physical gold in allocated accounts with a vaulting service (Texas operates one now).  Storage and security should be guaranteed, but you pay a small price for it.

You can buy claims to (unallocated) gold from some financial institutions.  Good luck ever translating those claims/shares to physical metal.  These are instruments for trading, not for insurance.

You can also buy ETFS that track the gold spot price, but they don't actually offer you a claim to real metal.  These are also just trading vehicles.

You can also buy contracts for metal on the comex, but after reading about it for several years now, I honestly have no idea how that wizardry actually works.  But I do know that it's a mine field.  MFGLOBAL holla. 

[Newdoc 3439]

15 hours ago, Trey3216 said:

  I store some of mine in my gun safe.  

this is where I store my gold.

[Aqua Buddha 14582]

On October 18, 2018 at 4:20 PM, happyfunball said:

Interest rates are still very low so while raising interest rates will dampen spending let's not delude people that money is tight. A proof point of this is that Private Equity multiples for M&A is at historical highs. 

For oil, volatility sucks but unless you are an airline or oil is a major COGs of your business then I don't think it has a huge impact. If we were looking at oil above $100 then I would have more concern about broader economy and market impact. 

When people raise concern about the slowing global economy, remember the US especially the consumer drives a big %.

I'd expect to see a lot of tech IPOs come to market in 2019 unless there is a significant global event / market correction.

Great chart.  I saw where Putin was talking about the end of American dominance the other day.  Meanwhile, Russia's economy is smaller than Italy's and without oil, they would be nothing.  Keep that in mind the next time Trump wants them in the G8.

[Goofyboy 3269]

When there is blood in the street, but!! Right?

[Fudge Nuggets 19189]

Due to recent volatility I figured my ERIC would be down and wanted to add to my position.  Nope, it’s up 10% this week.

[jimmyjazz 45892]

On 10/19/2018 at 4:55 PM, Tailgate said:

No one can time the market...

Uh, no.  It is completely possible to time the market.  In fact, it's trivially easy to show that it's possible using historical data.

 

Quote

Been there in 2007 and 2008 and advisors and family were like “sell” and I was like....”no.”

So you took it in the shorts and recovered?  Where would you be now if you had sold in the summer of 2007?

I absolutely loathe these silly "you can't" pearls of wisdom.  The historical growth of the market is an amalgamation of zillions of winners and losers.  Hedge funds wouldn't exist if there weren't at least SOME people who can "time the market".  Maybe you're speaking the truth about most ignorant novice investors, but it's insanely easy to beat the market using t utterly simple trading tools.  You can learn those tools in 5 minutes.

[Incredulity 11655]

Where do I send my 9.95 for your newsletter?

[Tailgate 3021]

2 hours ago, jimmyjazz said:

Uh, no.  It is completely possible to time the market.  In fact, it's trivially easy to show that it's possible using historical data.

 

So you took it in the shorts and recovered?  Where would you be now if you had sold in the summer of 2007?

I absolutely loathe these silly "you can't" pearls of wisdom.  The historical growth of the market is an amalgamation of zillions of winners and losers.  Hedge funds wouldn't exist if there weren't at least SOME people who can "time the market".  Maybe you're speaking the truth about most ignorant novice investors, but it's insanely easy to beat the market using t utterly simple trading tools.  You can learn those tools in 5 minutes.

Ok, have fun.

[jimmyjazz 45892]

1 hour ago, Tailgate said:

Ok, have fun.

Excellent counter.  I am having fun.  I'm making money.

[happyfunball 481]

4 hours ago, jimmyjazz said:

Uh, no.  It is completely possible to time the market.  In fact, it's trivially easy to show that it's possible using historical data.

 

So you took it in the shorts and recovered?  Where would you be now if you had sold in the summer of 2007?

I absolutely loathe these silly "you can't" pearls of wisdom.  The historical growth of the market is an amalgamation of zillions of winners and losers.  Hedge funds wouldn't exist if there weren't at least SOME people who can "time the market".  Maybe you're speaking the truth about most ignorant novice investors, but it's insanely easy to beat the market using t utterly simple trading tools.  You can learn those tools in 5 minutes.

1234

 

Howard Marks, cofounder of Oaktree capital, is on Bloomberg right now. Suggest people watch regarding economic and market outlook 

Edited October 22, 2018 by happyfunball

[troph 24781]

So you took it in the shorts and recovered?  Where would you be now if you had sold in the summer of 2007?
I absolutely loathe these silly "you can't" pearls of wisdom.  The historical growth of the market is an amalgamation of zillions of winners and losers.  Hedge funds wouldn't exist if there weren't at least SOME people who can "time the market".  Maybe you're speaking the truth about most ignorant novice investors, but it's insanely easy to beat the market using t utterly simple trading tools.  You can learn those tools in 5 minutes.

Okay if it only takes 5 mins I’ll listen. School me... not being surly, I’m serious.

[gsoda3 3248]

Algos had some to do with, and the rebalancing killed it off.  The thing that really killed it was not halting it prior to close and waiting til after hours to halt.  A large algo trade after hours triggered some other large trades to go through which triggered an implosion. Basically, if that listing falls a certain point over the course of one day, it could fold.  A huge Short trader came in and triggered a massive amount of algo stop losses, and those losses resulted in implosion

That's not what happened. Rebalancing and the timing of the halt didn't have anything to do with the implosion. XIV and SVXY were inverse vol instruments with accelerated redemption clauses that got triggered when certain events happened- that certain event was caused by a spike in volatility (I forget the exact number) that left the firms unable to roll forward their underlying contracts. The prices of VIX and SVXY traded according to their indicative redemption values, plus or minus a few pct points of premium or discount. As soon as the IV got updated at the end of the day the price followed.

So you took it in the shorts and recovered?  Where would you be now if you had sold in the summer of 2007?
I absolutely loathe these silly "you can't" pearls of wisdom.  The historical growth of the market is an amalgamation of zillions of winners and losers.  Hedge funds wouldn't exist if there weren't at least SOME people who can "time the market".  Maybe you're speaking the truth about most ignorant novice investors, but it's insanely easy to beat the market using t utterly simple trading tools.  You can learn those tools in 5 minutes.

Hedge funds don't time markets. I won't speak for every firm out there- most might make a few trades along those lines but it's definitely not the main trading strategy.

[nycHorn 2998]

15 hours ago, jimmyjazz said:

Uh, no.  It is completely possible to time the market.  In fact, it's trivially easy to show that it's possible using historical data.

 

So you took it in the shorts and recovered?  Where would you be now if you had sold in the summer of 2007?

I absolutely loathe these silly "you can't" pearls of wisdom.  The historical growth of the market is an amalgamation of zillions of winners and losers.  Hedge funds wouldn't exist if there weren't at least SOME people who can "time the market".  Maybe you're speaking the truth about most ignorant novice investors, but it's insanely easy to beat the market using t utterly simple trading tools.  You can learn those tools in 5 minutes.

 

8 hours ago, troph said:

Okay if it only takes 5 mins I’ll listen. School me... not being surly, I’m serious.

Seconded.

[jimmyjazz 45892]

Maybe we're working under different definitions of "timing the market", but in my view it's as simple as going to cash (or short) when conditions warrant.

At any rate, it's hardly a great secret:  a simple moving average crossover strategy historically beats the market.  I pulled this data off a chart of the S&P 500, so it's probably a little inaccurate inasmuch as I didn't take pains to trade the strategy on the EXACT days a crossover occurred.  I am literally too lazy to zoom in on a 25 year chart.  Consider it illustrative.

I used exponential moving averages -- one at 121 days (6 months), one at 252 days (12 months).  The strategy is insanely simple:  go long the market when the shorter moving average is above the longer moving average, and exit the market (to cash) when the longer moving average crosses over the shorter moving average.  For this model, I see 2 such periods from January 1995 until now -- 12/12/2000 through 8/8/2003, and 2/7/2008 through 11/3/2009.  (There is what appears to be a very short exit near the end of 2015 -- I ignored it, which in all likelihood means I didn't account for a small loss.  Sue me.  The right way to do this is to download daily price history to Excel, program the moving averages and the trade rules, and see where you are at the end.  I just don't have it in me to do that right at the moment.)  I selected the start date on a point when the crossover occurred, and I went far enough back in time to get some bear markets included.  Had I just started at the beginning of this latest run since the financial meltdown, buy and hold would come out slightly ahead.  Bear markets happen, though, which is where this kind of approach makes hay.

The results?  Over the last 23.8 years (Jan 1995 until now), this strategy would return ~ 10.5% annualized.  A simple buy and hold strategy would return ~7.8%.  That might not sound like a huge difference, but $100K invested in each strategy at the start would currently be $603K (buy and hold) versus $1.08M (moving average crossover strategy).  If one were to actually short the market instead of going to cash when the signals occur, then this difference would be even greater. 

Of course, many people aren't interested in actively managing their accounts, and I understand that.  One should also consider tax implications if trading in a taxable account.

[Tailgate 3021]

See how well you do here instead of losing your real money.

https://www.portfoliovisualizer.com/test-market-timing-model

[Wally Fairway 5541]

22 hours ago, jimmyjazz said:

Maybe we're working under different definitions of "timing the market", but in my view it's as simple as going to cash (or short) when conditions warrant.

At any rate, it's hardly a great secret:  a simple moving average crossover strategy historically beats the market.  I pulled this data off a chart of the S&P 500, so it's probably a little inaccurate inasmuch as I didn't take pains to trade the strategy on the EXACT days a crossover occurred.  I am literally too lazy to zoom in on a 25 year chart.  Consider it illustrative.

I used exponential moving averages -- one at 121 days (6 months), one at 252 days (12 months).  The strategy is insanely simple:  go long the market when the shorter moving average is above the longer moving average, and exit the market (to cash) when the longer moving average crosses over the shorter moving average.  For this model, I see 2 such periods from January 1995 until now -- 12/12/2000 through 8/8/2003, and 2/7/2008 through 11/3/2009.  (There is what appears to be a very short exit near the end of 2015 -- I ignored it, which in all likelihood means I didn't account for a small loss.  Sue me.  The right way to do this is to download daily price history to Excel, program the moving averages and the trade rules, and see where you are at the end.  I just don't have it in me to do that right at the moment.)  I selected the start date on a point when the crossover occurred, and I went far enough back in time to get some bear markets included.  Had I just started at the beginning of this latest run since the financial meltdown, buy and hold would come out slightly ahead.  Bear markets happen, though, which is where this kind of approach makes hay.

The results?  Over the last 23.8 years (Jan 1995 until now), this strategy would return ~ 10.5% annualized.  A simple buy and hold strategy would return ~7.8%.  That might not sound like a huge difference, but $100K invested in each strategy at the start would currently be $603K (buy and hold) versus $1.08M (moving average crossover strategy).  If one were to actually short the market instead of going to cash when the signals occur, then this difference would be even greater. 

Of course, many people aren't interested in actively managing their accounts, and I understand that.  One should also consider tax implications if trading in a taxable account.

Thanks -

One of the things that just looking at the returns ignores is the stress from holding during a market correction (and the stress from being out of the market when it turns positive). Investing is certainly a financial decision, but when it is your own money and not that of a professional money manager, the stress becomes an increasing portion of the decision. So factor in the cost of using a money manager to alleviate the stress vs. savings the investment fee and riding it out yourself.

I held through market up & downs when I was younger ('87, dot.com, 2008/09), with the idea that I can wait out the turnaround better than I can time the correction. However with age,  I will need to start to consider making withdrawals within the next 10 years, the emotional/stress component weighs heavier on these decisions.
I'm getting closer to making the Sharktank decision - I'm needing some of this money sooner rather than later, and therefore "I'm out'

[troph 24781]

Maybe we're working under different definitions of "timing the market", but in my view it's as simple as going to cash (or short) when conditions warrant.
At any rate, it's hardly a great secret:  a simple moving average crossover strategy historically beats the market.  I pulled this data off a chart of the S&P 500, so it's probably a little inaccurate inasmuch as I didn't take pains to trade the strategy on the EXACT days a crossover occurred.  I am literally too lazy to zoom in on a 25 year chart.  Consider it illustrative.
I used exponential moving averages -- one at 121 days (6 months), one at 252 days (12 months).  The strategy is insanely simple:  go long the market when the shorter moving average is above the longer moving average, and exit the market (to cash) when the longer moving average crosses over the shorter moving average.  For this model, I see 2 such periods from January 1995 until now -- 12/12/2000 through 8/8/2003, and 2/7/2008 through 11/3/2009.  (There is what appears to be a very short exit near the end of 2015 -- I ignored it, which in all likelihood means I didn't account for a small loss.  Sue me.  The right way to do this is to download daily price history to Excel, program the moving averages and the trade rules, and see where you are at the end.  I just don't have it in me to do that right at the moment.)  I selected the start date on a point when the crossover occurred, and I went far enough back in time to get some bear markets included.  Had I just started at the beginning of this latest run since the financial meltdown, buy and hold would come out slightly ahead.  Bear markets happen, though, which is where this kind of approach makes hay.
The results?  Over the last 23.8 years (Jan 1995 until now), this strategy would return ~ 10.5% annualized.  A simple buy and hold strategy would return ~7.8%.  That might not sound like a huge difference, but $100K invested in each strategy at the start would currently be $603K (buy and hold) versus $1.08M (moving average crossover strategy).  If one were to actually short the market instead of going to cash when the signals occur, then this difference would be even greater. 
Of course, many people aren't interested in actively managing their accounts, and I understand that.  One should also consider tax implications if trading in a taxable account.

I played around with this several times and depending on how you structure the chart the lines cross over at various places. 1 year, 5 year, 180 days, etc. seemed pretty difficult to actually manage at least for me.

[LurkingHorn 1066]

23 hours ago, jimmyjazz said:

Maybe we're working under different definitions of "timing the market", but in my view it's as simple as going to cash (or short) when conditions warrant.

At any rate, it's hardly a great secret:  a simple moving average crossover strategy historically beats the market.  I pulled this data off a chart of the S&P 500, so it's probably a little inaccurate inasmuch as I didn't take pains to trade the strategy on the EXACT days a crossover occurred.  I am literally too lazy to zoom in on a 25 year chart.  Consider it illustrative.

I used exponential moving averages -- one at 121 days (6 months), one at 252 days (12 months).  The strategy is insanely simple:  go long the market when the shorter moving average is above the longer moving average, and exit the market (to cash) when the longer moving average crosses over the shorter moving average.  For this model, I see 2 such periods from January 1995 until now -- 12/12/2000 through 8/8/2003, and 2/7/2008 through 11/3/2009.  (There is what appears to be a very short exit near the end of 2015 -- I ignored it, which in all likelihood means I didn't account for a small loss.  Sue me.  The right way to do this is to download daily price history to Excel, program the moving averages and the trade rules, and see where you are at the end.  I just don't have it in me to do that right at the moment.)  I selected the start date on a point when the crossover occurred, and I went far enough back in time to get some bear markets included.  Had I just started at the beginning of this latest run since the financial meltdown, buy and hold would come out slightly ahead.  Bear markets happen, though, which is where this kind of approach makes hay.

The results?  Over the last 23.8 years (Jan 1995 until now), this strategy would return ~ 10.5% annualized.  A simple buy and hold strategy would return ~7.8%.  That might not sound like a huge difference, but $100K invested in each strategy at the start would currently be $603K (buy and hold) versus $1.08M (moving average crossover strategy).  If one were to actually short the market instead of going to cash when the signals occur, then this difference would be even greater. 

Of course, many people aren't interested in actively managing their accounts, and I understand that.  One should also consider tax implications if trading in a taxable account.

 

Just out of curiosity, how did you arrive at those lengths of time for your moving averages? Is the thought being there are more data points so one wouldn't be as quick to overreact in the event of a few down days?

[troph 24781]

 
Just out of curiosity, how did you arrive at those lengths of time for your moving averages? Is the thought being there are more data points so one wouldn't be as quick to overreact in the event of a few down days?

Same question I had. Also is it SP500 ETF or index fund? What about the 2x/3x funds? What about when it’s time to pull out? Cash or inverse funds? 1x or an amplifier?

Curious minds want to know...

[52-80 19147]

11 minutes ago, LurkingHorn said:

 

Just out of curiosity, how did you arrive at those lengths of time for your moving averages? Is the thought being there are more data points so one wouldn't be as quick to overreact in the event of a few down days?

the time-tested academic method of tweaking shit until it fits a pre-constructed narrative.

[Trey3216 15345]

14 minutes ago, LurkingHorn said:

 

Just out of curiosity, how did you arrive at those lengths of time for your moving averages? Is the thought being there are more data points so one wouldn't be as quick to overreact in the event of a few down days?

Those lengths of time better reflect an investor's time horizon rather than a trader's/momo investor's.  

[jimmyjazz 45892]

27 minutes ago, 52-80 said:

the time-tested academic method of tweaking shit until it fits a pre-constructed narrative.

Well, you're exactly 100% wrong.  It was the first pair I tried.  I was trying to "think like a long-term investor" and show what types of downturns could be largely avoided without subjecting one's self to a lot of high frequency chop.  There are studies out there that mimic these results over longer timeframes with different moving averages.

Is past data an ironclad guarantee that one will beat the market going forward?  Of course not.  Is it worth considering given the significant excess return?  Of course it is.

Regarding the instrument -- choose whatever you want.  This was the index.  SPY would probably be a good proxy.  Historical SPY price data probably has dividend payouts rolled in, which I didn't want to mess with (if you're not in the market you're not getting dividends).  Then again, the "model" I showed also had one go to cash in downturns, and there are almost surely better interest-bearing opportunities available to help provide a meager (but non-zero) return during those times.  I'm thinking T-Bills, etc.  I'm not a bond investor, I literally know almost nothing about them.

This was merely an exercise to demonstrate that one can "time the market", at least to some degree.  Take it or leave it.

[troph 24781]

My questions were serious not CR argumentative. What time variable do you use?

[jimmyjazz 45892]

6 minutes ago, troph said:

My questions were serious not CR argumentative. What time variable do you use?

I'm not sure I understand the question -- I don't trade this strategy.  I trade on a near daily basis.  The price data for this exercise were daily closing prices.

I mentioned in the original post that the model assumes one goes to cash when the moving averages cross over.  Had one gone short (or more conservatively, bought puts or inverse funds) one would do even better for this particular set of price data.  The possibilities are endless, and no amount of historical modeling will predict future returns.  It's at least an indication of what is possible.

[Firemans4Horn 18036]

Just a fancy way of saying “hindsight is 20/20”. I’ve been pitched dozens of back tested algorithms over the years and they all had 3 things in common:

1. They worked perfectly at buying low and selling high and avoiding whatever black swan events that occurred

2. They never had any money invested in the strategy during the time periods they reference

3. They don’t mean shit going forward

 

 

[jimmyjazz 45892]

LOL.  It sure as shit meant something going forward in January 1995 (for instance).

[Anastasis 23463]

I don’t know about slk talks fancy talk but I’m within about 2% of capitulating.

[jimmyjazz 45892]

. . . and we're about 2% from a "correction".

[Trey3216 15345]

4 minutes ago, jimmyjazz said:

. . . and we're about 2% from a "correction".

CNBC will be all over that, all day.  

 

"We're within 170 pts on the Dow from correction territory, meanwhile, we're already in correction territory on the Nasdaq with many names already entering Bear Market Territory!!"

 

 

[Firemans4Horn 18036]

43 minutes ago, jimmyjazz said:

LOL.  It sure as shit meant something going forward in January 1995 (for instance).

 

Lol. All we need is a time machine. 

[Wally Fairway 5541]

1 hour ago, Firemans4Horn said:

 

Lol. All we need is a time machine. 

We have one - it's called "your life", you get to do with it as you wish.
Unfortunately this time machine does not go backwards and every 24 hours it goes forward one day

It is like Steven Wright and his map

Edited October 23, 2018 by Wally Fairway
speeling corectshun

[happyfunball 481]

    According to Credit Suisse, with nearly a quarter of companies reporting results, almost 80 per cent of them have beaten earnings estimates. That kind of overachievement is belied by the sustained fall in equity prices. There is unquestionably a re-pricing of the cost of capital currently under way. When the cost of capital is rising, even growing earnings can lead to lower valuations
 

[jimmyjazz 45892]

"Buy and hold" is a strategy.  It's no more prescient than what I proposed above.  For anyone to presume that their crystal ball is somehow more accurate than anyone else's crystal ball is preposterous, particularly in light of historical data that indicates one might just be wrong.  If we are to be intellectually honest and cling to the mantra of "past performance is not indicative of future success", then we have to do so for ALL models, not just the ones that get taught in sophomore business classes at McCombs.

[Goofyboy 3269]

I’ll stick with the teachings of a Random Walk Down Wall Street. Some people can time and beat the market. Most can’t.

[Chapo 104]

Buy and hold here,great time to do a tax loss harvesting

[Superhero 8828]

2 hours ago, Chapo said:

Buy and hold here,great time to do a tax loss harvesting

I’m generally a buy and hold guy, but have lost a little confidence after watching the market stall the past few months. Other than a treasuries fund, I already sold all of my losers for the year, and started selling the under-performers. 

Nearly all of my money was invested, it’s actually nice to have some cash on the sideline while the market figures out what to do for the future. 

[Blotto 23491]

I had some QQQ puts (mid Nov) as a hedge. Talked myself into adding to that position yesterday at the close. If the market wants to take a shit between now and the midterms, be my guest. If not, I'll be down some money. 

[troph 24781]

I'm not sure I understand the question -- I don't trade this strategy.  I trade on a near daily basis.  The price data for this exercise were daily closing prices.
I mentioned in the original post that the model assumes one goes to cash when the moving averages cross over.  Had one gone short (or more conservatively, bought puts or inverse funds) one would do even better for this particular set of price data.  The possibilities are endless, and no amount of historical modeling will predict future returns.  It's at least an indication of what is possible.

The question is when you formulate the chart what time period do you use? The averages cross at different points based on a chart spanning the last 90 days vs last 5 years. The ups and downs stretch and compress and the cross over lines vary based on this variable.

I had intended to trade this strategy a few years ago for a couple of points bump on over all returns over several decades and I actually believe in it as a modest timing strategy BUT its variable based on the chart inputs. SP500 will likely work fine, the averages are set so the last variable is the chart’s history - 90 days, 180 days, 1 year, 3 year, 5 year, 10 year or 20 year look back?

[jimmyjazz 45892]

15 minutes ago, troph said:

The question is when you formulate the chart what time period do you use? The averages cross at different points based on a chart spanning the last 90 days vs last 5 years. The ups and downs stretch and compress and the cross over lines vary based on this variable.

I had intended to trade this strategy a few years ago for a couple of points bump on over all returns over several decades and I actually believe in it as a modest timing strategy BUT its variable based on the chart inputs. SP500 will likely work fine, the averages are set so the last variable is the chart’s history - 90 days, 180 days, 1 year, 3 year, 5 year, 10 year or 20 year look back?

Well, the chart I showed above starts in January 1995.  Every point on each moving average is the (exponentially-weighted) average of the closing price for the previous 121 days (green line) or 252 days (red line).  These correspond to 6 months of closing prices and 12 months of closing prices, respectively.  (Full disclosure:  I think I should have used 126 days, I must have fat-fingered it.  This won't significantly change the results.)

So, on January 1 1995 (if the market was open) it uses the previous 121 & 252 days of closing price data to calculate the 2 averages.  The next day, the first closing price in each moving average gets dropped and the closing price for January 2 1995 gets added, and the averages are recalculated.  So on and so forth up until the present day.

Edited October 24, 2018 by jimmyjazz

[troph 24781]

2 hours ago, jimmyjazz said:

Well, the chart I showed above starts in January 1995.  Every point on each moving average is the (exponentially-weighted) average of the closing price for the previous 121 days (green line) or 252 days (red line).  These correspond to 6 months of closing prices and 12 months of closing prices, respectively.  (Full disclosure:  I think I should have used 126 days, I must have fat-fingered it.  This won't significantly change the results.)

So, on January 1 1995 (if the market was open) it uses the previous 121 & 252 days of closing price data to calculate the 2 averages.  The next day, the first closing price in each moving average gets dropped and the closing price for January 2 1995 gets added, and the averages are recalculated.  So on and so forth up until the present day.

right but if you change it from 1995 to today to something more like 2013 to 2018 you get a cross over at a different spot.  and that difference gets more exaggerated the closer you get in to the point that it doesn't make any sense.  my point is everyone talks about this strategy as the simpliest fool proof dummy way to beat the buy and hold strategy - but no one ever says and the chart must at least cover a minimum of 25 years or 10 years or 3 years. 

[jimmyjazz 45892]

38 minutes ago, troph said:

right but if you change it from 1995 to today to something more like 2013 to 2018 you get a cross over at a different spot.  and that difference gets more exaggerated the closer you get in to the point that it doesn't make any sense.  my point is everyone talks about this strategy as the simpliest fool proof dummy way to beat the buy and hold strategy - but no one ever says and the chart must at least cover a minimum of 25 years or 10 years or 3 years. 

No, the math doesn't change.  The crossovers are identical.  The strategy uses historical price data.  Those data don't change regardless of when I start the chart.  (Q:  what charting package are you using?) 

Here's an example, I'll run the same 2 moving averages starting in January 2007.  Note the crossovers that trigger movements out of and back into the S&P 500 happen at the exact same dates as on the longer chart I showed above:

 

Edited October 24, 2018 by jimmyjazz

[troph 24781]

Then I’m an idiot bc I’ve run charts and the lines elongate or compress and cross over differently. I’ll stick to buy and hold.