Snake Oil Strategies.
I have been getting a lot of queries about alternative strategies and backtesting – which is good, and keeps me on my toes. I try to check out all the strategies that are suggested. The main complaint is:
Why Not Get Out Sooner?
Most suggestions involve getting out sooner in a downturn so that we preserve more profits. The 10/50 SMA cross has been suggested several times and at first glance it seems like a good idea. Yes, it would have got us out in August when SPY was $445. But what then?
For a strategy you need 3 things:
> A GET IN signal
> A GET OUT signal
> A GET BACK IN signal
So, if we use the 10/200 golden cross to get in, the 10/50 death cross to get out – what do we do then? We can’t use the GET IN signal because that is still in force – we have our GET IN and GET OUT signals happening at the same time. We could use the 10/50 golden cross to get back in – but then we are trading ITM with different parameters. I have run both backtests and here are the results (no leverage):
You can see that the 10/200 has a higher success rate and larger average gains than the 10/50. This not saying that there isn’t a better way, just that this isn’t it. I am actively looking and testing strategies.
A Shocker of a Strategy
I don’t like criticizing people who set themselves up as experts. From experience I know that criticism always stings and it takes about 20 nice comments to be able to get some perspective on one nasty one. But when people have set themselves up as experts by disparaging others then I think they are fair game. This one got up my nose.
But here’s the shocker … all the dummies that put out these books never computer-tested a strategy in their life.
My 100% computer-tested trading strategies use exact and repeatable rules to beat the stock market
Well, this ‘dummy’ decided to check out your ‘100% computer tested strategies’ that are supposed to have beaten the market.
Last week I wasted time on your ‘Reverse MACD’. Then I tried your ‘Swing Trader Strategy’. Here’s your claimed results:
WOW!
82% profitable trades – that’s some record! Better than any other trader results I have ever seen. This guy must be good. Let’s get down to the details of how he does this. This is his strategy:
Seems reasonable at first sight, doesn’t it?
So, I set up an Excel spreadsheet and programmed it in. The results were nothing like his. Two frustrating hours later I did what I should have done in the first place: analyse his strategy.
And just by eyeballing his flow chart with an open mind it became clear what was wrong. Have you worked it out yet?
The Strategy Signals
What is the IN signal? Quite clear, trading above the 66 day SMA and the lowest close in 3 days. Seems reasonable. Trying to buy small dips in an upward trending market. No problem.
What is the OUT signal? Sell when you have the highest close in 19 days. Again, sounds quite reasonable. He is obviously trying to take profits along the way then buy back in again when it dips.
BUT – what if you are in a trade and SPY doesn’t make a new high before the prices retreat?
If you are following the above strategy, you are going to have to ride the bear all the way to the bottom.
How would this work in practice?
Let’s check out what would have happened in the GFC. IN the following graphs the blue down arrows are the buy signals and the green up arrows the sell.
You can see that the strategy would have kept us in for the duration of the GFC, riding the from the top of around 160 to getting out at 90. Not good.
Why doesn’t it work?
At first glance, the strategy sounds reasonable, even sensible, but you have to always look at the ‘universe of possibilities’ and see if the strategy covers them. All of them. This strategy clearly doesn’t.
How he can claim to get the results that he is claiming I don’t know. (Have deleted the next sentence, it was very uncharitable.)
OK back to the market.
The Last Gulf War
Last week was horrible on the markets, and much, much worse for the people caught up in this terrible situation. Everyone is on edge in case things escalate into all-out war in the Middle East.
Let’s look at what happened on the market the last time, in the second gulf war which was declared on March 20, 2003.
The market was in a downturn when 9/11 happened, and then it continued down for the next 18 months, all the while the world was on edge knowing that things were escalating in Iraq. Then almost to the day, as soon as war was declared the market started to climb and climbed for the next 4 years, almost doubling in that time.
Lessons from this? Just that sometimes markets don’t react the way you think they will.
SPY Chart
Wednesday, Thursday and Friday were terrible. Just when it looked like SPY was going to break through resistance, it tanked 3%. Not nice. We’re not yet at correction level ($412) and well away from a bear ($366) but the 10SMA is pointing down again and it is possible we will have a death cross this week.
On the other hand, there is support at $420 which hasn’t been broken so maybe this will turn out to be a consolidation phase. As with all things in the stock market, we won’t know for sure until afterwards.
SPYG Chart
Similar to SPY, it is just sitting on support. It has been in consolidation since early June, oscillating between $59 and $63. The SMA has been holding up better than SPY, but has now turned down again and could be heading for a death cross.
QQQ Chart
What about the MACD?
For all of them (SPY, SPYG and QQQ) the MACD histogram has turned negative. Which is traditionally a bad sign. However, I have been backtesting the MACD, and the results are quite terrible.
Over the last 30 years, if you had started with $10K and traded the 12/26/9 MACD you would now have $31,296. Nice. It has tripled. BUT – if you had just left is in the market it would be worth $97,199. So much for the MACD being a great way to trade!
VIX Chart
The VIX has ticked up, and is now at 20.73 so we are now out of the low volatility period.
ITMeter
The week ahead . .
The futures look slightly positive right now but there is still around 10 hours to go before market open. Like you, I am sure, I am completely fed up with the market right now, but it is there, and we have to deal with what is there not what we would like to be there. Although that would be much nicer!
Heather
Questions and Comments
As always, happy to answer questions and receive comments (just be kind!)
25 Responses
Hi Heather,
Thank you for your weekly thoughts of the stock market. I find it very helpful and your weekly blog keeps me invested in the stock market. Reading through your back testing of your ITM investing and trading strategy, I realize that we can grow our money of $10k to $10m in 50 years. That is like 15% annual return compared to the annual return of 10% of the SP500 Index.
Please see my estimated guess below: 1990 $10k 1995 $20k 2000 $40k 2005 $80k 2010 $160k 2015 $320k 2020 $640k 2025 $1.25m 2030 $2.5m 2035 $5.0m 2040 $10m
Sincerely,
George Halongton
Hey George – thank you!
I am just about to start a new round of backtesting more strategies, and using data going back 100 years. I really hope that your predictions are right, just remember it isn’t a monotically-increasing function, like interest in a bank. Some years are more, some less and this has an effect on your account balance.
After last night, with the Nasdaq down 2.4% I am feeling particularly bruised this morning. Hope tonight is better!
h
Hi Heather,
Thank you for your reply. I appreciate it. The PE ratio of QQQ and SPY are high in the low 30s and low 20s, respectively. Both of them need to come down because the stock market cannot go up forever. The death cross of 10sma below 200sma needs to happen.
Sincerely,
George Halongton in Los Angeles
Hi Heather, I would like to contribute to the discussion,
some people think this is the best timing:
We sell the SPY when it drops 20% from its all-time high , as we enter a bear market.
We buy the SPY after it closes 18 consecutive days above its 200-day SMA and hold it until the next bear market.
Greetings
Davide
Hey Davide – thank you for this let me test it out tomorrow.
h
Davide,
Sounds like you would also need a ‘get back OUT’ signal.
For example, spy goes down 20%, you’re OUT…on the 19th consecutive day you’re IN. (But let’s say you’re still below 20%) and it continues to go lower…you’re already still in bear territory so there is no ‘get out on the NEXT bear’ if you haven’t left the first one. Without a ‘get back OUT’ signal you would just be holding while it continues to go lower and lower.
Thoughts?
Allan M.
Hi Allan, you’re right but the advantage of this system is statistical and
that it doesn’t generate false signals like the various SMA crosses do,
18 consecutive closes above the 200 SMA in a bear market it’s a bull market signal based on statistics, the S&P 500 is up 100% of the time, 12 months after the signal is triggered.
The last time this signal was triggered was the 15 of February of 2023,
we’ll see if it works this time too.
Davide
One sma for entries, another for exits. Why is that a problem? Why do both entry and exit sma have to be the same? They don’t.
Rules:
1. Enter when 10/200 sma crosses above with space.
2. Exit when 10/50 sma crosses below with space.
3. Re-enter if 10/50 sma crosses above with space.
I really don’t think what I’m suggesting is “snake oil”.
Michael
Hi Michael
I’m starting to think that there is something that I am not seeing – so many people are making the same point. So feel free – anyone – to point out the flaw in my argument.
If we take your rules:
1. enter on the 10/200 golden cross – OK, same as ITM
2. Exit on 10/50 death cross – yes, that will be closer to the current price and limit our losses – no problem there.
3. Re enter on the 10/50 golden cross – doesn’t that mean we are now trading ITM but using the 10/50 crosses instead? We will never use the 10/200 cross again unless we are out after a 10/200 death cross ?
So just checking I have it right – while we are in a bull trend we use the 10/200 to get in, and thereafter use the 10/50 for in & out, until there is a 10/200 death cross and then we rely on the 10/200 to get back in again?
I have seen your other suggestions – thank you – and I will look at them in more detail and backtest if possible.
But if you can confirm my understanding of your strategy then I can backtest that relatively easily with minor changes – I think!
thanks
h
Hi Michael
I’m starting to think that there is something that I am not seeing – so many people are making the same point. So feel free – anyone – to point out the flaw in my argument.
If we take your rules:
1. enter on the 10/200 golden cross – OK, same as ITM
2. Exit on 10/50 death cross – yes, that will be closer to the current price and limit our losses – no problem there.
3. Re enter on the 10/50 golden cross – doesn’t that mean we are now trading ITM but using the 10/50 crosses instead? We will never use the 10/200 cross again unless we are out after a 10/200 death cross ?
So just checking I have it right – while we are in a bull trend we use the 10/200 to get in, and thereafter use the 10/50 for in & out, until there is a 10/200 death cross and then we rely on the 10/200 to get back in again?
I have seen your other suggestions – thank you – and I will look at them in more detail and backtest if possible.
But if you can confirm my understanding of your strategy then I can backtest that relatively easily with minor changes – I think!
thanks
h
Hi,
You hit the nail on the head. It is very hard to comment in a constructive way without appearing to criticize. I asked about Walk Forward because although not foolproof it is considered the gold standard for back-testing. The parameters you use do test out with the SPY in the time period stated. However, once you increase the data set there are issues. That is why I mentioned the 1987 crash. A 200ma got you out but a 200×10 did not. That was a major drawdown which would need to be rolled forward. I encourage you to use the SPX where there is much more data. Also, I would respectfully recommend the Stocharts.com articles on PPO vs MACD. I humbly suggest that you move to the PPO for the reasons they explain. Thanks for all you do!
HI Sherman,
OK, I have just downloaded the SPX data back to 1928, and tomorrow will run it through the model.
Re the PPO – Its rather funny, I already had figured that the weakness of MACD was the fact that is was unbounded, making it hard to compare different period with different underlysing prices. I had worked out my own indicator – I called it the ‘Normalised MACD’ – which is almost exactly the same as the PPO. Now I see someone what beaten me to it, its a thing.
Dammit.
Thought I was being smart and original.
Humph!
OK, getting over my bruised ego, will report back on the results tomorrow.
h
Hi Heather,
I bought your book a while back and reviewed it but couldn’t locate where it is. I do remember some of the mechanics but don’t remember exactly all 3 signals you just mentioned in the blog.
For a strategy you need 3 things:
> A GET IN signal
> A GET OUT signal
> A GET BACK IN signal
Can you let me know what the current rules you follow for these 3 signals?
Hi Ilan
the ITM strategy just uses the 10/200 SMA golden and death crosses.
h
Finally, do you favor spreads of whatever variety using DITM LEAPS, and if so, which have worked for you?
I have never found that using spreads on DITM SPY has been profitable, mainly due to the low time value and tight spreads. It all depends on what premium you can get.
For example, if you look a 1 year to expiry $200 strike the bid / ask is 225.96 / 229.71, and the $210 strike is $216.61 / 220.35. If you buy at the ask and sell at the bid this will cost you $13.10 ($1,310) which means you are always making a loss of $310, no matter where SPY is because you have bought a spread of $10 for $13.10.
If you can manage to get a spread of $10 for less than $10 you will make a profit. If you were able to buy at the mid point of both bid / asks then it will cost $9.40 and you will make a profit of $60 if SPY is above $209.40 at expiry.
For example, if you look a 1 month to expiry $210 strike (there is no $200 strike) the bid / ask is 211.74 / 212.59, and the $220 strike is $201.79 / 202.63. If you buy at the ask and sell at the bid this will cost you $10.80 ($1,080) which means you are always making a loss of $80, no matter where SPY is because you have bought a spread of $10 for $10.80.
If you could buy at both mid points then you would be paying $9.96 for a $10 spread, which will cost you $996 and gives you a maximum profit of $4.00, or 0.4%. Although that is pretty well guaranteed, SPY is unlikely to drop below $220 in the next month!
Where spreads become profitable is OTM and near dated. I think I may cover this in a blog article as this answer is already getting too long!
Hope this helps
h
Stop Losses on option positions: In one of your books I think I read that one should not take more than a 10 % loss on a position. I think this must be an error on one of our parts because even DITM 2-year options can touch beyond ten percent changes on any given day.
Hi there
I am pretty sure that I never recommended it – it is not something I do or would suggest that others do. I may have mentioned it when discussing stop losses or strategies involving stop losses, but it is definitely not part of the ITM strategy. In fact, I will post the diagram I used to show the problem with stop losses at the bottom of the comments.
So, no I don’t recommend it – I try to use symmetrical indicators – as in the same indicator for IN and OUT. If you use differe0t ones – say a golden cross to get in, then a 10% stop loss to get out – that’s fine for 1 trade, but what then? What do you do after you get out? It is quite possible that you have one signal sayng one thing the other saying another, so which one do you act on? If it is a subjective decision then it is impossible to back test accurately.
Our GET OUT signal – the death cross is effectively acting as a stop loss. As the price increases so does the 200 SMA – the SMAs both move up with the prices so you are not getting out at the price you got in.
Hope this helps
h
How do you decide whether it is too late to take a long position after a golden cross has climbed a good way above market price?
Hi, I am always mindful of Jesse Livermore’s quote (he was the original Wolf of Wall St) : “a stock is never too high to buy, or too low to short”
I can’t advise you what to do, just tell you what I do – which is that when I have decided to get into a trade I get in. No waiting for the market to drop to what you think is ‘reasonable’, because in my experience it never seems to (or no for me anyway). But, as is the way of the market, once I am in it often drops to where I wanted it afterwards! We just have to accept that the only traders who get perfect trades are those who do it in hindsight.
But to address your question: it is hard to quantify what is a ‘good way’ above your buying price – is that 10% 20% 30%? It depends partly on the length of your trade. If you are a short term trader then it might work, as it forces you to take profits. But if you are doing long term trades (and ITM has averaged just over one per year since 1993) then you could be missing out on a large part of your profits by sitting on the sidelines.
In short, I don’t know. It is something I am playing around with in models, seeing if there is a good way to do it – haven’t found it yet.
So I suggest that if the golden cross is in force then get in if the market is trading ‘normally’. If there has been a sudden move in either direction of some magnitude (say 2% on one day) then maybe waiting a couple of days for things to settle down would be wise.
Sorry can’t be more definitive – still working on new strategies.
Hope this helps
h
Heather,
“So much for using MACD as a way to trade”
Considering that the MACD is also used in the ITMB, would your comments suggest having another look at ITMB strategy? (I understand that its in conjunction with other signals) or are you just suggesting that the macd /alone/ is not a great way to trade?
Thanks,
Allan M.
———
Hi Allan, I was testing the traditional way to trade the MACD, and found it wanting.
That isn’t to say the MACD isn’t useful.
It is basically 2 EMAs, and a third derived from these 2. Moving averages are mathematical (objective) indicators which may or may not work. In this case, using the crosover for trading signals, which is the traditional way to trade it, doesn’t work very well at all.
In the Bear strategy I use a slightly different measure, which is that the magnitude of the space between 2 of the EMAs is over a certain level, and in conjunction with other SMA signals.
Re revisiting tITMB – it works, and bear markets are few and far between so I prefer to concentrate on bull strategies which is what I am working on now.
Hope this helps.
h
Heather,
I’ve also found that many market “gurus” tend to cherry-pick the timeframes where their particular strategy outperforms. So, I truly appreciate how you’ve displayed the entire history of your trades in Appendix 1 and Appendix 2 of the Bear Market Strategy book.
In my own backtests using TradingView, I’ve found the 50/200 SMA is the only strategy that beats buy-and-hold. However, my analysis didn’t account for the leverage used in ITM and is based only on buying and selling SPY.
What are your thoughts about the traditional 50/200 SMA? I know that the main weakness is volatility/drawdowns and that the signal completely missed the Covid crash.
In my view, the trade-off is higher volatility but fewer buy/sell signals, reducing whipsawing.
Hi Ankur – just signing off for the day – will run it through my models tomorrow!
h
Hi Ankur
I ran it through the model, and it definitely beat the market. Here i’ve put the results at the bottom of the comments (you cant put an image in a comment, unfortunately)
You can see that there are significantly less trades, a higher percentage of winning trades, but the ratio of $win to $loss not so good.
AS you point out, that is without leverage so if that was added at double leverage then in the ITM method wins by a lot.
Hope that helps
h