In my 20 years of trading I have noticed this cycle play out again and again with traders that "make it":

1. Overconfidently reach for returns
2. Get humbled by the market
3. Simplify + concentrate on clear, high probability edges.

Nearly everyone starts with a lack of respect for how hard it is to consistently make money trading.

That leads them to pass over high-probability sources of returns in favor of more marginal ideas.

Or they overcomplicate the trading of a good edge.

Here's an example...

It is 2015. You look at a simple strategy.

You hold an equal dollar exposure to:
- Cap Weighted US Stocks (VTI)
- 20+ Yr US Treasury Bonds (TLT)
- Gold (GLD)

And rebalance each month.

(I've extended back a bit with mutual fund prices.)

It made money consistently with a few blips.

You'd have 4x your money from 1997 to 2015.

It makes sense. It's harnessing two big sources of risk premia (equity + duration).

It diversifies across 3 macro assets that tend to outperform in different econ environments

But, naturally, you turn your mind to how you could do better...

And here is where it gets dangerous!

There are good ways to try to "do better", and there are bad ways to try to "do better".

The "good ways" involve trying to maximize the probability this will be profitable *in the future*.

For example:

1. The asset universe is US-centric. Global diversification would reduce concentration risk that the US underperforms in the future.

2. Gold is doing a lot of heavy lifting there - more asset classes would be helpful.

3. The assets have different levels of volatility and their volatility changes over time - managing this so that each asset's contribution is more equal over time would be helpful.

These kinds of things are "good ways" to try to improve things. Things that increase the chances of you making money in an uncertain future

What are the bad ways?

The bad ways *decrease* the probability of you making money in an uncertain future. They add ways to screw up

For example, you might notice that, in the past, assets that tend to have out-performed over the previous 12 months tend to outperform in the short term too.

We call this x-sectional momentum.

So, you, say, "I'm gonna use this effect to improve returns"

Now at the start of each month, instead of holding each of the 3 assets in equal dollar amount, you're going to:
1. Rank each asset by its total return over the past 12 months
2. Hold only the top asset (with the highest returns).

You can see the past performance of this approach in cyan vs the diversified equal weight strategy in red.

It was more volatile but would have out-performed in returns quite significantly...

So, emboldened by the glory of the backtest, you trade the momentum strategy.

How did you do from 2015 - 2021?

Oops... not so hot.

You underperformed the diversified buy and hold strategy quite significantly.

What happened?!!

You started with a very good high-probability edge: Diversified Risk Premia.

Then, in overconfidently reaching for extra returns, you added a much lower probability active bet (x-sectional momentum) to it.

You added a new way to screw up!!

But did we just get unlucky?

Was there ever enough evidence that we should have been overconfidently overlaying that momentum strategy?

Let's look....

Here we:
1. Take monthly observations for each asset
2. Rank each asset each month by their previous 12-month total returns
3. Calculate the difference between the next month log returns of the top and bottom-ranked asset
4. Plot the mean of this difference by year

There's a small hint that there may have been a strong x-sectional momentum effect pre-2000.

But much less since 2000...

But imo, there was really nowhere near enough evidence available to us in 2015 to be making such a drastic active bet on x-sectional momentum.

You have to understand what is driving performance in something like this.

Our edge comes from being exposed to risk premia.

And the smooth performance (vs just buying stocks) comes from the volatility-reducing diversification effects of holding uncorrelated assets.

Cutting ourselves off from those effects with an active timing rule is cutting ourselves off from the things that we *want* to be exposed to. The things that are driving our returns.

It's an overconfident bet on an uncertain effect.

The edge (diversified risk premia) is too good to add the chance of screwing it up.

If you wanted to trade the x-sectional momentum effect, you should have done it *in addition to* the simpler diversified strategy, not *instead of it*.

This way, you'd gently increase weights in the highest momo asset each month, and gently decrease weights in the lowest.

You'd have some exposure to the momentum effect - but you wouldn't be cutting yourself off from the high-probability diversified risk premia effects.


Don't add ways to screw up high-probability good things.

Look to maximize the chance of making money in an uncertain future - rather than reaching for lower probability excess returns.

Keep it simple. Don't try to be a hero.

BTW here's what that x-sectional momo time series plot looks like extended out to 2020...

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Robot James

Robot James Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @therobotjames

20 Apr
In Australia, if you're serious about getting the job done effectively and efficiently, you might say:

"I'm not here to f*** spiders"

Many traders act like they are, indeed, here to f*** spiders.

A thread about getting serious about making money trading 👇👇👇

1/n Image
If you're making soup, you first need a good stock.

Stock isn't exciting. Everyone has stock.

Garnish is exciting, but you can't make soup from just garnish.

You need some stock in your trading portfolio

You need at least one reliable, stonkingly obvious way to get paid

Here's a non-soup analogy...

If you start a business venture, it's clear that you need an obvious, reliable way to make money.

You wouldn't just try to blag it.

"I am smart and hard-working" is not a business case.

You need a stonkingly obvious way to get paid.

Read 30 tweets
14 Apr
A common mistake is to make implementation decisions or parameter choices based on "what improves the summary performance of a backtest".

Quant research is not "changing random stuff and picking the best performing backtest"

A backtest is a very complicated thing.

In the best case, the cause -> effect relationship between what you are changing and the performance of the backtest (say) is highly non-linear.

More likely there is no clear relationship.

Quant trading is not "changing stuff until you get a backtest you're happy with".

You need to split what you're doing into small component chunks and model those chunks as best you can.

Read 5 tweets
11 Apr
It's easy to lose money trading if you:
1. Trade too much (paying fees + impact on each txn)
2. Size positions too big (high vol hurts compounding ability + gets u rekt)
3. Shorting positive drift/risk premia

It's hard to lose money consistently if you avoid these things.
However clueless you are, you get to trade at market prices.

Imagine we can know that an asset has a fair value of $100.

You might think it's worth $150.

But if it's quoted $99 / $101, you can buy now at $101.

You were totally wrong but you still bought close to fair value.
The same mechanisms that make it hard to get an edge also make it hard for you to trade at really bad prices.

In a simple model, you might say that prices are set by:
- (risky) arbitrage and relative value in the short term
- pricing/valuation models in the long term
Read 15 tweets
30 Mar
Allow me to ramble for a bit about how I think about edge in trading. 👇👇👇

First, what we're trying to do is trade deviations from fair value.

We want to repeatedly:
- buy what's cheap
- sell what's expensive
- offset risk as cheaply and efficiently as possible.

We'll concentrate on the first two here.

Let's take a really simple example to start with. Imagine you have the same asset trading on two different exchanges.

Let's pretend it's some altcoin trading on two crypto exchanges (cos I want to look cool.)

Remember we want to be trading deviations from fair value?

Well, I don't have a clue what the fair value of some altcoin should be.

But maybe I don't need to.

I can certainly identify when it might be *relatively* cheap or expensive on each exchange.

Read 16 tweets
21 Mar
Most beginner traders don't realize just how variable the p&l of even a very high-performing trading strategy is.

I simulated 10 5 year GBM processes with annual return 20% / annual vol 10%.

(Simulating a strategy within known Sharpe 2 characteristics.)
I plotted the path with the highest ending equity (green), median (black) and lowest (red).

All paths are from exactly the same process, with the same known return distribution.

You might think of the green line as trading a strategy with a known large edge and being lucky.
You might think of the red line as trading a strategy with a known large edge and being unlucky.

Even when you were really lucky, you were underwater for 130 days.

When you were unlucky, you were underwater for 508 days (about 2 years)
Read 8 tweets
21 Mar
@InBraised If you can trade for free, then your optimal trading strategy (given reasonable return estimates) would be incredibly hyperactive.

You would continuously change portfolio weights according to your latest return estimates.
@InBraised In the real world, this would kill you, because trading frictions would eat away at your PnL.

So, one way to avoid hyperactive rebalancing is to only calculate your return estimates periodically (say once every day, or every week, or something).
@InBraised But this isn't optimal because, if your alpha is good, you want to be calculating it as often as possible. You just only want to be trading when the increase in expected returns from the new position is much better than the old position.
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!