Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
SqueezeMetrics Profile picture
SqueezeMetrics
@SqueezeMetrics
Dark pools, options, and volatility.
27 subscribers
Mar 7, 2024 • 6 tweets • 2 min read
Gold, the original meme coin, is breaching all-time-highs. Since there's no real reason to ever buy or sell gold, positioning and flows are an important part of what drives price. Image One particular data point has proven to be very useful through time—and it's fun because it's illustrative of the underlying reality.

From the CFTC CoT report, "Swap Dealer Long" positions: Image
Feb 12, 2024 • 5 tweets • 2 min read
What follows is some information that tends to induce people to make bad decisions. So, disclaimer: Don't make bad decisions.

There's some unique positioning in S&P options right now.

Dealer SPX vanna exposure (VEX) is at all-time highs. Specifically, VEX is around $500mm.

That means that when VIX rises, there's an unusually high mechanical bid for the S&P 500, which reinforces an "auto-dip-buying" mechanism. It's very supportive. Image
Dec 9, 2023 • 7 tweets • 2 min read
People are still saying stupid things about 0DTE options, so let's dredge up an old note to point people in the right direction.

Non-blue-check, non-revenue-sharing, character-limited free-tier Xitter thread below. Balderdash. Image
Aug 9, 2023 • 4 tweets • 1 min read
Predicting inflation has little to do with money supply or the velocity of money or the Fed balance sheet or whatever.

Inflation is a price (a meta-price) in a betting market like any other betting market, and it responds chiefly to the bets that people have taken on it. When people are afraid of a price(s) going up, they hedge that with some kind of swap or option or future, and when a lot of people hedge, the thing that they're hedging against won't happen. Because betting markets are reflexive.
Jun 11, 2023 • 4 tweets • 1 min read
Let's talk about this:

JPM's notoriously huge collar strategy got attacked at its last roll. Its next roll happens at the end of this month. The strategy's short call strike is 4320.

What happens this time?

Commentary from April 2nd. I see, I see.

+3.00%
May 17, 2023 • 4 tweets • 1 min read
Today in the exciting world of quantitative volatility analysis: If you are backtesting a volatility strategy or calibrating your vol-time, remember that the Juneteenth national holiday was signed into law on June 17th of 2021, but did not result in a National Market System closure until June of 2022 and that future market closures will [partly] follow the July 4th Independence Day convention which means that when the holiday falls on a weekday the closure will occur on that weekday and if the holiday falls on a weekend the
Feb 15, 2023 • 4 tweets • 1 min read
The Dark Index (DIX) has never been this high for this long.

Liquidity is excellent. Dealers are happily supplying S&P to customers, who are buying hand over fist. The CFTC CoT (TFF) data covering E-minis has been delayed for many weeks due to a "cyber incident," but when they finally release data from after January 24th, you'll likely note that Dealer Intermediaries have taken on a large short position in the intervening time.
Jan 27, 2023 • 6 tweets • 2 min read
When you

1. segment a stock's time-series into a "price" (up, down) and "volatility" (rising, falling) component, and then
2. plot each of those on an axis, and then
3. color the coordinates according to forward returns at each of the [x, y] coordinates... Image ...you learn a lot about the way a stock trades that you wouldn't be able to learn by simply looking at a chart.

The above stock, for example, responds negatively to increases in volatility (↑), and positively to decreases on volatility (↓).

Seems useful to know.
Jan 25, 2023 • 4 tweets • 1 min read
Higher S&P implied volatility (VIX) relative to S&P realized volatility means worse intraday S&P performance. Image And vice versa: Lower VIX relative to higher RV means better intraday performance for SPX.
Jan 19, 2023 • 4 tweets • 1 min read
At current VIX (~20), the customer vanna-gamma ratio (the relative impact of vanna vs. gamma on existing customer option positions) flips to positive around SPX 3850.

That's where the deltas of existing sold put positions start going up really quickly, putting pressure on folks. Realistically, though, VIX probably scoots closer to 22 before that happens, which means the vanna of those sold puts starts spiking deltas a bit before SPX reaches 3850. Maybe 3875.
May 5, 2022 • 4 tweets • 3 min read
@AlexLachance11 @stephenharlinmd For the reason that large "dark pool" volume in ETFs likely relate to share creation/redemption, yes, I think so. @AlexLachance11 @stephenharlinmd Specifically in the case of QQQ, I recall that low DPI is actually a more interesting condition, and I suspect that's because low DPI is indicative of low intermediation, and low intermediation is often exciting.

What *kind* of exciting depends on circumstances.
Jan 25, 2022 • 12 tweets • 4 min read
A bit of color:

DIX printed 37.2% because retail got rug-pulled today.

dix.sqzme.co The Dark Index (DIX) measures off-exchange short sales. More short sales -> higher DIX.

Usually, more short sales is more bullish, because it means money managers are sending buy orders to ATSs (passive midpoint orders), or passive agency orders to BDs, and getting filled.
Jan 20, 2022 • 13 tweets • 5 min read
@ChrisClemonz The "lit/dark" distinction isn't really the right way to frame it, though it's often easier to describe it that way. Rather, there's a sliding scale of "urgency," or "time preference," that's implicitly tied to any given order. @ChrisClemonz The practical means through which this is expressed in the market is in the type of order that's being used. A resting limit is low time-preference. A midpoint order is medium. A bid/ask-hitting order on a lit exchange is high time-preference.
Jan 18, 2022 • 4 tweets • 1 min read
DIX 45.5% 45.1%

They're buying the dip again.
Oct 11, 2021 • 5 tweets • 2 min read
DIX 44.9% into that dip. -> dix.sqzme.co

Experienced DIX practitioners know that this is not only strong, but seasonally strong for a Monday. Image
Sep 8, 2021 • 6 tweets • 1 min read
SPX down 20 points, VIX up 3 points.

And more to go. SPX now down 40, VIX flat.

And more to go.
Sep 2, 2021 • 6 tweets • 2 min read
We talked a lot about the Vanna-Gamma Ratio (VGR) yesterday, and how it tells us about the fragility of customer positioning in SPX.

But it didn't show us the whole picture.

Today, let's add the last piece to our discussion of "non-dealer gamma." The missing piece is Net Put Delta (NPD). Image
Sep 1, 2021 • 7 tweets • 2 min read
What are "bad times?" A "bad time" is when everyone is overexposed to the same thing. This makes the market fragile.

As you know, an option's delta is not only sensitive to spot price (gamma), but also to volatility (vanna). We've called vanna "gamma's evil twin." She's sneaky.
Aug 30, 2021 • 5 tweets • 1 min read
Moving on...

What is "non-dealer gamma?" It's the other 50%.
Aug 27, 2021 • 7 tweets • 2 min read
You are completely certain that SPY will not close below $440 tomorrow (Friday).

SPY is $446.26.

You spend every last penny of your life savings, and your brokerage margin, on Friday-expiring... A. You sink your life savings into the 440 calls. SPY closes at 440.00. You have lost everything.
Aug 24, 2021 • 4 tweets • 1 min read
S&P 500 (SPX) option dealers hold more and more inventory.

"But what about the other 50%?" you wonder. Normally, this would be an opportunity for us to plug gamma exposure (GEX).

"Dealer gamma is only getting more important!"

But that's just not true anymore. As dealer inventory gets bigger, and as people model it with increasing accuracy, it actually becomes *less* important.