Here's Nomura's Mcelligott: Trader Who Correctly Called Tuesday's Crash Lays Out What Happens Next
Unbeknownst to moi every bit I was typing "Quant blame game over stock sell-off pits Nomura against Nomura":
This is a placeholder until ZeroHedge gets or then to posting Charlie McElligott's latest on the CTA crowd....
ZH had already posted but no 1 told me.
Here it is from ZeroHedge:
I induce got to become induce got a discussion alongside some immature people, dorsum inwards a bit.
This is a placeholder until ZeroHedge gets or then to posting Charlie McElligott's latest on the CTA crowd....
ZH had already posted but no 1 told me.
Here it is from ZeroHedge:
Having correctly predicted the "breaking point" trigger for the S&P's Tuesday plunge - which incidentally coincided alongside the 200DMA - at which CTAs would collectively puke, Nomura's caput cross-asset quant Charlie McEllgiott, is dorsum alongside a banknote slamming his critics (apparently those who failed to last right decided to deflect their lack of added value past times criticizing the Nomura strategist), spell also mapping out what's adjacent for the marketplace should the electrical flow run a peril waterfall continue....MORE
First, nosotros become right to McElligott's well-deserved victory lap, inwards which he steamrolls his "quant" critics, as well as writes that he received a release of pass-along notes yesterday from or then the Street "which questioned the validity accuracy of our CTA model ‘deleveraging’ telephone telephone from Tuesday, which in 1 trial again “nailed” the S&P futures degree where the marketplace would come upward nether pregnant notional selling pull per unit of measurement area (as good every bit Russell as well as Nasdaq “trigger” levels every bit well)."
What he is proverb is simple: Wall Street is perplexed past times how he could larn it then right, as well as all the other "experts" did not, as well as is accusing the Nomura quant of a lucky one-time fluke. Needess to say, McElligott volition induce got none of it, as well as highlights that every bit business office of the sellside criticism of his take, at that spot were "mis-categorizations/inaccuracies" on a release of fronts which remove to last highlighted:
Not satisfied alongside the evisceration of his well-nigh song critics, the human who is apace emerging every bit the truthful quant "Gandalf" writes that inwards calorie-free of "the incredible business of prognosticating 58 unique cross-asset futures contracts every bit our QIS team’s model does—and inherent requirements of merchandise direction / sizing / leverage of course—the index replication model is brilliantly accurate on performance-matching vs the index", to wit:
- Misinformation every bit to the notional size of the selling which nosotros estimated vs what was reported
- Inaccuracies of the AUM scale of the CTA universe as well as seat sizing / leverage resources allotment therein
- Whether this “trigger” was a “deleveraging” of a long (size reduction of the long which it was) vs outright “shorting”(which was misreported / misinterpreted past times some)
- Omissions / lack-of-context surrounding macro- as well as positioning- / performance- catalysts which I’ve been documenting inwards recent notes that genuinely “kicked off” the motility to said trigger levels
- A full general lack-of-awareness every bit to the make-up of the model--i.e. that nosotros contain 2w, 1m, 3m, 6m as well as 12m windows to capture the broad-spectrum of CTA lookback periods across the tendency universe
- No context every bit to the incredible accuracy of the model—not only via the success that the tool has had inwards identifying “market inflections,” but alongside regards to our CTA replication model’s incredible track-record vs benchmark
And his crushing parting words at his desperate-for-publicity-and-page-views critics:...
- Since 2016, the model has exhibited an average departure from benchmark of only 54bps, alongside a median departure from benchmark of 48bps
- Over the incredibly volatile terminal 6m both from a cross-asset realized volatility- as well as CTA performance- perspective, the average departure of the replication model from benchmark is only 56bps, alongside the median departure from benchmark beingness 58bps
I induce got to become induce got a discussion alongside some immature people, dorsum inwards a bit.
No comments