Thread by @0xdoug

1/ Great, thread. But I strongly disagree with this sub-point. The vast majority of HFT MM alpha realizes in under 1 second post-fill. Even to the extent that retail flow is toxic cuz “apes strong together”, it’s impossible for WSB to sync activity at millisecond timeframes.

https://twitter.com/drewg__/status/1443731135880744972

2/ You don’t have to take my word for it. Just look at the relative distribution of which firms excel in which areas. If PFOF was linked to an informational advantage, we’d expect the firms with the most captive retail flow to have the strongest alphas.

3/ And we’d expect the firms with the strongest alphas to have the highest rate of liquidity taking aggressor trading on the lit exchanges. Larger magnitude alphas mean you’re more likely to have signal which exceed [spread]+[fees]

4/ Instead we see the opposite. Virtu which forms the one half of the duopoly virtually never crosses the spread. Their prop MM operation heavily leans on rebate capture and queue position over directional alpha.

5/ It’s much more debatable which firms have the best alphas/trade most aggressively. But a good candidate would be HRT. And They don’t even have a retail PFOF (yet).

6/ Some argue that retail flow used to have no alpha (at HFT timescales), but meme stocks changed this dynamic. Whereas before informed fund managers drove the market, now apes run this show. I have no direct experience with meme stock HFT, but am skeptical.

7/ Remember alpha is synonymous with toxicity. To the extent that retail all of a sudden had alpha, that would also mean PFOF internalizers are eating a ton of toxicity. We’d expect the GME explosion to have nuked the profits of internalizers.

8/ VIRT’s Q1 earnings tell a very different story. Meme stock mania has been very profitable for PFOF firms. That’s not consistent with retail meme flow being toxic/having alpha.

9/ As it turns out trading against counterparties who drive the market is not a good business strategy. Which is why Robinhood is paid more for their flow than Interactive Brokers who is paid more than Credit Suisse

10/ Again, all of which is the opposite of what one would expect if proprietary flow info pumped into prop trading alpha was any significant component of the PFOF business model.

11/ The one exception I can think of is a game called “follow-on” trading. Certain gamer shops will target large institutional managers like Vanguard, who they know have large persistent blocks to execute.

12/ The shop tries to send a really enticing price improvement for a really small size to profile Vanguard’s flow. They then turn that flow into an alpha, then stop quoting.

13/ The most important part to understand about this game is to that you internalize the *smallest* amount of volume you can get away with. Think of it like pinging a dark pool.

14/ The idea is that you bleed on a small amount of highly toxic, highly persistent flow, because you think you can turn around and execute in lit markets at larger size.

15/ (FWIW don’t feel too bad for Vanguard. They absolutely know the rules of this game. Just as often the gamers get counter-screwed when the block execution desks tease them into taking as large a clip as possible at the end of a run)

16/ Again this is entirely inconsistent with the behavior of retail PFOF. Citadel and Virtu would be fighting to get the *smallest* market share possible. Not competing to handle more flow. Remember the point of follow on is “just a small taste”— not to eat the whole buffet.

17/ All of which is to say that none of the stylized facts or behavior of the market participants indicate that retail PFOF creates any significant informational for exchange market making.

18/ Occam’s Razor: Robinhood’s flow is valuable because you want to trade *against* it. If one wanted to trade with it, it would defeat the entire point of internalizing it in the first place.