Rengan Rajaratnam is free.

Raj Rajaratnam is a former hedge fund manager who did quite a lot of insider trading, so it stands to reason that, if you hung out with him a lot, you might have done some insider trading too. Based perhaps on that plausible logic, federal prosecutors in New York brought insider trading charges against Rajaratnam's younger brother Rengan, who hung out with Raj a bunch, including in particular in wiretapped conversations, which is pretty much the worst place to hang out with him. But despite "boast[ing] to his brother that he had a friend working for the consulting firm McKinsey & Company who was a 'little dirty'" -- really who doesn't? -- Rengan was acquitted yesterday by a jury of conspiring with Raj to insider trade, after various other insider trading charges against him were thrown out by the prosecutors or the judge.

The case is "the only trial loss in the crackdown on insider trading at hedge funds for Manhattan U.S. Attorney Preet Bharara and the Federal Bureau of Investigation," after at least 81 convictions, and one question you might ask is, what is the right number of losses? Like, if the prosecutors never won a jury trial, that would obviously be bad, but if they never lost that's not ideal either. That might suggest that they're only taking easy trivial cases to trial and avoiding the hard ones. There are those who think of Bharara's perfect-until-now insider trading record as a bit of an embarrassment: Yes you've gotten 80-odd insider trading convictions, but that's only because you've pursued 80-odd insider trading cases. Insider trading cases are easy and dumb, and even the big ones are sort of penny-ante. Where are the criminal cases against the people who brought down the financial system etc. etc.

Another mortgage settlement.

"The Justice Department and Citigroup Inc. are close to a deal for the bank to pay about $7 billion to settle allegations it sold shoddy mortgages in the run-up to the financial crisis." So there's that.

Greatest fat finger story ever?

On January 30, 2014, the price of HSBC Holdings Plc shot up from 629 pence to 688 in less than a minute, triggering a halt in trading; the stock reopened back around 630. The U.K.'s Financial Conduct Authority tells the story of how that happened (via FT Alphaville). A broker got an order for $2.5 million worth of stock and misread it as 2.5 million shares ($25.9 million worth). Bad, but not fatal. The broker submitted it algorithmically to its broker, seeking to buy 20 percent of volume until it acquired the 2.5 million shares. Again not that bad; HSBC had averaged 24 million shares a day over the previous month. But then the broker's broker's computer said, no, this order is too big. And the broker and the broker's broker talked about it -- like, the actual humans called each other to say "these computers, amiright?" -- and they decided the fix was for the broker to re-submit five orders for 500,000 shares each. This, again: not necessarily wrong. But then the human at the broker did a magic thing: He submitted all five orders at the same time, each seeking to buy 20 percent of the volume. A computer will tell you that 5 times 20 is 100, and that 100 percent of the volume is the wrong percentage of the volume to buy. But the broker had doggedly and dumbly evaded the computers' checks, so he managed to buy all the stock there was to buy and pushing up the price by ten percent instantly.

Equity market structure.

Meanwhile across the pond the U.S. Senate was holding a hearing on regulation and high-frequency trading and dark pools and all that good stuff. These hearings are generally not, you know, the best way to learn about market structure, but this one was pretty good, though also pretty book-talky. Here are summaries from Bloomberg News and Dealbook, and here's a list of speakers and their written testimony. One notable moment was Jeffrey Sprecher, who runs the Intercontinental Exchange (and so owns the New York Stock Exchange), describing how horrified he is with the NYSE's proliferation of order types: "I live in a glass house. I’m trying to clean it up before I criticize others." (He also said that he aspired to get down to IEX's 4 order types, though not its 1 percent market share, buuuuuurn.) But my favorite moment might have come when a senator asked the panel what the biggest problem in market microstructure is and they all gave different answers, with some worrying about excessive intermediation leading to inefficient pricing, others about the perception of market "rigging" eroding investor confidence, and Ken Griffin of Citadel worrying about market stability, that is, the risk of flash crashes and so forth. Which makes sense: If you run a fund that does a lot of algorithmic trading and is basically good at it, your main worry is not that algorithmic trading is bad, but rather that your competitors are bad at it. Judging by that FCA story, they are.

Elsewhere, Finra is looking into retail broker routing practices, which should give you some sense of how much of an impact "Flash Boys" has had. And Eric Schneiderman would like to remind everyone that New York law preempts the law of supply and demand.

Bond market structure.

Here is Scott Skyrm on why you shouldn't worry all that much about fails in the repo market. Here is Cardiff Garcia on Zolton Pozsar on reverse repo as a new source of money-like claims for big cash pools. And here Jim Leaviss ponders the problem of credit market liquidity, which he attributes to the many bonds per issuer, each with a different coupon and maturity and terms. He concludes that each issuer should issue only one, perpetual, floating-rate bond that it can reopen if it wants to take on more debt or buy back if it wants to retire debt, and that bonds should be traded on exchanges. In other words he's discovered stocks? And everyone loves stock market structure.

Things happen.

Gold company investors don't like gold hedging, because they love only gold. Short sellers aren't short selling any more because prices are so high, okay. Bill Gross "made it clear that he wanted no more leaks to the press," several of his employees leaked to the press. Bart Chilton has stopped calling high-frequency traders "cheetahs" now that he's in the private sector, hmm. Argentina, on the other hand, is not toning down the rhetoric. Jim Simons is the best. The components of the Dow Jones Industrial Average mapped over time; don't you miss U.S. Cordage and Standard Rope & Twine? "A summer law associate probably isn’t half as productive as a good software engineering intern." The PR firm for a "garden-variety tech bro" chief executive of a social app company "wanted stuff on transiency and ephemerality; self expression; creativity and what drives it" from his blog-post ghostwriter.

To contact the writer of this article: Matt Levine at mlevine51@bloomberg.net.

To contact the editor responsible for this article: Tobin Harshaw at tharshaw@bloomberg.net.