StreetEYE Blog

Stories Are Powerful, But Check the Math

The first principle [of scientific inquiry] is that you must not fool yourself – and you are the easiest person to fool – Richard Feynman

In God we trust; all others must bring data. – attributed to W. Edwards Deming (ironically without any primary source backing up the attribution)

This Amy Cuddy TED talk was electrifying.

Video spoiler: If you adopt a “power pose” for 2 minutes, Amy Cuddy says it will not only change your posture, image, and attitude, but even your body chemistry, with more production of testosterone and anti-stress hormones.

It’s a great story, which is probably why it’s currently the second most-viewed TED talk.

Unfortunately, the published study study had only 42 participants. And other studies haven’t replicated the results on hormone production. Andrew Gelman even uses the opprobrious term p-hacking: data-mining to find a spectacular result.

The curse of dimensionality: the more things you measure, the more things will significantly deviate from the median.

The math can be counterintuitive.

Take a sample of apples. Grade each apple with a single number, like weight. For a contrived example, let’s say weight is uniformly distributed between 0 and 1.

What percentage of objects lie between 0.25 and 0.75 (the middle 50%?).

50% number line

Obviously, the blue line is 50% of the orange line.

Let’s grade apples along 2 dimensions, e.g. weight and redness.

What percentage of objects lie in the middle along both dimensions? Assuming weight and redness are uncorrelated, the answer is 50% squared, i.e. 25%.

How big a circle do we have to select to get to 50% of objects? We have to solve

    \[ \Pi r^2 = 0.5 \]

which gives r = 0.3989.

We see that we need a circle with almost 80% diameter to capture 50% of the square.

Number square

Let’s grade apples along 3 dimensions, e.g. weight, redness, and sweetness.

What range do we have to select to get to 50% of objects? We have to solve

    \[ \dfrac{4}{3} \Pi r^3 = 0.5 \]

which gives r = 0.492373.

We need a sphere with almost 100% diameter to capture 50% of the cube.

Sphere in cube

The point is, as you add more variables, the central 50% (or any x%) contains more and more extreme values. As you add dimensions, the outlying regions get bigger faster.

We can extend to higher dimensions which we can’t visualize, and chart the width of the 50% hypercube as we increase the dimension:


If you have 14 dimensions, the 50% hypercube is 95% of the length of the unit hypercube.

With enough features, anything or anybody is an outlier on some dimension.

Suppose you do an experiment measuring the variation of testosterone after assuming a power pose.

Suppose the power pose in fact has no effect on the level of testosterone (the ‘null hypothesis’).

If you observe a change due to chance variation, 95% of the time it will be statistically insignificant at the p > 0.05 level, and significant (p < 0.05) 5% of the time.

If testosterone and corticosteroids both exhibit no effect, the measured change in both will be statistically insignificant 0.95 * 0.95 = 90% of the time (assuming no correlation between them). As you measure more variables, the chance of one of them being significant goes up rapidly.

If you measure 14 insignificant variables, there’s a 50% chance one will be significant at the p < 0.05 level.

If you measure 50 insignificant variables, there’s a 92% chance one will be significant. 92% of that 50-dimensional ‘hypercube’ is in its outermost 5% region.

That’s how you get a prank paper to go viral showing chocolate helps people lose weight.

This sort of thing could be avoided if it was standard practice to hold back some test data, and do an out-of-sample test on any scientific finding. The methodology as practiced, to assume errors are unsystematic, and report p-values and significance on that basis, even on small samples tested for multiple relationships, seems weak and unscientific.

Returning to Amy Cuddy, you can interpret this a couple of different ways.

One interpretation: Statistics do not back up her story, that power poses raise hormone levels.

Another interpretation: Statistical methods are weak at finding complex stories, and you have to come up with a story to understand the world, and look for statistical confirmation where you can find it.

Acting with confidence and joy is contagious, to your own psyche and how others view you. That’s a story. Stories let humans understand and remember very complex phenomena.

For instance, attach a story related to their personal experience, and people solve tricky logic problems easily. Show them the same version as an abstract math problem, they fail miserably.

Feynman, quoted above about not fooling yourself, also said you must develop your intuition, thinking through examples and understanding the story of how things work as more than mathematical abstractions.

Stories are powerful. The more interesting things in the universe are complex interactions, like stories: evolution, the Big Bang, the French Revolution.

The curse of dimensionality means that as you absorb more features of the world, the possible states and explanations and oddities rise according to factorials and exponents. Things get curiouser and curiouser. There are complex interactions that can’t easily be explained. Stories are how humans make sense of a complex world.

Stories can mislead. A great story can be spurious, T. H. Huxley’s “great tragedy of science – the slaying of a beautiful hypothesis by an ugly fact.”

Stories are a powerful shortcut (Kahneman’s ‘thinking fast.’). But they are a shortcut that can lead you astray, so you also need to stop from time to time and make sure you know where you are going (Kahneman’s ‘thinking slow’).

So use your evolution-given power to understand complexity through narrative — but check the math.

Even if poses don’t elevate hormone levels, Superman and Wonder Woman were depicted that way for a reason. Don’t slouch through life due to lack of statistical evidence you shouldn’t!

(Mathematica notebook.)

iPhone Backdoors for the FBI, a blockchain approach for transparent due process, and why it’s a bad idea

So, the security complex is putting on the full court PR press for encryption back doors. See here and here.

Basically this is about giving someone a TSA lock to your phone and promising to keep it really really safe unless a legit law enforcement request is received. Of course, legitimacy is in the eye of the beholder.

One place where the real-world analogy breaks down is that any backdoor, in theory, enables industrial-scale exploitation. Potentially, it’s not just making it possible to open a car trunk that contains a body but more like requiring cars made out of material that’s transparent to the state. And then counting on due process to make it not infringe on the 4th Amendment and freedom from constant state surveillance.

The problem is, the folks at the NSA, CIA, DIA are in the deception business, and feel they have a moral imperative that demands deceiving the enemy, which in turn demands deceiving (lying to) the public.

I don’t even blame them, they have a job with a lot of risk, no real glory. Their job is to do what they can with the tools they’re given, and probably take the fall for ‘not connecting the dots’ even when they pretty much connected the dots.

Hillary talks about a Manhattan Project for cybersecurity…the truth is there is no way to create a magic bullet that can only be fired by the good guys, and there has been a $10b annual Manhattan project for years to enable the NSA to undermine and exploit the tech industry’s security.

So, I really thank Tim Cook for standing up to useful idiots who say Apple enables terrorists.

But with the current level of stupidity, there’s a very real possibility it’s a losing battle against that accusation, especially if there actually is a terrorist attack that hits an investigation roadblock due to iPhone encryption.

If you go down the backdoor road, there has to be maximum real-time transparency and due process. That’s kind of what the blockchain is: a secure, open public ledger of transactions. They can be money transactions, or transfer of ownership of other rights or responsibilities, or any bits, really.

So anyway, here is, as a thought experiment, how you can use the blockchain to enable transparency and due process in a key escrow scheme.

1) When Apple generates keys to encrypts your phone, they keep a copy. The copy is kept in such a way that the only way to release it is through due process.

  • Records of keys to be kept in one location, e.g. basement vault at Apple HQ.
  • They are kept only on physical media.
  • There is no network access to that storage
  • The location is electromagnetically shielded, physically secured per DoD standard for most secret information.
  • The keys are generated and conveyed to that storage securely, and any copy outside the room is destroyed. How to actually guarantee that is another giant bag of worms that is beyond my pay grade. But it has to be done per a checklist like generating nuclear launch codes, and the process audited regularly, and e.g. Tim Cook to certify annually under criminal penalty that the procedures were followed, and any shortcomings or attempts at circumvention publicly disclosed.

2) When law enforcement wants access to a phone for a criminal investigation, they post the request on public blockchain that is jointly maintained by all the interested parties, including watchdogs like the ACLU. The request records

  • requestor (state attorney general, US attorney, etc.)
  • target device
  • specific major felony accusation
  • specific individual or witness

3) Judge approves the request and posts approval on the blockchain.

4) There is a reasonable delay e.g. 72h to allow challenging/appealing the request.

5) Public signature by e.g. Tim Cook that he personally authorized access after finding it was legit and all necessary information was public on the blockchain, and appeals/challenges exhausted.

6) Keys transmitted to law enforcement by similar nuclear launch code checklist, e.g. all access to the physical location and media where it’s stored by two people who follow the checklist and record that it was followed, under criminal penalty for exfiltrating information inappropriately, or not documenting any attempt at circumvention. And again, procedures and logs subject to annual 3rd party audit, and management to certify that all procedures followed, any gaps or attempts at circumvention publicly disclosed.

The point of this exercise is that once you have a backdoor, you need real, public due process with teeth.

This process will satisfy no one. It’s a huge hassle for e.g. Apple. The security community wants something where they can ask for an inch and take a mile, and blame civil authority when they don’t find the threats. The civil liberties community will rightly suspect there is a hole in there somewhere, or that one will be created at the next ‘national security emergency,’ because that’s what the public raised on ’24’ and ‘Homeland’ expects.

And of course China and Russia will demand their own, much more leaky version of this, and Apple will end up in the Stasi-enabling business.

More and more, your whole life is on the phone. It leaks plenty of information semi-voluntarily about everywhere you go, everyone you spend time with, communicate with, what sites you browse, who you transact with. The security guys can do all kinds of other things to track you, GPS monitors, hack your phone, search your garbage.

Better to not go down this road of giving the surveillance state unfettered access to everything. And maybe it’s time to try to use technology for cryptographically secure transparency and due process.

The most popular keywords and sites of 2015

Here’s a word cloud of StreetEYE headlines in 2015 (click to embiggen).

word cloud

Greece (remember Greece?) beat out China for the biggest headline-bait of the year. Tsipras even beat out Yellen, although Grexit ended up a non-event. (To my surprise actually…Schäuble and Varoufakis were both apparently playing for Grexit, so I thought it would take a miracle. The center held, but the political cost to Europhiles like Merkel, Hollande, Draghi, Renzi hasn’t been counted yet.)

One thing that makes me happy: we posted stories from 1536 unique domains in 2015. We (or you, our readers and curators) posted about 65 headlines a day. About half were from the ‘Big 5’ of Bloomberg, Wall Street Journal, Financial Times, New York Times, and Reuters. The balance were from the long tail of domains (see below).

The most-clicked stories of 2015 were quality features, but sometimes a little click-baity. We’re all about the good headlines.

Details below. Thanks for coming along on this journey. If you have any comments or suggestions, please let us know. Have a great 2016!

1 It’€™s sleazy, it’€™s totally illegal, and yet it could become the future of retirement (Tontines, per Moshe Milevsky (I agree))
2 What the Smartest People in Finance Think You Should Read (books)
3 What is code? If you don’t know, you need to read this
4 The CEO Paying Everyone $70,000 Salaries Has Something to Hide
5 How Two Guys Lost God and Found $40 Million
6 Carly Fiorina failed to register this domain.
7 I Had a Baby and Cancer When I Worked at Amazon. This Is My Story
8 Inside Hunt & Fish, where beauties trawl for sugar daddies (throw’em back!)
9 What The New York Times Didn’€™t Tell You (about Amazon)
10 ‘Shell-shocked’ CNBC staffers had long flight home

Top domains of 2015

1 Bloomberg
2 Wall Street Journal
3 Financial Times
4 New York Times
5 Reuters
6 The Guardian
7 Bloomberg View
8 Business Insider
9 Washington Post
10 The Telegraph
11 VOX
13 Re/code
14 The Economist
15 European Union
16 Fortune
17 Vox
18 Quartz
19 MarketWatch
20 Project Syndicate
21 Medium
23 Forbes
24 Federal Reserve
25 New York Fed
26 A Wealth Of Common Sense
28 ZeroHedge
29 TechCrunch
30 Politico
31 The Reformed Broker
32 New Yorker
33 NY Post
34 Dealbreaker
35 The Verge
36 Calculated Risk
37 mainly macro
38 Marginal Revolution
40 Yahoo
41 Fusion
42 Econbrowser
43 The Atlantic
45 IMF
46 Huffington Post
47 BuzzFeed
49 Stumbling and Mumbling
50 New York
52 CNNMoney
53 BBC
54 Wired
55 Economist’s View
56 Barron’s
57 Gawker
58 LA Times
59 Yanis Varoufakis
61 Bank Underground
62 YouTube
64 Noahpinion
66 USA Today
67 Slate
68 Bank of England
69 The Independent
70 Der Spiegel
71 Worthwhile Canadian Initiative
72 BBC
75 Vanity Fair
76 Macro and Other Market Musings
77 Conversable Economist
78 InvestmentNews
80 BIS
84 The Grumpy Economist
86 The Times
89 Time
96 Foreign Policy
97 The Big Picture
100 Institutional Investor

The End of the PC? On Intel’s Apple and ARM problems

apple-iphone-6s-live-_0752.0Tim Cook has been running around heralding the end of the PC. A self-serving assessment, but Intel and the PC ecosystem are going to struggle to maintain their traditional relevance. In this post, I will look at 1) the narrowing Intel/ARM performance gap, and 2) what the ‘end of the PC’ might look like.

1. The narrowing performance gap

At the introduction of the new iPhone 6s, Phil Schiller made the claim that the phone’s ARM-based SOC (system-on-chip) is more powerful than chips powering 80% of laptops…in other words faster than the low-end Intel Atoms, Celerons, i3s, and on a par with the Intel’s bread-and-butter Core i5s which powers pricey MacBook Pros.

iPad Pro benchmarks

Benchmarks bear him out (scroll through for Intel comparisons).

Comparing the new iPad Pro to the latest Intel-based Microsoft Surface Pro 4, the new iPad has a bigger screen, weighs less, and has longer battery life than Surface.

The Surface Pro 4 seems to have similar single-threaded performance and higher multicore performance. It seems positioned as a good laptop, which can also function as a tablet.

(Aside: Mystifyingly, no built-in LTE option. For many people having a single mobile plan and tethering other devices via Wifi/Bluetooth seems like the best option, but for many corporate use cases LTE is still a needed option.)

Bottom line: Right now, ARM can’t offer the raw performance of the high-end Core i7 and Xeon processors. But it can hold its own against the bread-and-butter i5, at a superior performance per watt and performance per dollar.

This is a big problem for Intel.

2. How did this happen?

Historically Intel has had a number of key advantages over its competitors in CPUs:

  • Above all, the most advanced chip fabrication. In process, Intel typically was a generation ahead of its competitors like Samsung, TSMC etc. That means smaller chips, lower power per transistor, higher performance.
  • CPU R&D – more advanced architecture design with higher transistor counts, larger caches, out-of-order execution pipelines etc.
  • Scale and network effects – more investment in compilers, tools for the x86 platform.

That added up to a mega franchise:

Investment in fab and architecture
-> most advanced and powerful CPUs
-> market share, volume, industry standards
-> huge margins
-> plowed back into massive investment in fab and architecture
-> rinse and repeat.

Intel had 2 ‘high quality’ problems:

  • Backward compatibility – Acts as a performance tax, requiring legacy features / more transistors/ more power / hamstrings architecture. Unclear how significant due to Intel’s other advantages but it’s there.
  • High margins / innovator’s dilemma. Intel could no doubt have offered an x86 mobile architecture that was cost/performance competitive with ARM. But they could not do it without having it installed in PCs and servers and cannibalizing their PC/server business.

The tick slip: Intel has operated on an alternating tick/tock model. On the tick, they shrink the existing architecture, putting it on a new manufacturing process that runs on smaller chips with closer-packed transistors that draw less power and run at higher clock speeds. On the tock, they introduce a new architecture with more transistors and design optimizations.

These tick/tocks about a year apart have kept them ahead of the competition.

In September 2014, they started shipping Broadwell chips manufactured on a 14nm process.

In August 2015, they started shipping Skylake chips, the new micro-architecture on the same 14nm process. But they also announced the 10nm successor process would be delayed until 2017.

In the meantime, Samsung began shipping 14nm SOCs in the Samsung Galaxy S6 smartphone in early 2015, over a year behind Intel. Apple then released its iPhone 6s, also on 14nm Samsung and 20nm TSMC SOCs, in late 2015.

I hasten to add, all 14nm chips are not identical. For instance, the iPhone you get may have either a 14nm Samsung or a 20nm TSMC chip: they are dual-sourcing the SOC. Some testers and pundits proclaimed the TSMC iPhones on the larger die/older fab process actually used less power and performed better, contrary to what one might expect. Samsung’s 14nm may not be higher density than Intel’s, and clearly not even a knockout punch vs. TSMC’s 20nm.

Nevertheless, right now Intel’s competitors are nipping at Intel’s heels. And Moore’s law is running out of room. At this point each transistor is a few dozen atoms. We have maybe 6 50% ‘shrinks’ before we hit a single atom. People have been saying Moore’s law has reached its limit for a long time…but perhaps Intel’s struggles to stay ahead are the real deal this time.

ARM is big in cheap, high performance, PC-incompatible Chromebook laptops. But ARM servers haven’t had an impact yet. Nevertheless, for loads highly distributed across numerous servers like Google and Facebook’s immense Web server farms, they would appear to make a lot of sense. Companies like Calxeda have tried before and failed. But the ‘tick slip’ seems to create a window of opportunity where the Intel fab edge is limited, for the next year or more, and could get closed entirely if ARM fabs improve further. (Both Samsung and TSMC say they will match Intel’s roadmap, but talk is cheap.)

The key metric in massive web farms is performance/watt. If ARM OEMs can achieve fab parity with Intel, the case appears to be compelling. Intel would then have to cut margins to compete. It would seem incumbent on the Googles, Facebooks to be testing ARM and developing standards for ARM servers via the Open Compute Project.

From there you might see ARM start showing up in corporate server farms, cloud infrastructure providers like Amazon AWS. Eventually, ARM CPUs with larger caches, more execution pipelines could be designed to compete with Xeon and make inroads in the largest single-server applications, like databases.

It’s also worth pointing out this rumor that the next iPhone will be on an Intel-manufactured ARM SOC. That would be a huge Intel hedge against a decline in its x86 business. Dell’s purchase of EMC and attempt to sell its PC business can also be seen in this light.

3. The phone as PC

The other issue is…the phone is the primary computing platform for more and more people.

On many dimensions your phone is more advanced than your PC. It features

  • 20+ radios communicating on any mobile network known to mankind, WiFi, Bluetooth.
  • A phone properly integrated into the OS and messaging, which PCs never really got working right.
  • Voice control and input with Siri, Google, Cortana
  • Touch screen UI with multi-touch and Apple’s pressure-sensitive 3D Touch, biometric fingerprint ID.
  • Integration of watches, fitness trackers, internet of things; sensors galore (2 cameras, multiple mikes with noise canceling, GPS, accelerometer, gyroscope, digital compass, ambient light detector, proximity sensor)

The mobile phone is ludicrous technology, the nexus of a technology singularity. It may not have as much RAM, or a competitive equivalent to an Excel or Powerpoint. But if most of what you do is email and Web browsing, and you can make do with Google apps, you’re in good shape when you just have your phone/tablet.

It’s perfectly fair to say that a lot of office workers could do all their work on their phone/tablet. A lot of sales guys just need a phone and a CRM app on a tablet.

The Mac was a miracle to my generation in college. A generation raised on an iPhone and iPad may well view a Mac as a step backwards.

With the iPad Pro, Apple is aiming at content creators. Give the iPad Pro a decent keyboard, stylus, and mouse and you can use it as your main screen, even if you’re a power user. I’m kind of expecting it to bomb, near term. Not enough apps or a large enough market for those users.

But Apple is just one killer app away.

In offices, as iPhones became popular, most companies went to BYOD – bring your own mobile device for email. This makes users happy, and to some extent, system administrators.

A few companies have gone to VDI – virtual desktop infrastructure. The actual PC runs on a virtual machine in the cloud, users view it on a local thin terminal display. Your Excel and Powerpoint, even Bloomberg and trading systems live in the cloud, you connect over the internet, just like RDP you may be familiar with (Remote Desktop Protocol).

VDI is a far better disaster recovery posture. Sit anywhere with a terminal and a network connection, get all your apps and data, everything backed up to the cloud.

Both BYOD and VDI are a far better security posture. Harder for malware on your phone to spread throughout a company, since it’s basically a foreign device outside on the Internet. Get an infection on your VDI image, restore it immediately from a pristine image.

You can see where I’m going with this…VDI will potentially be a killer app on an iPad Pro.

All your servers and desktops go into the cloud using e.g. Amazon cloud infrastructure as a service and VDI. Then you give your users a tablet with the keyboard and VDI app. Presto, no more PC desktops or servers in your front office. The end of the PC world as we know it.

I could see quite a few knowledge workers being issued hybrid PC/tablets like Surface Pro and using them primarily as tablets. Over time they may find they don’t need the PC functionality. And departments and even companies go 100% BYOD.

Of course there would be many Intel CPUs in that cloud infrastructure. But over time perhaps not as many, unless Intel retains their fab edge even as they lose a chunk of the revenues that support it.

The phone would displace the desktop PC, while the PC platform would complete its displacement of the old centralized mainframes.

Back in the 80s, who would have thought that was in the original IBM PC’s future. Big iron mainframe guys were laughing at the PC as a toy. But then again so did the the first iPhone. The next big thing often starts out looking like a toy.

Strange times. And possibly risky ones for Intel.

Zombie army

Is China’s sale of Treasurys ‘quantitative tightening’ for the US?

There’s this notion going around that since the Fed buying Treasurys was QE, therefore China selling Treasurys constitutes monetary tightening.

Nope. The root cause of the disequilibrium and resulting capital flows is capital flight from China to the US.

An oligarch wants to buy a condo in New York. He takes yuan to the Chinese central bank and exchanges them for USD. In order to provide the dollars, the Chinese central bank sells some Treasurys.

How can capital rushing into US risk assets be ‘quantitative tightening?’ It’s the opposite. At the end of it, the Chinese foreign sector has exchanged an American safe asset, a Treasury, for a condo at 432 Park, a risky asset. The US has built a big building, generating jobs, demand, maybe a little inflationary pressure.

The Chinese central bank could do nothing, and the yuan will drop until US assets look expensive to Chinese oligarchs and equilibrium is restored.

Or they could intervene to limit the drop in the yuan by buying yuan for dollars, which requires them to sell Treasurys.

The sale of Treasurys is a second-order reaction to partially stem the drop in the yuan and accommodate capital flight. At the end of it, at best, it’s an even swap of Treasury bonds for condos.

It’s possible that a central bank might sell Treasurys to buy non-US assets, gold, etc. In the case of the Saudis selling reserves to support their local economy, pay for imports from a lot of places, there might be some ‘quantitative tightening’, ie capital flight from US assets.

In China, though, it seems that the sale of Treasurys is a second-order effect from flight into US assets.

It’s a good principle in life and in economics that second-order effects partially offset first-order effects. It’s more probable that on balance, the capital flight from China into the US is a stimulus, and the Chinese FX intervention and sale of Treasurys partially offsets that stimulus.

Tontines: strange name, great idea

A huge problem in retirement planning is a safe spending rate, so you don’t outlive your money.

One side of the problem is, how much can you spend and not risk running out of funds in say, a 25-year retirement? See, for instance, our Cat Food Calculator mad science experiment. One can even calculate a spending solution which maximizes your certainty-equivalent spending based on historical returns, and your risk tolerance.

The other side of the problem is, what happens if you live to 105 and have to fund a 40-year retirement? If you plan for 40 years, you are likely to under-spend and leave a large estate. If you plan a 30-year retirement at 65, and it gets you safely to 95, there is still a small but potentially catastrophic possibility of outliving your savings.

tontine can address this problem. Here’s how it might work:

  • Imagine a Kickstarter-like page: “Seeking 1,000 men (or women) who were 50 years old on Jan 1, 2015 to start a tontine.” When we get 1,000 eligible participants commit, we launch the tontine.
  • Each participant funds the purchase of 200 SPY ETFs — at current prices about $40,000. (The math works out if you let people vary their contribution, within limits, but lets keep it simple).
  • The ETFs go into a trust.
  • When the cohort reaches age 801, about 53% of males will still be alive. At that point the tontine begins paying out.
  • We set up a fixed annual distribution of shares to each survivor. The amount is set at, say, 10% of the shares divided by the surviving participants.
  • Let’s hope we get a 4% real return on the SPY, and assume1 that we can re-invest dividends tax-free. After 40 years your initial $1 investment in a SPY has grown to about $4.80.
  • In addition your equity share in the trust has almost doubled because 48% of the original participants are no longer with us. Your $50,000 initial investment has bought you a $366,000 old-age fund.
  • So 10% of the SPYs are distributed. Their value has gone up ~5x. The survivors are about 50% at this point. So you can get a distribution valued at $36,600 in year 1. The trust can afford to distribute the same number of shares to each survivor every year, and never run out of money assuming the current life table. With luck, the S&P continues to appreciate. Your initial $40,000 has bought about that much per year for the rest of your life.
  • If the tontine runs out of money because cancer has been cured and everyone is living to 120, that’s it, it’s out of money. It’s a risk you take, along with S&P fluctuations. As your retirement progresses, you will want to monitor the expected payout from the tontine, plan and adjust accordingly.
  • When there are say 10 people surviving, the remaining SPYs are distributed pro-rata. Saves on administrative costs and gives the lucky ones a payoff and an estate.

What is insurance except a pool of people coming together to share risks? What better product for our era than a crowd-sourced, peer-to-peer, sharing economy life insurance solution.

The great thing about this tontine is, for a small investment you fund a big part of your needs in the event you are one of the lucky ones to live a really long time. You can focus on saving funds for the earlier part of retirement when you know you are likely to be alive. You can plan to spend pretty much your entire nest egg over the first decades and don’t need to worry about the tail risk of living to 105.

There is an existing product which also addresses this problem: an insurance company variable annuity.

The problem with the variable annuity is, it can be quite expensive — it is a market that is a bit of a minefield. Variable annuities are complicated, and are often high commission products. They often have high expense ratios. (Vanguard is a great provider, but if I read it correctly, they charge 0.46% to 0.77% per year on top of the management fees for the funds you invest in. Adds up over 40 years.) The insurance company takes the other side of the longevity risk, ie they keep whatever is left over when people die quicker than expected. There is a small, but potentially non-negligible credit risk. If an insurance company goes broke because, for instance, its overall returns are too low and it takes too much of the wrong kinds of risk, you might not get the expected payoff.

At scale, the tontine should be feasible at index-fund expense levels, like 25bp. Or, possibly, more like account maintenance fee levels. The problem with the tontine is… it’s illegal in the USA.

But it really would be a great product. It’s extremely unfortunate that our system allows all sorts of expensive, gimmicky products, not to mention outright shenanigans, but blocks relatively simple, legitimate, useful peer-to-peer products.

I encourage any renegade entrepreneur to take a crack at it, set up a Kickstarter-like website, and dare the authorities to shut it down. Uber didn’t ask permission. It  takes boldness to disrupt financial services. But regulators may not be sensitive to the public interest, and insurance companies may not be as easily overcome as taxi commissions and fleet operators.

Like Warren Buffett on space exploration and tech stocks, I applaud the endeavour but may prefer to skip the ride.

(Disclaimer: it would be immoral to take Uber-like risks with people’s retirement savings. Perhaps it’s time for a group of willing participants to try a test case and challenge the law. Or maybe some very smart lawyers can identify a workaround. Or maybe there’s a Bitcoin-like solution. That’s a joke, mostly: you can’t keep a large pool of real-world assets like SPYs outside the law.)

1 The tontine could, of course, invest in something besides the S&P, like a bond fund, a balanced portfolio, etc.  It might also make sense to let investors fund the tontine over say, 5 or 10 years. It could also start paying before age 80, for instance it could fund an entire retirement plan starting at age 65. I chose this example because it highlights how the tontine simply, effectively, and cheaply mitigates longevity risk.

2 Unfortunately, an unwarranted assumption: In a taxable account, dividends would be taxable. In an IRA, minimum distributions would be tricky, one might have to start distributing at 70 ½.

God help us

A quickie, light rant on politics <cough> for a Labor Day weekend.

As a jumping-off point, supposedly the majority of Republicans think Obama is a Muslim. That’s probably bullshit. As the Brits say, they are taking the piss. They don’t like Obama and pick the most other and derogatory (in their mind) response. If you asked them to bet $5 on what religion he was raised in, what services he attends, they would not say Muslim.

It says something that you take the most experienced elected officials the GOP has to offer, and they can’t break out of single digits, and Trump is kicking their butt by 20+ points. Poll responders who say they support Trump are also taking the piss. A lot of Republican voters do not like the establishment GOP. Given a free shot, they’d as soon kick’em in “Deez Nutz” as admit to supporting one of them. Whether a lot of voters have actual enthusiasm for Trump in a binding primary, or a general election, or as a President may be a different matter.

But it’s more than just a protest. There’s a fundamental split between the donor base and establishment candidates, on the one hand, and the voter base on the other. Practically no one in the voter base wants to gut Social Security. But you have to say that as a candidate, or bye-bye PAC money. On the other hand, no one in the donor base is anti-immigration. It keeps wages low. The business establishment played the game of the Reagan amnesty, with the caveat that employers would be on the hook for checking legal status. Then they gutted enforcement. What started off as a fair and humane compromise turned out to be a bait-and-switch. It’s a sad day when the USA acts like Dubai, de facto inviting a foreign worker army that can be taken advantage of with impunity in fruit orchards, slaughterhouses, and Home Depot parking lots. Economic pain, and latent and overt racism and anti-immigrant xenophobia play their role, but righteous anger is there too.

Previous populist Tea Party movements have been a bust. Like grass-roots wack jobs like Sharon Angle or “I am not a witch” Christine O’Donnell. Like the Dick Armey Tea Party scam, where they rile up the base, affinity-scam them for money, and then try to deliver them as voters to establishment legislators. A lot of Republicans would welcome a Trump hostile takeover, if he would deliver an actual populist movement. The question is whether that’s what Trump wants to do, or whether he just wants the attention to inflate his ridiculous ego. Seems equally probable he’s the one taking the piss out of all of them, and us.

The GOP is a strange alliance of pro-business types, a theocratic American Taliban1, libertarians, and nationalist populists. Oddly, they don’t seem to have a credible candidate or shot at taking the Presidency. Meanwhile the Founders’ Senate gerrymandering and a number of states’ Congressional gerrymandering doesn’t give the Democrats much of a shot at taking the legislative branch back for a while. Sounds like a recipe for gridlock. What a total clusterfuck.

I feel sort of the same about Hillary as I do about Tom Brady, Boston’s finest Deflator in Chief. He tampered with the football and covered it up. That is a threat to the integrity of the game, and needs to be nipped in the bud. But it’s not really that different from a pitcher spitting on the ball. Throw him out of a game, slap him with a fine, and get on with it. At most a 2-game suspension and a fine, one for tampering and one for covering up. Same as Ray Rice’s initial suspension. At some point the greater threat to the integrity of the game is a lynch mob against a terrific competitor, or an arbitrary star chamber settling unrelated scores. Fuck Boston and fuck Tom Brady, but let the guy play.

So Hillary placed her own IT guy at State and paid him under the table to maintain her server. Now, Colin Powell and all manner of Bush White House staffers had their own off-government email addresses. This seems to go a bit beyond that. But let’s face it, anything Hillary puts in an email is going to be out in public. Either the Republicans are going to get it via a witch hunt subpoena over some manufactured scandal, the intelligence services domestic or foreign are going to get it, some Snowden type is going to release it. (The White House and State Department systems were both massively compromised by foreign intelligence services and got shut down.)

So she said fuck that, I’m going off the reservation, going full Cheney, doing it my way. If she came out and said she did it that way so she could do her job right and fuck the political consequences, it could even be called a gutsy decision. The problem is, she did it to protect her ass and her political future against perceived persecution.

So slap her on the wrist, tell her that’s not how you do it, set some binding rules on official communication. (FFS, it’s fine to have your personal text messaging or e-mail account, just don’t use it for government business. And maybe, allow a legal purge of official emails for personal or embarrassing irrelevant stuff, as long as a complete record of official business is preserved.)

The problem with Hillary is, she has been subjected to all kinds of insane attacks and witch-hunts her whole life. And unlike Obama2, she hasn’t reacted with mostly good humor and the occasional on-target dig. She gets paranoid. She sounds evasive, entitled, petulant. She fires travel agents and White House staff she thinks might not be loyal. And now this, her own personal IT department to flout government rules.

So, as President would she put the past aside, say she’s attained the equanimity of her years, and work for the future of the country with anyone who was willing? (Which doesn’t seem to count many Republicans.) Or would she go full paranoid score-settling Nixon? I wish I were more optimistic.

God help us.

1 FFS, that Kentucky born-again bumpkin is not in jail for exercising her freedom of religion, she’s in jail for trying to use the power of a cushy state bureaucrat to impose her religion on everyone.

2 Which immediately sets all the assholes puckering up about how un-Presidential and divisive Obama is. Now seriously, on what planet is the guy who can’t shut up about Obama’s birth certificate going to make America great, while Obama is the great divider?

Smart Beta: Maybe Smart, But Definitely Not Beta

A donut with no hole, is a danish. – Ty Webb

Live your life as though your every act were to become a universal law. – I. Kant

So-called “smart beta” is having a day in the sun. Pioneered by the very smart Rob Arnott, the basic idea is: don’t invest in an index weighted by market cap — invest in one that weights according to a non-market fundamental value measure, or even equally. Smart beta funds and ETFs have grown faster than the market. I’ve even seen versions of this quote in a couple of places: “market cap weighting is the worst way to own a broad index.” This makes me rage a little, hence this post.

When someone badmouths plain-vanilla market cap passive investing, hold onto your wallet.

Point: If you buy a market-cap-based index like the S&P, the higher the price of a stock, the higher the market cap, and the more dollar value you’re supposed to own. So, the argument goes, you’re going to own too much of the overpriced stocks that are going to underperform and too little of the cheap ones that are going to outperform. ‘Smarter’ to buy an equal-weighted index, or one weighted by earnings, or dividends or another indicator of fundamental value!

Counterpoint: On average, the S&P investor will pay the market price for value. (Jane, you ignorant slut!)

Suppose Enron is in the S&P 500, and you’re an investor in an equal-weighted S&P 500 portfolio. It goes down to 0 and gets delisted. Your equal-weight index or fundamental-weight index is buying it all the way down (assuming no change in reported fundamentals). When it hits a sufficiently small price, you’re going to own the whole company.

Nice public service, dedicating 1/500 of your portfolio as insurance to bailing out shareholders of any piece of s**t company that fails its way out of the index. The ‘smart’ part, I guess, is dedicating only 1/500 to any one disaster. But that’s a bounded definition of ‘smart’.

If you just buy the market weight and hold it all the way down without throwing good money after bad, that seems like it would be a little smarter.

Likewise, the equal weight investor buys a 2% position in Microsoft when it enters the index and keeps selling it all the way up, which, as it turned out, was not the best strategy. (Selling on the way up, not buying an initial market overweight position.)

The point is, maybe ‘smart beta’ will help you avoid lunatic bubbles like Cisco in 1999, but there are always other traps it will fall into. The overpriced stock bubble may not be the most persistent or expensive anomaly. (Maybe the worst is to lose money in a bubble, then pay for expensive anti-bubble gimmickry.)

You could get even ‘smarter’ beta, and buy a dividend- or earnings-weighted index. Buy more of the stocks with high earnings or dividend yields, less of the stocks with low earnings or dividend yield. This will overweight cheap stocks. Indeed, I’ve never heard an investor say they prefer to pay higher prices. We’re all value investors, just some of us are willing to pay a little more for growth. An efficient market theorist might say the value premium is a liquidity premium, a small-stock premium, a 60-year-flood premium since the cheap unloved stocks are going to be out of business in a Great Depression scenario (when 20% of companies went out of business).

If earnings-weighted indexes are really just a stealth factor model, you could weight your portfolio based on the best factors you can get paid for, like low price-to-book, high return on invested capital, etc.

But at some point you have to say, sufficiently advanced smart beta is indistinguishable from active management.

And something like the equal-weighted S&P is a gimmicky hack, and in the long run, gimmicky hacks usually don’t work.

Beta is the market risk-return profile. Anything that deviates from that is, by definition, seeking to outperform the market. In other words, active management. Alpha. The smarter smart beta gets, the more it looks like alpha.

‘Smart’ beta, by definition, is not beta.

If smart beta isn’t smart, it’s just beta. If it is smart, it’s just alpha. It can be smart or beta, just not both at the same time.

Smart beta is really fraidy-cat alpha. It’s an investor claiming to be passive while following an active strategy that is highly diversified and based on the index.

I’d rather have a guy claiming to sell me beta with a sprinkling of alpha, than a guy selling me alpha and calling it beta.

Thought experiment: Suppose, in some alternate universe, every investor sought to invested in the equal-weight index.

The market is an election. Everybody in the market ‘votes.’ The equal-weight investors vote. The active investors vote. The index investors apathetically say they’ll go along with whatever everyone else decides.

The market cap is the current equilibrium of all those investors’ strategies.

If everyone is an equal weight indexer, as soon as a stock is added to the index, the stock gets bid to the price at which its market cap gives it equal weight in the index. That doesn’t seem like a rational price. Or a smart price. That seems like a dumb price.

Being an equal weight indexer violates the Kantian/Nashian categorical imperative to invest the way, in a rational world, everyone should invest. It assumes everyone else is doing it wrong, in a very naive way, and will persist in doing so.

The current market cap is what current marginal buyers and sellers think the ‘right’ weight in the market should be. That’s the crowdsourced answer to what the company’s market cap ‘should’ be in the typical investor’s portfolio, as determined by investors of all stripes — indexers, active investors, and ‘smart beta’ investors.

Those strategies’ popularity are themselves determined by their own equilibrium. When you have too many active investors, the return on active investment goes down, the expenses are not worthwhile, the least successful active investors switch to passive investing.

When you have too few active investors, the returns to active investing go back up. See Stiglitz and Grossman: If a market is perfectly efficient and prices are arbitrage-free, arbitrageurs don’t get paid and exit the business. If there are no arbitrageurs, prices get out of line. In general, some arbitrageurs get paid, and prices are approximately, but not perfectly efficient.

If prices get very inefficient, arbitrage capital and talent enters the market. If prices are too perfect, arbitrage capital and talent exits the market. And then maybe even a poor capital allocator like Donald Trump can beat the market.

This is worth exploring further, since some investors (astoundingly) make an argument that increased indexing and herding are bad for active investors. The argument is that markets are like a poker game, and when dumb money turns to indexing, there are fewer underperforming investors to fleece, and less divergence between good and bad stocks, good and bad strategies.

By that argument, if you were an active investor like Warren Buffett, and could pass a law forcing all other investors to abandon active management and switch to an index, would you do it? Or would you prefer not to eliminate your competition?

Another thought experiment: Let’s think through what happens if active investors leave the market and people switch to indexing. Take it to the extreme, where there’s one active investor left in the market.

An IPO comes out. Mr. Active Investor solely determines the IPO price. He doesn’t have anyone to compete with or have anyone to trade with, since it’s not yet in the index.

The IPO at some point gets added to the index. Indexers have to buy the stock. Mr. Active Investor solely determines the price at which it gets added to the index. If a company gets delisted in favor of another company, he solely determines the price which the exit takes place.

Seems like a sweet deal. Demand a big premium when a stock goes into the index, demand a steep discount when one leaves the index.

But let’s ignore IPOs and individual stocks and suppose you can only trade the index.

Suppose the passive investors decide to liquidate, they need cash to fund retirement, or just turn panicky. Mr. Active Investor solely determines the prices at which he is willing to take the index portfolio off the hands of the passive investors.

Later on, suppose passive investors have cash to invest, or turn optimistic. Again, Mr. Active Investor is the only person who can sell them the index, and he can set the price and sell them at a nice markup.

Well, my point is this: If everyone indexes, in the short run it’s not a stockpicker’s market. Anyone who owns any stock in the index is just getting index performance.

But Mr. Active Investor can time the market and make a bid-ask market for the index, selling when it’s x% above his estimate of fair value, and buying when it’s x% below fair value.

The indexers are all going to match the performance of the index every single day. And yet, Mr. Active Investor is going to crush them. Because he’s always buying low and selling high. In a sense, he’s going to perfectly time the market by determining what price he’s willing to buy and sell at.

The more herding, the greater the volatility over time, and the more Mr. Active Investor crushes the herd.

Getting past that important aside, and back to our original point, I haven’t seen any ‘smart beta’ fund or ETF that significantly outperforms, after fees, transaction costs, and taxes, including Rob Arnott’s PRF. None of them have gotten really big or done really well over the long haul. The most popular ‘smart beta’ funds seem awfully niche or gimmicky.

I could see a place for some ‘smart beta’ funds or ETFs to fill a role in a portfolio, such as the Russell ‘value’ or ‘growth’ to boost those factors, especially when one style or the other is unusually out of favor. As long as you’re aware of the liquidity issues and other potential pitfalls of ETFs, which didn’t cover themselves with glory amid the recent volatility.

The dumbest way to hold a broad index isn’t a market cap weighted index. It’s something that pretends to be active management, or is poor active management, and charges high fees. ‘Smart beta’ is just a marketing gimmick for systematic active management.

As the great man might have said, investing is a dark ocean without shores or lighthouse, strewn with many a wreck… and out of the crooked timber of investment managers, no straight thing was ever made.

P. S. After posting this opus, I became aware that Cliff Asness et al. did part of it much better, showing that empirically, fundamental indexing is identical to systematic value investing. The relevant passage is here on page 10.

The Mathematics of Bluffing

A quick post about poker! That seemingly simple, deceptively complex game with a number of interesting parallels to investing. I just watched the MIT lectures on ‘Poker Theory and Analytics,’ an ‘Independent Activities Period’ mini-course, and for our mutual amusement, I worked through the math on bluffing, which is an interesting problem I had never done the full deep dive into. Here it is, including a Mathematica notebook.

Here’s how we set up a pure bluffing scenario:

  • There is a $1 pot.
  • There are 2 players. Player 1 flips a coin.
  • Player 1 looks at the coin, which represents his ‘hand.’ Player 2 does not see the coin.
  • If it’s heads, Player 1 has ‘the nuts,’ the winning ‘hand.’
  • If it’s tails, Player 1 has the worst possible ‘hand’, loses to whatever Player 2 has.
  • Player 1 gets the option to bet $1, or check.
  • If Player 1 bets, Player 2 can call the $1 bet, or fold. If Player 2 folds, Player 1 wins the pot without ‘showing down’ the coin. If Player 2 calls, the coin is revealed and best hand wins the pot (heads: Player 1, tails: Player 2).
  • If Player 1 checks, the coin is revealed and best hand wins the pot.

This maps pretty well to a pure bluffing scenario on the river. You either have the nuts 50% of the time, or the worst possible hand the other 50%. This only covers whether Player 1 should bluff and whether Player 2 should then call. Player 2 doesn’t have the option to bet if you check, raise if you bet, and the bluff amount is fixed.

How should Player 1 play?

  • If Player 1 has the nuts, he (or she) should always bet for value. Why? Betting is always at least as good or better than checking:
    • If Player 2 folds, Player 1 wins the $1 pot (same as Player 1 checking first).
    • If Player 2 calls, Player 1 wins $2, the pot plus the called bet, better than checking first. Betting always has an expected value (EV) >= checking.1
  • If Player 2 has no hand, it gets more interesting!
    Suppose Player 1 bluffs when coin comes up tails:

    • If Player 2 then calls, Player 1 loses the $1 bet. EV: -1.
    • If Player 2 folds, Player 1 wins the $1 pot. EV: +1.

    Suppose Player 1 checks when coin comes up tails:

    • Player 2 checks and Player 2 wins the pot. EV: 0. (Only 1 outcome since we’re not letting Player 2 bet.)

When Player 1 has nothing, neither strategy dominates.

Here is a ‘pure strategy’ matrix.

P2 tight
Fold to any bet
P2 loose
Call any bet
P1 passive
Value bet heads, check tails
P1 EV: 0.5
P2 EV: 0.5
P1 EV: 1
P2 EV: 0
P1 aggressive
Value bet heads, bluff tails
P1 EV: 1
P2 EV: 0
P1 EV: 0.5
P2 EV: 0.5

There is no stable outcome to this game if each player sticks to a single strategy.

If Player 1 is aggressive, it’s better for Player 2 to be loose: he catches all the bluffs for $2, and loses all the value bets for only $1. If Player 2 is loose, it’s better for Player 1 to be passive: he always gets $1 value for betting, and never gets caught bluffing. In each cell, one player is better off moving counterclockwise to the next cell, and they chase each other around the matrix.

In the lingo, there is no pure strategy Nash equilibrium.

Now suppose each player can choose a mixed strategy. Player 1 randomly picks ‘aggressive’ pbluff% of the time. Player 2 randomly picks ‘loose’ pcall% of the time.

Now, if Player 2 calls 50% of the time, Player 1 is indifferent to betting or checking. Calling at random 50% of the time is an unexploitable strategy for Player 2.

Now suppose Player 1 bets 50% of the time he has no hand, and gives up the pot 50% of the time. This strategy presents Player 2 with a ratio of 2 value bets for every bluff.

  • 50% of hands Player 1 has the nuts and bets
  • 25% of hands Player 1 has no hand and bluffs.
  • 25% of hands Player 1 has no hand and folds, giving up the pot.

If Player 2 calls, 1/3 of time he wins $2 ($1 pot + $1 bet), 2/3 of time he loses $1. This has an EV of 0, same as folding. Player 2 is indifferent to calling or folding. Bluffing at random 1/3 of the bets (50% of time the coin is tails) is an unexploitable strategy for Player 1.

If Player 1 always bets for value on heads and bluffs 50% of the time on tails, while Player 2 calls 50% of the time, this is a mixed-strategy Nash equilibrium: neither player can improve by changing the mix of of strategies.

Player 2 breaks even on calls, Player 1 breaks even on bluffs, but wins $1.50 on average each value bet ($1 pot plus additional $1 50% of the time the bet is called). Player 1 gets $0.75 of the overall EV vs. $0.25 for Player 2. By bluffing, Player 1 gets 50% more EV per hand vs. only betting for value.  As an exercise, think about what happens to each player’s EV if one of them switches strategies.

So, that’s the picture of the problem and the Nash equilibria. Now let’s solve it more generally and get some more intuition for what the solution and P/L look like for different bet sizes.

Suppose we set up the problem more generally:

P: initial pot =1
S: bet size as fraction of pot
pbluff: probability of bluff
pcall: probability of call

Q: How often should Player 1 bluff?
A: Often enough to make Player 2 indifferent to calling or folding.

EVcall: 1+S when Player 2 calls a bluff. -S when Player 2 calls a value bet.
EVfold: 0

Setting Player 2’s EVcallEVfold

pbluff (1 +S) – (1 – pbluff) S = 0

Solving for pbluff:


There’s a simpler way of expressing this. Define ratio of bluffs to value bets as

o_{bluff} = \frac{p_{bluff}}{1-p_{bluff}}.


o_{bluff} = \frac{S}{1+S}

In our example, S=1, as bet = pot size; our game-theory optimal bluffing ratio is 1/2; We should bluff half as often as we value bet.

Q: How often should Player 2 call?
A: Often enough to make Player 1 indifferent to checking or bluffing.

EVbluff: P=1 when Player 2 folds. -S when Player 2 calls a bluff bet of S.
EVcheck: 0

EVbluff = (1 – pcall) – pcall S
EVcheck: 0

Setting Player 1’s EVbluffEVcheck

(1 – pcall) – pcall S = 0

Solving for pcall

p_{call}=\frac{1}{1 + S}

Plotting bluff ratio, call probability, and EV as a function of pot size:


Interpretation: When bet size (S) is close to 0 as fraction of pot, it is always worth it for Player 2 to call a small bet to get a chance at a big pot and to make sure Player 1 is honest. As S → 0, calls approach 100%. Best response for player 1 is to never bluff, and they just split the pot. As bet size goes up, it’s more costly for Player 2 to call a big bet to win a small pot and keep Player 1 honest. Therefore bluff frequency goes up as S goes up, and Player 1 gets a higher fraction of the pot. As S → ∞, Player 1’s EV → 1 and Player 2’s EV → 0.

If we do a 3d plot of EV against each player’s strategy with S = 1, we get this:


Here is an interactive version you can explore from various angles by clicking and dragging.

It’s a saddle anchored in 2 upper corners and 2 lower corners at EV=0.5. The four corners represent pure strategies of 100% aggressive/passive/tight/loose. At the Nash equilibrium of bluff ratio = 0.5 and call % = 0.5, the 2 axes along which each player can adjust his strategy form a horizontal plane: neither can unilaterally improve by changing strategy. However, if one player moves away from the Nash equilibrium on the axis representing his strategy, he becomes vulnerable to exploitation by the other: movement along the other player’s strategy axis can improve the other player’s EV.

If we change S from 0 to 1, we see something like this (rotated 90° clockwise from above):



Conclusions: 1) Poker is a pretty deep game (this wasn’t as quick a post as I thought!) and 2) This is a way to get some game theory intuition about one part of the game – bluffing and calling. If you’re into this sort of thing, I recommend the MIT online poker class mentioned above (this is a deeper dive into part of the last lecture by Matt Hawrilenko), and Mathematics of Poker, by Bill Chen, a math and finance quant at Susquehanna who has won a couple of WSOP bracelets.

This is just an analysis of a simplifed model of one street! And yet, heads-up limit hold’em was recently weakly solved. Computers simulated 2 players playing each other as well as possible for 900 CPU years, iteratively improving complex strategies (terabyte databases of all possible moves and responses) until they were provably so good that no one could have an advantage of more than 1 big blind per 1,000 hands. See if you can beat that near-perfect strategy!

No-limit hold’em with up to 10 players adds giant levels of complexity…No-limit bots are pretty good, but the pros still beat them.

Note: the first version of this post had a dumb error and some wonky graphs but I think it’s all good now. If you see any mistakes let me know!

1 We say that the Player 1 strategy of always betting when the coin comes up heads dominates the strategy of checking: it is sometimes a better decision and always at least as good. If it is always strictly better than an alternative strategy, we say it strictly dominates. In our case Player 1’s strategy of always betting heads sometimes leads to the same outcome as checking (if Player 2 never calls), so we say it weakly dominates the strategy of always checking.

Through the Looking Glass

Curiouser and curiouser! — Alice in Wonderland ZJ7jjMD

Well, having written about Greece the last couple of weeks, why stop now? Quickie post to get my thoughts in order.

TL; DR – Still seems possibly under-appreciated chance of bailout deal falling apart before it really starts, and likelihood of Grexit medium term.

Let’s review where the key players are.

Schäuble keeps agitating for Grexit. Rough justice. And yet one can see the logic and consistency of his argument. At some level he is fighting the good fight. The deal’s long-term success is mathematically impossible. The politics preclude a deal where success is possible. So an orderly Grexit may be best for everyone.

Varoufakis is no longer a key player, but he was saying essentially the same thing as Schäuble. Fix the Eurozone or pull the Grexit ripcord. He was ready to issue a new currency and seize the Bank of Greece, but Tsipras nixed him. He suffered from smartest-man-in-the-world syndrome: right about everything except the need to find common ground with others to arrive at a solution.

Tsipras is an enigma. He went along with Varoufakis all the way to the last step, which is sort of like going all in, getting to the showdown and then folding. If you’re going to get cold feet, better to do it before burning all the bridges. Like Stalin’s son, he can’t seem to shoot straight when committing suicide. Maybe he’s constrained by his allies, maybe he has some invisible master plan. But he sure looks like he’s in over his head and screwed the pooch.

Merkel likewise seems to have gone along with Schäuble down the Grexit road, only to chicken out at the last step. She offered just enough that Tsipras would have to pull the ripcord himself, and he didn’t. I suspect she doesn’t mind triggering Grexit, she just doesn’t want to be blamed for it.

Lagarde seems to have missed a trick. Or the IMF is at war with itself. She went along with a bailout that seems to require further billions from the IMF. And yet, there’s doesn’t seem to be any way the IMF board can approve it. You can’t lend tens of billions of euros if there’s no sustainable way to pay them back. And the IMF staff and pretty much everyone agrees it’s unsustainable. Are they going to walk away after Lagarde agreed? Are they going to waive sustainability under pressure from Obama, Merkel, Hollande et al.? I wouldn’t bet on it.

Draghi seems to be doing everything in his power to help Greece, having stood up to Schäuble in the Eurogroup and already loosened ELA. If Greece goes, who knows who is next. He’s Italian, and US-educated, so he follows a pragmatic line toward the periphery.

The question of IMF participation looms large. Then, it seems unlikely the banks can actually re-open and lift capital controls, without Greeks rushing to empty them and move cash abroad.

Then, there’s the question of what happens next to the Greek economy. If the deal moves forward, the government and others pay all their overdue bills, the ECB fully stands behind the Greek banks and they reopen ahead of a possible restructuring, Greece could start to normalize.

But then, tax hikes and spending cuts will start to materialize. The deal is supposed to have automatic triggers that increase austerity if budget targets aren’t met. (Isn’t that special, a hard-wired pro-cyclical fiscal policy.) It’s like a doomsday self-destruct button for the economy. There seems a distinct possibility of a runaway collapse.

So, there still seem to be perhaps under-appreciated prospects of the deal running into trouble. And little prospect of long-term sustainability. And some risk of having the economy go south and needing another bailout earlier than expected.

And maybe new elections will bring an even further-left government, or bring neo-fascists into a government, or blood in the streets.

No one is willing to make the sacrifices that would allow Greece to recover within the Eurozone. But no one wants to be blamed for Grexit. The result is an unsustainable deal that will lead to more pain, and presumably Grexit once the alternatives have been exhausted. I was wrong since I thought inability to reach an reasonable deal would lead to no deal and Grexit. (and Schäuble and Varoufakis seemingly playing for Grexit.) The surprise is the determination to do an unreasonable deal that solves nothing.

In behavioral economics terms, each side frames leaving the euro as a policy defeat. And yet, the point of the euro is to work together for peace and prosperity. It seems misguided to hang on to it when it is doing the opposite.

Austerity has led to economic collapse. Ratcheting up austerity is not going to fix it. Fanaticism consists of redoubling your efforts when you have forgotten your aim. — George Santayana

It’s a deal no one seems to want. Greece’s political will to implement a tough program is questionable at best. Even if they had the will, the numbers don’t add up.

Eventually, Stalin’s son got a prison guard to shoot him.

27 queries in 1.406 seconds.