I'm struggling to see how the advice applies to their own examples. Take Jane, who took a great offer but was hit by a recession:
- potential reward of the decision: checks out
- probability of getting it: 95% if you already have the offer in hand
- resources (time/effort/money) to bet to get the reward: not significant
Then there is 'imagining a bet' and 'analysing your past failures'. The first one might lead you to the 'what if the company goes bust?' question, but would that really help in the decision making or be a reason to not accept the better job? The unlucky event is a recession, not this particular company doing badly (i.e. no signals to see).
If a recession hits, there is no reason to believe the consumer goods co. would fare any better and her old job more secure. I'd really like to see the 'poker player' mindset at play but this just feels like a complete miss.
I think like a poker player, I skimmed the article. What they seem to fail to mention is that variance is important as well. In poker you need to play at least 10000 hands in order for variance to average down a bit. 10000 hands is still very little.
So, if you have a 95% chance of going 10x that's great, but the question is: can you handle a 5% chance of having being potentially ruined for at least a decade? If not, then you cannot take the bet despite the fact that EV is super high.
Put more extremely, if I offer you a 99% chance to take make one billion dollars (legally) and a 1% chance of being killed, would you take it? I definitely wouldn't.
How about the following? I offer you an unfair coin flip advantage of 51% versus 49%, we're going to flip 1 million times, each time the bet is 1 dollar. Would you take it? I would.
The heuristic that poker teaches is: take a lot of small bets with huge upside with little to no downside. It doesn't matter how slim the chance of winning is. Slightly more nuanced (still a heuristic): the chance should be about as big as your volume of chance-taking. Say, you take 10000 chances, then the slimmest chance you can take (on average) is 1/10000.
This is off-topic, but the mobile wikipedia link you posted is surprisingly nice to read on desktop. Seems a lot more pleasant than the wide lines on the desktop site.
This comment comes up repeatedly on HN. While I somewhat agree, I would also suggest to not maximize your browser window by default, to avoid wide lines. I usually have my browser window square-ish.
> Put more extremely, if I offer you a 99% chance to take make one billion dollars (legally) and a 1% chance of being killed, would you take it? I definitely wouldn't.
I would even if it offered 50+% of getting killed. The small chance you strike a billion is worth the risk.
Sure but at some point the money gets too low to risk it even if you are in the dumps, no? Would you take 1 million for 50%? 10k for 50%? 10 bucks for 1%? 1 dollar for 1%?
At some point it goes from a life changing amount of money to just "meh" money.
Any job that earns money has an implicit acceptable $ for death risk, because all jobs have a risk of death (proxy measurement of injury using ACC rates in New Zealand).
Your sum risk might be reduced (e.g. your job pays health insurance), but many jobs increase your risk e.g. just driving to work is dangerous.
Of course your primary finite resource is usually time (years of life), and your object function usually relates to quality of life (not $) . . . Win $ for risk of death is a poor metric.
I wouldn't do it for 50%, I can see how some heuristics would get you there.
- Right now we spend 33% of our life working, nd the stress of work impacts another 33%.
- A billion is also way more than out life expected income, but there's a big marginal reduction of value to money. But you also get it now when you're younger rather than parsing it out.
- As a random 25 year old male, you would only have a 1 in 16 chance of surviving to retirement age at 60 on your own.
I personally would only take the deal for 2.5% chance of death though.
I find that fascinating. I'm not even a millionaire, yet money is rarely a constraint for me and life is great.
The only reason I can fathom desiring a billion dollars is the leverage it would give me to influence the world in what I think is a positive direction. But when it comes to money for my own sake, $10M is about the most I can imagine having use for.
It's a fun thought experiment under the many-worlds interpretation of quantum mechanics. Even at an infinitesimally low chance of getting the billion dollars, if the many worlds interpretation of quantum mechanics is true, you could argue that it is worth it from a selfish perspective.
In all the worlds where you get the money, your life is improved. In all the worlds where you don't, well, you're dead, so you don't really care.
Obviously, this doesn't take into account the impacts to everyone else in the worlds where you died.
I don't think the advice doesn't apply (they still claim Jane made a good decision, even though it led to a bad outcome: their point is that Jane should keep making good decisions, and not worry about bad outcomes because you can't control luck — if you mostly make good decisions, luck won't matter in the long run), it's just that it's pretty much useless because
> "Probability of getting it — you cannot know the exact probability, but estimate it the best way you can (remember the bet with a friend)"
is the hard part. Jane's case was clear (good decision), but what about Mike? Is there a proportion of his savings he should have invested that would have turned his "bad decision" into a "good one"? We have no signal to tell us what the probability of any cryptocurrency going up or down is over, say, 6 months (they've all gone up and down completely arbitrarily). Basically, finding the probability of any particular "success" (or actually, a failure) will easily tell you how much you should risk.
And that's what it's like with the most of life decisions as well: we don't know the probability — not even the ballpark range. Likelihood that any one person you meet is going to be your life-long partner is basically nil: but we still invest in building relationships before fully committing to either decide they are not, or increase the probability that they will be. But we still do that relationship building investment based on very few signals (appearance, short chats and potentially what social circle someone is from if from a shared acquaintance).
There is no probability to know in the sense you mean -- probability is what we call our personal assessment of the situation, incorporating the (little) knowledge we have.
So sure, there is your probability. Will you make money knowing it? That's the question you're alluding to.
(As a concrete example: there is no "true" probability of rain six days from now. Either it rains or it doesn't. But a skilled meteorologist can give you a probability that would probably make you a little money over the climate average.
Both the climate average and the meteorologist's assessments are correct probabilities. They just reflect different amounts of information.)
I was not referring to discreet outcomes at all (the old joke about everything having a 50% chance: it either happens or it does not).
My point is that with the available information when making most life decisions, we usually have no idea at all of even the ballpark chances of something turning one way or another.
Basically, the advice from the article is to incorporate luck into your estimations (both good and bad), and to use that to determine if a decision is good or bad, and then only go with good decisions (like a poker player would). And don't stress over bad (or good) outcomes if they were mostly due to really bad luck (i.e. something improbable has happened).
But if you don't know if chances of something happening are 10% or 90%, how do you incorporate that into your decision making?
What I am saying is that in life, we implicitely work to reduce the range (eg for a romantic partner, get to know them much better), but we've already decided to invest that much time with very slim chances of them turning out to be our lifelong partners.
I've also heard this topic referred to as "results-oriented thinking," which is generally what you don't want: you shouldn't judge a decision based on its result, but on the information you had at the time.
The key to this idea, which I don't see covered in the blog post, for making future decisions is that you shouldn't let past bad luck affect your future decisions. If you lose a positive EV bet, it shouldn't shy you away from making the same bet again.
Some examples off the top of my head about decision-making traps people could make:
- Continuing to bet at the roulette wheel to "regain" what you've lost
- Not going to a well-rated restaurant again just because you had a random bad experience
- Not investing in ETFs after being burned by past downturns
- Playing MTG and not burning them out just because they had a counterspell last game
> Not investing in ETFs after being burned by past downturns
Helloooo Japan!
ETFs are great, it will never happen to the US economy :)
In other words, based on market behavior from multiple countries, it is definitely a possibility that ETFs won't return much in a period of 30 to 40 years.
And what's your alternative? Cash under the mattress gets killed by inflation. Gold has its runs but usually underperforms. Bonds also get killed in a downturn.
However, the example is simply wrong (IMO) since it assumes that within a 30 year timeframe you'll have gained 8% on average (adjusted for inflation). If this would be true, thne yes it would be a decision-making trap. I think that's what morley is getting at.
I'm arguing it's a tough sell that the S&P 500 works like that. If one would agree that the S&P 500 might not continue to give 8% ROI on average over a 30 year time frame, then one might consider doing something else with their money. For example, maybe it's more fruitful to invest in yourself to upskill even more rather than putting your money into the markets. I don't know I haven't researched it, I myself try to beat the market, it's a fun endeavour. The jury is still out.
That's a good point, betting on the world is a more viable strategy. That is at least, if the world is growing in an economic sense. With that said, since the world population will grow, I'd be willing to make that bet. In this case, I'd say the example holds up.
The point of investing in an ETF is not exactly to bet that the economy will grow no matter what, but to minimize variance and regret with the money you had at the beginning (i.e. you may end up in the red, but in terms of expectations, your likelihood of being in the red was no worse than the average).
Exactly - for me, investment is less about "making it big" and "beating the market," and more about, "storing the value of the work I did to earn this $X."
Such that, if the "burn rate" for my household today is $60k/yr, and I store $60k in investments, then when I withdraw at some future date $X will still be enough to support my household for a year.
The revenue of the S & P is 50% global so pay attention that 1) US ETFs already have huge global exposure and 2) many country ETFs have global corporations in them
There is no pure financial alternative. But a reasonable alternative (or complement) is to invest some money in projects/investments that will for sure have a worse expectation than ETFs, but simply bring you joy in life or new learnings.
Physical cash is terrible but to an investor, "cash" tends to mean something like t-bills, which over the past century have done slightly better than inflation. Gold is horrible if you have too much of it, but it tends to do well in equity crashes, so mixing in a little can be helpful. Bonds also sometimes do well when stocks go down, though I think that's more likely when you start at higher interest rates. Commodities can have 15-year bull markets independent of everything else. An investor in Japan would have done fine if they had a lot of foreign stocks. Etc.
> I've also heard this topic referred to as "results-oriented thinking," which is generally what you don't want: you shouldn't judge a decision based on its result, but on the information you had at the time.
Had this happen at work recently and shapes why I was ok with the outcome even though it wasn't what we wanted.
I was building an integration and got it working and trying to push to Prod when I realized the integration with the 3rd party required a security audit... Given the information at the time, it was the right thing to build but didn't do deep enough research. Lesson learned.
Results-oriented thinking is just one of many human cognitive quirks that are often maladapted to the modern world.
A major one is how we tend estimate probability. We naturally do that by equating it with how easy it is to imagine or remember it happening. This worked well in a tribal situation where our world was very small...if Grog got attacked by a tiger, there's a good chance that tiger attacks are a serious danger that I should consider. But this breaks down in the face of a global society of billions of people and a media that profits from making people afraid. Rare events get magnified by media attention and feel, to us, like they're incredibly likely. Plane crashes are incredibly rare. Even in a plane as fundamentally broken as the 737 Max was, passengers were safer traveling that way than by car. But since every airplane crash is covered extensively on the news whereas car crashes rarely merit a mention, people's primitive cognitive quirks kick in and they're more afraid of flying. You see it with mass shootings too. Even if you get 1-2 per day in the US, that's still maybe a few hundred people per day who are directly affected. Over the course of a year, that's roughly 70k people out of 330 million, or %0.02 of the population. Which isn't to say that we shouldn't do everything we can to prevent it, but when you hear people saying that they don't feel safe, that's entirely on the media and how it's warping our perception of the danger rather than it being a real threat to our lives. We also saw it with the spate of Asian attacks that became a media favorite. There was a story of Asians in New York starting to contort their lives to avoid becoming victims. But when I looked up the actual statistics, there were 20 such attacks at the time in a city with 1.2 million Asians. Which, again, doesn't mean that we shouldn't be treating each of those attacks as terrible and be doing everything we can to stop them, but when you consider 20 victims out of 1.2 million, that's just not a risk that's worth going to considerable inconvenience to avoid.
There was an interesting TED talk many years ago from a guy who specialized in these kinds of cognitive quirks. And he discussed the findings of one particular study that always stuck with me. In it, participants were presented with one of two hypothetical situations. In the first, they were going to the theater to see a movie and they pre-purchased the ticket for $20, and also brought along a $20 bill. When they arrived at the theater, they found that the ticket had gone missing. Almost all participants in the study said they'd turn around and go home rather than using the other $20 to purchase another ticket. In the second case, they were going to the theater to see the same movie, but instead of having pre-purchased a ticket, they were intending to buy one when they got there. But they had, instead, brought two $20 bills with them and managed to lose one on their way to the theater. Almost all participants presented with this scenario chose to use their remaining $20 bill to buy a ticket. Despite the situation being basically identical, when you're presented with both scenarios, we have a cognitive quirk where we don't want to pay for something twice whereas we can rationalize the loss of money that we never spent. But what got to me about that study wasn't the results, but how intuitive both decisions felt to me. I could feel how unpleasant it would be to buy a second ticket and, yet, how easy a decision it would be to buy a ticket for the first time even after losing some of my money. The lesson, for me, was that while I've always thought of myself as a logical person and someone who would always let statistics or other scientific basis guide my thinking, it's still really hard to follow through with that.
Do you drive a car? I am asking because statistically speaking, it's the number one killer of people from age 21-65. Very few drivers go their whole life without getting in an accident.
Statistically, on any given day, the odds of being in an accident are quite low. However, if you drive for 30-60 minutes every day, you are exposing yourself to a low amount of risk consistently over time, which inevitably means that one day, your number will come up.
Most people don't have a background in probability or they haven't taken the time to look at driving, but for those who consider themselves normally "rational", they're often very surprised to see just how dangerous driving a car really is, but they just weren't aware of the amount of risk.
Analyzing the risk of driving every day is an exercise worth doing. That being said, I have a fairly high level of risk tolerance, so I drive to work every day because I live just a little too far to reasonably walk, and riding a bike would be a higher risk than driving because of lack of adequate road shoulders for much of the route between my workplace and home.
> riding a bike would be a higher risk than driving because of lack of adequate road shoulders for much of the route between my workplace and home.
Is this an empirical statement or based on the same gut feeling that makes people feel safe in a car?
I'm asking because when you actually run the numbers, in most cases I've been involved in, you get to be very surprised to see just how safe cycling really is.
I don't know what the culture and road type is around your neck of the woods, but I would prefer a road without a proper shoulder most of the time -- simply because the existence of the shoulder provides very little actual safety, but it does give a sense of safety that makes drivers behave more dangerously near cyclists. Not to mention lower quality road surface, collected debris, etc.
I've found that frequently bluffing in life pays off. Generally you don't get caught very often and the cost of getting caught generally isn't very high, even if the stakes are. You have to be prepared to take your way out of sticky situations though.
People who describe themselves as super honest usually are self-deprecating. They tend to under value their knowledge and experience.
The reality is, 95% of scenarios don’t require what is asked for, and stating your capabilities in the most generous way is the optimal decision. Just back it up with work.
This is another example I'd use of a bluff: basically saying that you can do a job even though you've never done it before. If you're good and confident in your abilities, it all works out in the end. Everyone wants ridiculous experience but at the end of the day they're just trying to make sure you can do the job.
Not OP either. Here there aren't turnstiles to access the commuting train, but there are controllers on board from time to time. With seldom use, it's just better not to buy a ticket (read: to bluff).
Numbers made up and simplified but in the same ballpark:
Round-trip ticket across town: 10€
Monthly pass: 50€
Fine: 70€
The chance to be controlled is roughly 1 in 10 for a round-trip. So if one did 5 round-trips in a month, on average one would pay 70 x 5 x 0.1 = 35€. Which is a good price with acceptable variance fur such usage. When caught just pay without fuss.
At some point (between 7 and 8 round-trips) it's more cost effective to just buy the monthly pass, as the average cost of not buying approaches/goes over such threshold.
As much as "fake it till you make it" is a much hated approach, there's a grain of truth to it. Many people only got the chance to prove their merit by bluffing (faking?) it first. Sure, they had to ultimately deliver on the expectations, and perhaps there are more of those who faked it and then failed to deliver. It doesn't change the fact that often you don't even get the chance if you don't project an aura of confidence.
Generally speaking, being confident gets you to places. This really applies to every aspect of life.
If you act like you belong somewhere and that there can be no questions about you belonging there, people in general will not question your presence. If you seem lost, or confused people will be curious why.
Walk into a private event acting like you belong, and there are good chances nobody will realize you shouldn't be there. Walk around looking all confused and you are likely to be asked to validate your presence.
Naturally, this doesn't apply only to places, but also groups, communities, companies...
The first thing that comes to mind is bargaining. I always claim I have a better choice elsewhere and that I need a discount to close the deal. So I was renting a flat in SF in 2010 and claimed I had a better offer elsewhere and got 5% off in order to take their flat instead.
I think people who tend to “bluff” often also tend to seriously underestimate how often they’re “caught”. Just because no one has called you on your bluff to your face doesn’t mean you haven’t been caught.
Bluffing in poker doesn't mean something silly like going all in with 72o. It can be something like playing a relatively weaker hand more strongly because you've evaluated the range of hands that your opponent can have and the range of hands that you could have. Likewise, in real life, bluffing doesn't have to mean outrageous lies.
So in a requested example below I give "I was renting a flat in SF in 2010 and claimed I had a better offer elsewhere and got 5% off in order to take their flat instead."
How do I get caught? If they say no I get up, pause then change my mind.
I've never suffered from being caught by a bluff, but I structure them sensibly.
Everyone else who rented a flat negotiated 10% off. Your landlord guessed you were lying so altered subsequent planned rent increases to more than make up the difference - because they don’t like you.
It surprises and slightly delights me to see how often people who try to ‘game’ others, queue jumpers, and others who display selfish behaviour are then given special (negative) treatment from those who notice their behaviour.
Some seem to go around thinking that people are awful, but really people are just awful to _them_. People are generally kind and help each other.
The main problem that I have with lying regularly is that in order to lie effectively, you need to somewhat believe your own lies.
Eventually, you really do believe the lies you tell and then your ability to see the world for what it is is diminished.
Additionally, when dealing with other people, we tend to fill in the blanks about the things we don’t know about them using information we have picked up from other people as a template. We know ourselves better than any other person, so we often use our own motivations as a stand in for the motivations of others. If we are liars ourselves, we assume other people are liars too, by default (and, due to confirmation bias, once we look for evidence of this, we find it everywhere).
This mindset means we end up trusting people less than is probably optimal, but we are unaware of this fact.
Bluffing is indeed lying, but in a fairly innocuous way.
Say you're a physical coward and you're being picked on for a fight by someone. You can puff your chest up, put your shoulders back and put a sneer on your face, possibly scaring the bully off.
Yes you're misrepresenting yourself. You're lying. But how exactly is that corrosive to your character?
I'm a serious but hobbyist player with ~15k hands last month.
The thinking like a poker player is mostly about being upfront about your risk tolerances and then having a culture supporting people who make the best risk adjusted decision, even if it doesn't work out.
Expected Value is complicated because a 50% chance for $100 is the same as a 10% chance at $1000. It's the variance, not the EV that makes a lot of decisions hard.
You (and businesses) need to decide what risk is acceptable, communicate that clearly. Reward people who manage risk in a way that's aligned with the business even when it doesn't work out. Get rid of people who are either take too big risks, and people who don't take risks.
I also like the discourse on bankrolls and bets sizing. I think being able to correctly size your bets is such a great, albeit challenging skill to have in the real world.
One of the rules of thumb you quickly learn when studying the Kelly criterion is that overestimation can lead to ruin, underestimation can at worst lead to slower growth.
The golden days are sadly gone. It has pretty much all been dried up after the US crackdown.
Before that, sites were full with very loosely playing Americans who viewed a loss of a $100 just 'entertainment' or equivalent of a night out.
With that all gone, you have tons of very tight Chinese and Russian players these days.
Lots of hunters and not a lot of moose.
As someone who used to live with a pro poker player that might be the beginning but it really comes down to the psychological aspect of being able to disconnect and deal with losing a TON of money in the short term. You need to be able to quit, consistently, after losing massive amounts and then come back and do it all again the next day. Over and over. And the losing streaks can be long.
I think you meant a 50% chance for $100 is the same as a 5% chance for $1000 right? Just... funny math error as I was like, wait, no... I'll take the 10% chance at $1000 every time! (unless I'm desperate for my next meal, then, as you say, the lower variance would point to higher certainty of some money vs max expected value)
EDIT: I see someone else noticed this further down than I looked in comments, sorry for dupe.
If your bankroll is small enough that Kelly vs naive EV calculations are giving you significantly different answers for what you should do, you are probably playing at stakes higher than you should be (or you're playing in a tournament, in which case the optimal strategy is indeed different than in a cash game).
Pot odds are just the ratio of the current bet to size of the pot.
If you have not yet made your hand (e.g. only 4 spades) then you consider your drawing odds, which is the chance of making your hand.
But say there is a pair on the board. Then you need to consider the probability that your opponent has or will make a full house, which would beat your flush if you were to hit. Likewise the probability that someone will have a higher flush than you.
Analyzing this in aggregate is what gives you the expected value, which is what poker players consider.
If you are simply playing a cash game with unlimited buy-ins. Then assuming your overall bankroll is sufficient, you will always make plays with positive expected value, even with high variance, because it is a continuous game and you expect the total return to be positive over sufficient iterations.
If you are playing a tournament with a set number of places in the money, then you need to take into account the variance as well as the expected return because it is an episodic game and the types of plays you will make will depend on not just expected value of the single play but what place you are in currently and the number of players left.
And probably the most difficult to estimate are the implied odds. If you do make your draw, how much money can you expect to make in later betting rounds.
Implied odds are interesting in that they can make very bad hands more playable because if you do hit, then the likelihood that your opponent believes that they are winning is higher (because the chances of such a bad hand being played rather than folded immediately should be low) and thus the probability that you can elicit more money from them is higher.
Small pairs and low suited connectors are good examples of these kinds of hands. Your pre-flop chance of winning with a pair of 2's on a full table is pretty low. But if the flop is 2-K-A, you will likely get a lot of action.
The non buzzwordy way to express this is to use some basic concepts in probability to make accurate decisions in the face of randomness. Specifically when making a decision you should think about the expectation[1] of the outcomes overall and some properties of the distribution. This requires you to think through what all the outcomes are and what you think their probabilities are.
1)All other things being equal you should generally prefer a higher expectation decision to a lower-expectation one
2)But you have to avoid decisions which have outcome distribution properties you can't tolerate. Most obviously you should generally ensure that your risk of ruin is zero because even if the probability is very low, that outcome is close to impossible to recover from so must be avoided.
3)For two possible strategies with similar-ish expectations in real-life terms you may well want to choose the one with the lower standard deviation of payoffs. Like say you're choosing between an offer at medical school to train to be a dentist and pursuing your long-shot idea of going to Hollywood and trying to make it as an actor. Imagine that when you look at the outcomes you estimate that in acting you have an insanely low probability of making it but a huge payoff if you do and in dentistry most people do pretty well but no-one's partying on superyachts with Jay-Z and Beyonce. Well even if the expectation of acting works out slightly higher, if they are close enough you should probably pick dentistry because the variance of outcomes is just way lower. If you think about your future as a monte carlo simulation you want to end up "ok" on most of the paths in the simulation even if that means giving up on some "lights out" outcomes.
4)most people agonise the most about decisions where the expectation is really pretty similar so it just doesn't matter that much either way. So don't beat yourself up about whether you made the right decision - just try to learn from the decision and move forward
A huge mistake people make is to evaluate the quality of the decision based on the single possible outcome that crystalised into reality by actually happening rather than using the framework above. So resist that temptation and instead evaluate your decision based on whether you chose to maximise expectation while avoiding risk of ruin and excessive variance.
[1]In the sense of being the weighted average of all possible payoffs where the weights are the probabilities and the payoffs are the utility of each outcome (usually just in cash terms).
The author sounds like a losing poker player. If my calling frequency is based entirely on pot odds, a good opponent would just start bluffing me out of every pot by increasing their bet size to a point where I can’t profitably call.
But note that unexploitable, game theory optimal strategy in poker does not differ based on the opponent. It’s not just based on pot odds, but on the whole sequence of actions and cards. Still not a very good article, though.
There are only two ways they can be losers: they are not GTO or the rake is higher than their winnings. Hold’em is pretty much solved (almost because approximations were used for bet sizing). It is possible that you can do better by deviating from GTO to better exploit bad players but that makes you vulnerable to be exploited yourself.
That's sort of a no-true-scotsman argument: They lose because they're not playing true GTO.
Well, yes, but the point is that a fully GTO non-exploitable strategy is not solved for Hold'em. At least not outside of limit (and kinda no-limit) heads up play.
(There is also an argument that Libratus actually just approaches Nash equilibrium and isn't fully there, since there are 1e+160 decision points in a heads up no-limit game, but compared to human players the difference is probably meaningless. )
The closest we have seen for larger games is Pluribus - https://www.science.org/doi/10.1126/science.aay2400 - but the researchers there aren't even attempting to say they have solved 6max, just that they could be less exploitable than some of the best in the world. It is not particularly useful for online play because it does not try to counter variance or manage bankroll, and in further results it began to lose quite heavily as the pros adapted to it, losing 700BB over the 10k hands.
By no means am I saying GTO-based strategy is ineffective - I'd probably be wasting my DTO subscription if I did think so ;) - but we're only part of the way there and the missing bits mean that you can't play perfectly unexploitable poker.
We have no idea what kind of player they are because the article isn't about being the best poker player. I don't recommend making a habit of bluffing in real life. Unlike in poker and certain television shows, people aren't likely to respect you for it and it may expose you legally.
> a good opponent would just start bluffing me out of every pot…
People try this.
> …by increasing their bet size to a point where I can’t profitably call.
A good poker player will adjust your expected range based on your sizing and frequency of bets/raises, and they will adjust their calling/raising ranges accordingly.
Most poker games, especially big bet games are played against a range/distribution of hands, not a specific hand.
Yes. The article makes its analogies with a "simplified" view of poker. I don't think the author's point was to go deep into the game. It's enough to describe the basics like bankroll management, pot odds and expected value, and then show real world examples of it.
I'll also point out that constructing ranges and thinking in terms of ranges/ hand distributions is a relatively new development in poker. I mean, the very good players were subconsciously doing it all along, but it's only been written down and analyzed in the last 10 years or so.
> but it's only been written down and analyzed in the last 10 years or so
I agree with you overall points.
For historical reference for folks who are interested in the poker boom of the 00s but were not there, it’s more like 20 years, if not more.
The book Let There Be Range was published in 2008.
High stakes players were discussing the issues covered in that book amongst ourselves (live and online) with quite a bit of sophistication for several years before that.
All that said, and supporting the point made above, advanced computational support for ranges (esp. via “solvers”) didn’t really start becoming a thing until the early 2010s.
> Annie Duke has cashed 39 times at the WSOP and has won at least four million dollars at tournaments alone.
This statistic is almost certainly not counting tournament buy-ins. It doesn't represent net profit. If you want to evaluate somebody's performance as a poker player, you want to look at net profit (among other things). Literally every poker player will have some wins, so if you only count the wins without counting losses, it will sound impressive but not actually mean anything.
> has won at least four million dollars at tournaments alone.
This is most certainly not true, because it doesn't subtract the cost of all the entry fees, to the tournaments she cashed and the dozens or hundreds she didn't.
"alone" implies she has won additional money in non-tournament play, when there is no evidence she is a winning cash player.
Finally, there is a LOT more to the Annie Duke story. Morally, ethically, business-wise. A winner? A little research goes a long way.
> This is most certainly not true, because it doesn't subtract the cost of all the entry fees
It is fairly standard in the poker world to discuss total winnings in tournaments rather than net winnings.
When discussing net winnings, it’s usually discussed using an ROI measure (e.g., 30% roi over their past 100 MTTs between $1000 and $5000) rather than just dollars, since the biggest winners in dollars are almost all winners of super high roller tourneys like the Big One for One Drop.
Furthermore, live MTT results in general are often skewed due to the unmatched (?) softness of the $10k wsop main event for a field that size.
With regards to Annie Duke in particular, I never considered her a “good” pro poker player. That said, she was pretty good at fleecing amateurs, and she was able to play a tight game when she was outclassed.
An additional comment that I copied from another reply:
With regards to Annie Duke in particular, I never considered her a “good” pro poker player. That said, she was pretty good at fleecing amateurs, and she was able to play a tight game when she was outclassed.
It goes both ways. If you increase your bet size too much on bluffs, it becomes profitable to just wait for a good hand to call you with and fold everything marginal. The bigger your bluffs, the fewer I need to catch.
You'll also become predictable if you only overbet when bluffing. If you try to deal with that by making all your bets larger, then you lose out on value from your strong hands because I can now profitably fold some hands that I would have had to call you down with otherwise.
Put another way: you can't exploit someone who makes decisions based on pot odds just by changing your bet size, because changing your bet size changes the pot odds.
I attended a talk by Richard Garfield, the game designer behind Magic: The Gathering. He made a point that adding randomness to a game can be a way to give it that property of "easy to learn, hard to master", and in turn make it easier for a community to grow around the game. When you add randomness, it becomes possible for novices to sometimes beat more experienced players through sheer luck. Contrast that with e.g. chess, where it doesn't take much skill difference before the most skilled player is almost certain to win. So randomness makes it easier for newcomers to get the occasional win, to keep them motivated while learning. But at the same time, randomness can make it harder to master, because the cause-effect-relationship between good decisions and good outcomes becomes blurred, like the article points out, defeating the trivial "reward function" we usually use for evaluating our performance. It takes a lot of discipline to say "I won, but it was luck, because I actually misplayed" and learn from that. And on top of that, high level play will become a game of analyzing and estimating probabilities, minimizing or maximizing variability depending on the strength of your position, basically embracing randomness as a gameplay element to be understood and sometimes even manipulated, despite being so intangible compared to most other gameplay elements. So used the right way, adding randomness to a game can both lower the barrier of entry, and raise the skill ceiling at the same time.
This is an interesting comment, and I'd add one other element to this, which is that randomness is way more realistic, if we think of games as something intended to represent the real world in some way.
In real life the best marksman can get wiped out on the first day of a war, the best business strategist can get done in by a once in a generation disaster, and a hapless idiot in business can get an early big break. And so on.
I feel like it feels much more natural and intuitive to us to play games that are a mix of skill and random chance.
That depends on whether you think that games should be simulations. Even if you think they should be simulations, you still might only want to reward optimal play over lucky play. If I'm using a simulation to train someone, I only want them to be rewarded for ideal play.
This is incorrect, as far as the research into training for expertise goes. You actually want simulations to be like the real world: complex, ambiguous, sometimes with delayed feedback and presenting multiple forms of the problem. Of course, it also takes a good mentor to debrief with and untangle what was bad luck and what was bad decision making. (See the Oxford Handbook of Expertise if this stuff interests you.)
An alternative to randomness is handicap. For example, Go game has ranks and based on them either you or your opponent get a handicap. Equal ranked players don't use handicap. It is counted that a stone of handicap is worth ~6.5 points, and that is what's used in handicap - one player gets additional stones placed at the beginning of a match. The system is aiming to provide you with 50% win/lose percentage, and every match to feel equal.
The issue with a handicap is that if you win you don't feel as good because you know that the only reason you won was the handicap. It's harder to draw such a conclusion when you win due to randomness.
Agreed. I've seen comments in similar threads in the past where people rag on games like poker or backgammon because they believe the randomness factor moves them from skill-based to luck-based.
In truth, managing randomness is actually one of the core skills of the game. Not just on the decision in front of you, but on future decisions as well. It's a hard skill to master. Moreover, the timelines to evaluate one's skill are long, sometimes hundreds or thousands of games, incorporating tens of thousands of decision points. It's not a wonder it's often misunderstood.
> But at the same time, randomness can make it harder to master,
Except that Magic: The Gathering is particularly bad for locking into specific "meta" (metagame) which then persists until new cards come out/old cards rotate out. Hardly random and hardly difficult to master.
The best MTG players are really good at what I call the "game mechanics corner cases". The know all the weird ways to play and resolve the cards--this gives them a quite persistent couple percentage point advantage.
However, it won't give them enough advantage to beat the "meta".
When you put it this way, it sounds like you are referring to plain deception, and not actual game-theoretic bluffing.
Bluffing is about making non-optimal choices for the express purpose of making it more expensive for someone else to make optimal choices. If someone knows you're engaging in bluffing, that's fine, because by bluffing you're still forcing them onto non-optimal lines.
On the other hand, you would prefer people not to know that you're engaging in deception.
Well you can play millions of poker hands online. Life isn't like that. You have only ~30 years, and you have 30 * 365 = 10950 hands if we count each day as a hand.
So what you need to work on is how you define "success" in life. Money is a high variance objective, whereas self-improvement is a low variance objective. I think finding a good balance of variance and tolerance to risk is a key to happiness.
Slight offtopic, but this is why the "if you try hard enough you will succeed" advice sounds like BS to me. Even if the most dedicated entrepreneur can only try so many startups in their lives. If the odds are vanishingly small then success is far from guaranteed.
Rejecting the role of luck in our outcomes is delusional, and people who support it have something to peddle (if only their own personal brand as thought leaders).
If you're good, your profit depends on the number of hands you get to play. And that number of hands you get to play is based on your bankroll.
In the non-poker real world, most people maybe get to play one hand in life, and that’s it. One at-bat in a baseball game, to use a different analogy. You dump 20 years of savings into buying that laundromat, and that’s your only shot of making it. If we are really lucky we might get one or two more hands to play but that’s it for the vast majority of us.
The rich get as many hands/at-bats in life that they want to play, so they will eventually win one. They can just serially start business after business until one of them lucks out and succeeds, then do talks for the rest of their lives on how to be good at business.
Alternately, if you find yourself in the situation where you were not born rich then that should influence your strategy. Apply the Kelly criterion and carefully make a number of appropriately sized bets based on your currently available capital. Don't dump 20 years of savings into anything, but carefully build up investments over time.
Measuring your net worth against those born rich is a fools errand, but you can definitely make your children have a better shot against their children. True wealth is almost never built up in a single generation, despite the fairy tales about the "American Dream" that we tell our children.
Yeah, it’s more like “if you don’t try hard you’re unlikely to be successful” combined with “you shouldn’t underestimate your chances of having some kind of success if you put in some grit”.
High side? If I am understanding OP correctly he says one only has 30 years to "make it in live" or you're fucked and if I am understanding you correctly you think that is too high. If that is the case both of you are wrong.
Probability distributions are complicated. Even determining what kind of distribution you are look at is difficult in many cases. Such distributions are also skewed in real life by things like insider information (in poker, that would be cheating). Even so the list is rather intimidating, assuming fair play:
If financial institutions (and crypto players) are using these kind of approaches to make their bets, then isn't the individual investor hopelessly outgunned in the vast majority of cases? Plus, not having a big pool of capital to absorb temporary losses makes that situation even worse.
Investment capitalism, in other words, is just a casino for the uber-wealthy. Letting it rule the economy is a serious mistake.
It's even more complicated than you think! The theoretical distributions you list are just that: theoretical. No real-world process generates values exactly according to a theoretical distribution.
So you're not even dealing with a somewhat countable list of distributions -- everything has its own unique distribution!
Fortunately, you don't need to model things exactly. A very rough approximation often gives me miles better results than plain gut feeling.
Its not only complicated, its bordering on the impossible. They are simply hand-waving the "analyze your past decisions" part. Its extremely difficult to be introspective like that. You don't know why someone said "yes" to your deal, and not to someone else. You can guess, but there are so many micro things that fall in your favor when you're successful. Context matters, environment matters, and while we can capture our thoughts, we can't always capture the environment in which those events occurred, etc, etc. This is Harvard business review type stuff, excepting each person to accurately weigh and analyze all that is unrealistic. You're going to find examples of people who "read the book" or "attended the lecture" or "sought advice from X" and struck gold, and these examples provide buoyancy for such ideas.
>If financial institutions (and crypto players) are using these kind of approaches to make their bets, then isn't the individual investor hopelessly outgunned in the vast majority of cases?
When its your job, you're better at it compared to someone for whom its a hobby. And also, larger firms are able to manipulate the market which tilts the balance in their favor.
I think I first heard this quote in one of Taleb's books (paraphrasing):
In Ancient Greece, you were a hero for what you attempted to do and not for what you actually accomplished. This was because the Greeks considered the outcome to be primarily governed by the whim of the Fates. What you attempted to do was a function of your courage especially given that a big task could lead to a big failure outside of your control.
I think of this quote often anytime the "outcome vs process" discussion comes up.
This article, like Duke's book, has a solid premise, but fails to provide any actionable advice aside from a simple risk/reward framework. When I read the book, I was hoping there would be more information about how to properly handicap various situations, but there just wasn't.
Business and life decisions aren't as simple as calculating pot odds and outs. Anyone who has estimated a complex and unfamiliar programming task knows that the unknown-unknowns are the biggest part of any equation.
The largest fault in the practical (life) applications of probabilistic thinking is that estimating the odds in real-time is often impossible. Poker is a constrained environment where the odds are computable.
It's a useful framework for thinking in various situations, but it is almost never going to reduce to an equation that can tell you some objectively correct answer or decision.
If the main reason your effort estimations are off is that you don't allow for unknown unknowns, then that's easily fixable.
After all, although we don't know which the unknown unknowns are, the possibility of them is known. And they do, in my experience, tend to increase the required effort by, say, 1--30 × depending on task complexity and familiarity.
So even in the most complex and unfamiliar of tasks, you can adjust the upper end of your estimate by 30× and there you go! Unknown unknowns accounted for in your effort estimation.
(Simpler or more familiar tasks require smaller adjustments to their upper end. Knowing how much adjustment is appropriate is a matter of deliberate practise.)
Is this site related to Knowledge Project / Farnam Street (fs.blog), a competitor, or just a ripoff???
I noticed because I read what seems to be the exact article there, with the same illustrations and everything on fs a couple weeks ago. Maybe it was a link in the newsletter to this article.
Looking around there is a lot of overlap in content.
I think life is way more complex than the poker. Yes the mindset could be useful, but mindset without proper data is not going to help a lot.
Here is what I'm thinking: How about we go up a level, and accept that life is basically random and a lot is decided by luck (your gene is luck, your childhood education is also luck, these two pretty much decide a lot of things), as indicated in the article, but maintain a (meta) mindset that can manoeuvre around or even mitigate negative emotions?
For example:
- Realize that probability is useful in poker but not that useful in life, thus it is impossible to calculate mathematical expectation.
- Learn to harden one against impulsive emotional acts (impulsive purchase, suicidal thoughts, etc.).
- Learn to control one's material requirements and save some $$ for rainy days.
- Keep connections active so when really bad days come maybe someone can get you out.
Probability is incredibly useful in real life. But it takes deliberate practise. You can improve your "sense of uncertainty" and assign fairly accurate probabilities to things. This process is known as calibration.
It also helps to know some probabilistic identities to derive the same probability multiple ways and see if there's consistency in your assessment.
> Realize that probability is useful in poker but not that useful in life, thus it is impossible to calculate mathematical expectation.
That's very backwards. Sure, "exact" expectation no. But you can add errors bars to all your estimates, that's the basis of statistics. Being able to correctly estimate things and adjust for uncertainty is incredibly useful in life. Unlike poker, you have time to actually punch things into a calculator and consider different scenarios if you want to be super diligent but most things can be intuited if you have a solid grasp of probability/statistics. To be super pragmatic, the whole thing with fake news and misinformation wouldn't exist if people just applied Bayes theorem. For instance, people will say something like P(vaccine is dangerous | bad pharma company) = 1 but then P(bad pharma) = 0 in the sense that Pfizer doesn't want to kill people.
Life's a game of chance where we control the odds.
Just this morning I was reading Scott Adams's How to Fail at Almost Everything and Still Win Big, and he was talking about making bets in Life that don't cost any money (they only cost time) and if we do this there's ~100% certainty of winning. Of course he's more eloquent and elaborate. It's also another way to look at "we miss 100% of the shots that we don't take".
A great concept of Texas holdem is that of chasing the pot: calling when the odds aren't in your favor. In other words, just because you've come this far it doesn't mean you have to keep putting money towards a hand you'll likely lose.
155 comments
[ 3.7 ms ] story [ 197 ms ] thread- potential reward of the decision: checks out
- probability of getting it: 95% if you already have the offer in hand
- resources (time/effort/money) to bet to get the reward: not significant
Then there is 'imagining a bet' and 'analysing your past failures'. The first one might lead you to the 'what if the company goes bust?' question, but would that really help in the decision making or be a reason to not accept the better job? The unlucky event is a recession, not this particular company doing badly (i.e. no signals to see).
If a recession hits, there is no reason to believe the consumer goods co. would fare any better and her old job more secure. I'd really like to see the 'poker player' mindset at play but this just feels like a complete miss.
So, if you have a 95% chance of going 10x that's great, but the question is: can you handle a 5% chance of having being potentially ruined for at least a decade? If not, then you cannot take the bet despite the fact that EV is super high.
Put more extremely, if I offer you a 99% chance to take make one billion dollars (legally) and a 1% chance of being killed, would you take it? I definitely wouldn't.
How about the following? I offer you an unfair coin flip advantage of 51% versus 49%, we're going to flip 1 million times, each time the bet is 1 dollar. Would you take it? I would.
The heuristic that poker teaches is: take a lot of small bets with huge upside with little to no downside. It doesn't matter how slim the chance of winning is. Slightly more nuanced (still a heuristic): the chance should be about as big as your volume of chance-taking. Say, you take 10000 chances, then the slimmest chance you can take (on average) is 1/10000.
https://en.m.wikipedia.org/wiki/Kelly_criterion
I would even if it offered 50+% of getting killed. The small chance you strike a billion is worth the risk.
What would make a difference would be my life situation - if it would be worse or totally shitty I could take even 25% for a billion.
That was kind of theme of "Squid Games".
At some point it goes from a life changing amount of money to just "meh" money.
Your sum risk might be reduced (e.g. your job pays health insurance), but many jobs increase your risk e.g. just driving to work is dangerous.
Of course your primary finite resource is usually time (years of life), and your object function usually relates to quality of life (not $) . . . Win $ for risk of death is a poor metric.
- Right now we spend 33% of our life working, nd the stress of work impacts another 33%.
- A billion is also way more than out life expected income, but there's a big marginal reduction of value to money. But you also get it now when you're younger rather than parsing it out.
- As a random 25 year old male, you would only have a 1 in 16 chance of surviving to retirement age at 60 on your own.
I personally would only take the deal for 2.5% chance of death though.
The only reason I can fathom desiring a billion dollars is the leverage it would give me to influence the world in what I think is a positive direction. But when it comes to money for my own sake, $10M is about the most I can imagine having use for.
In all the worlds where you get the money, your life is improved. In all the worlds where you don't, well, you're dead, so you don't really care.
Obviously, this doesn't take into account the impacts to everyone else in the worlds where you died.
Isn't that just baccarat without ties?
I'd take it in a heartbeat. I've done things that have a 1% chance of death just to have a good story to tell.
https://tompollak.me/blog/ev-v-eg/
> "Probability of getting it — you cannot know the exact probability, but estimate it the best way you can (remember the bet with a friend)"
is the hard part. Jane's case was clear (good decision), but what about Mike? Is there a proportion of his savings he should have invested that would have turned his "bad decision" into a "good one"? We have no signal to tell us what the probability of any cryptocurrency going up or down is over, say, 6 months (they've all gone up and down completely arbitrarily). Basically, finding the probability of any particular "success" (or actually, a failure) will easily tell you how much you should risk.
And that's what it's like with the most of life decisions as well: we don't know the probability — not even the ballpark range. Likelihood that any one person you meet is going to be your life-long partner is basically nil: but we still invest in building relationships before fully committing to either decide they are not, or increase the probability that they will be. But we still do that relationship building investment based on very few signals (appearance, short chats and potentially what social circle someone is from if from a shared acquaintance).
So sure, there is your probability. Will you make money knowing it? That's the question you're alluding to.
(As a concrete example: there is no "true" probability of rain six days from now. Either it rains or it doesn't. But a skilled meteorologist can give you a probability that would probably make you a little money over the climate average.
Both the climate average and the meteorologist's assessments are correct probabilities. They just reflect different amounts of information.)
My point is that with the available information when making most life decisions, we usually have no idea at all of even the ballpark chances of something turning one way or another.
Basically, the advice from the article is to incorporate luck into your estimations (both good and bad), and to use that to determine if a decision is good or bad, and then only go with good decisions (like a poker player would). And don't stress over bad (or good) outcomes if they were mostly due to really bad luck (i.e. something improbable has happened).
But if you don't know if chances of something happening are 10% or 90%, how do you incorporate that into your decision making?
What I am saying is that in life, we implicitely work to reduce the range (eg for a romantic partner, get to know them much better), but we've already decided to invest that much time with very slim chances of them turning out to be our lifelong partners.
The key to this idea, which I don't see covered in the blog post, for making future decisions is that you shouldn't let past bad luck affect your future decisions. If you lose a positive EV bet, it shouldn't shy you away from making the same bet again.
Some examples off the top of my head about decision-making traps people could make: - Continuing to bet at the roulette wheel to "regain" what you've lost - Not going to a well-rated restaurant again just because you had a random bad experience - Not investing in ETFs after being burned by past downturns - Playing MTG and not burning them out just because they had a counterspell last game
Helloooo Japan!
ETFs are great, it will never happen to the US economy :)
In other words, based on market behavior from multiple countries, it is definitely a possibility that ETFs won't return much in a period of 30 to 40 years.
However, the example is simply wrong (IMO) since it assumes that within a 30 year timeframe you'll have gained 8% on average (adjusted for inflation). If this would be true, thne yes it would be a decision-making trap. I think that's what morley is getting at.
I'm arguing it's a tough sell that the S&P 500 works like that. If one would agree that the S&P 500 might not continue to give 8% ROI on average over a 30 year time frame, then one might consider doing something else with their money. For example, maybe it's more fruitful to invest in yourself to upskill even more rather than putting your money into the markets. I don't know I haven't researched it, I myself try to beat the market, it's a fun endeavour. The jury is still out.
Such that, if the "burn rate" for my household today is $60k/yr, and I store $60k in investments, then when I withdraw at some future date $X will still be enough to support my household for a year.
Physical cash is terrible but to an investor, "cash" tends to mean something like t-bills, which over the past century have done slightly better than inflation. Gold is horrible if you have too much of it, but it tends to do well in equity crashes, so mixing in a little can be helpful. Bonds also sometimes do well when stocks go down, though I think that's more likely when you start at higher interest rates. Commodities can have 15-year bull markets independent of everything else. An investor in Japan would have done fine if they had a lot of foreign stocks. Etc.
https://www.google.com/finance/quote/GLD:NYSEARCA?comparison...
10 year Treasury outperformed NASDAQ https://www.google.com/finance/quote/GLD:NYSEARCA?window=YTD...
That's diversification, even if it just means losing less.
Had this happen at work recently and shapes why I was ok with the outcome even though it wasn't what we wanted.
I was building an integration and got it working and trying to push to Prod when I realized the integration with the 3rd party required a security audit... Given the information at the time, it was the right thing to build but didn't do deep enough research. Lesson learned.
A major one is how we tend estimate probability. We naturally do that by equating it with how easy it is to imagine or remember it happening. This worked well in a tribal situation where our world was very small...if Grog got attacked by a tiger, there's a good chance that tiger attacks are a serious danger that I should consider. But this breaks down in the face of a global society of billions of people and a media that profits from making people afraid. Rare events get magnified by media attention and feel, to us, like they're incredibly likely. Plane crashes are incredibly rare. Even in a plane as fundamentally broken as the 737 Max was, passengers were safer traveling that way than by car. But since every airplane crash is covered extensively on the news whereas car crashes rarely merit a mention, people's primitive cognitive quirks kick in and they're more afraid of flying. You see it with mass shootings too. Even if you get 1-2 per day in the US, that's still maybe a few hundred people per day who are directly affected. Over the course of a year, that's roughly 70k people out of 330 million, or %0.02 of the population. Which isn't to say that we shouldn't do everything we can to prevent it, but when you hear people saying that they don't feel safe, that's entirely on the media and how it's warping our perception of the danger rather than it being a real threat to our lives. We also saw it with the spate of Asian attacks that became a media favorite. There was a story of Asians in New York starting to contort their lives to avoid becoming victims. But when I looked up the actual statistics, there were 20 such attacks at the time in a city with 1.2 million Asians. Which, again, doesn't mean that we shouldn't be treating each of those attacks as terrible and be doing everything we can to stop them, but when you consider 20 victims out of 1.2 million, that's just not a risk that's worth going to considerable inconvenience to avoid.
There was an interesting TED talk many years ago from a guy who specialized in these kinds of cognitive quirks. And he discussed the findings of one particular study that always stuck with me. In it, participants were presented with one of two hypothetical situations. In the first, they were going to the theater to see a movie and they pre-purchased the ticket for $20, and also brought along a $20 bill. When they arrived at the theater, they found that the ticket had gone missing. Almost all participants in the study said they'd turn around and go home rather than using the other $20 to purchase another ticket. In the second case, they were going to the theater to see the same movie, but instead of having pre-purchased a ticket, they were intending to buy one when they got there. But they had, instead, brought two $20 bills with them and managed to lose one on their way to the theater. Almost all participants presented with this scenario chose to use their remaining $20 bill to buy a ticket. Despite the situation being basically identical, when you're presented with both scenarios, we have a cognitive quirk where we don't want to pay for something twice whereas we can rationalize the loss of money that we never spent. But what got to me about that study wasn't the results, but how intuitive both decisions felt to me. I could feel how unpleasant it would be to buy a second ticket and, yet, how easy a decision it would be to buy a ticket for the first time even after losing some of my money. The lesson, for me, was that while I've always thought of myself as a logical person and someone who would always let statistics or other scientific basis guide my thinking, it's still really hard to follow through with that.
Statistically, on any given day, the odds of being in an accident are quite low. However, if you drive for 30-60 minutes every day, you are exposing yourself to a low amount of risk consistently over time, which inevitably means that one day, your number will come up.
Most people don't have a background in probability or they haven't taken the time to look at driving, but for those who consider themselves normally "rational", they're often very surprised to see just how dangerous driving a car really is, but they just weren't aware of the amount of risk.
Analyzing the risk of driving every day is an exercise worth doing. That being said, I have a fairly high level of risk tolerance, so I drive to work every day because I live just a little too far to reasonably walk, and riding a bike would be a higher risk than driving because of lack of adequate road shoulders for much of the route between my workplace and home.
Is this an empirical statement or based on the same gut feeling that makes people feel safe in a car?
I'm asking because when you actually run the numbers, in most cases I've been involved in, you get to be very surprised to see just how safe cycling really is.
I don't know what the culture and road type is around your neck of the woods, but I would prefer a road without a proper shoulder most of the time -- simply because the existence of the shoulder provides very little actual safety, but it does give a sense of safety that makes drivers behave more dangerously near cyclists. Not to mention lower quality road surface, collected debris, etc.
I've seen many people getting a 3 year contract without having a clue of what they had to do, just buzz words.
Public sector
The reality is, 95% of scenarios don’t require what is asked for, and stating your capabilities in the most generous way is the optimal decision. Just back it up with work.
Numbers made up and simplified but in the same ballpark:
Round-trip ticket across town: 10€
Monthly pass: 50€
Fine: 70€
The chance to be controlled is roughly 1 in 10 for a round-trip. So if one did 5 round-trips in a month, on average one would pay 70 x 5 x 0.1 = 35€. Which is a good price with acceptable variance fur such usage. When caught just pay without fuss.
At some point (between 7 and 8 round-trips) it's more cost effective to just buy the monthly pass, as the average cost of not buying approaches/goes over such threshold.
As much as "fake it till you make it" is a much hated approach, there's a grain of truth to it. Many people only got the chance to prove their merit by bluffing (faking?) it first. Sure, they had to ultimately deliver on the expectations, and perhaps there are more of those who faked it and then failed to deliver. It doesn't change the fact that often you don't even get the chance if you don't project an aura of confidence.
Generally speaking, being confident gets you to places. This really applies to every aspect of life.
If you act like you belong somewhere and that there can be no questions about you belonging there, people in general will not question your presence. If you seem lost, or confused people will be curious why.
Walk into a private event acting like you belong, and there are good chances nobody will realize you shouldn't be there. Walk around looking all confused and you are likely to be asked to validate your presence.
Naturally, this doesn't apply only to places, but also groups, communities, companies...
How do I get caught? If they say no I get up, pause then change my mind.
I've never suffered from being caught by a bluff, but I structure them sensibly.
It surprises and slightly delights me to see how often people who try to ‘game’ others, queue jumpers, and others who display selfish behaviour are then given special (negative) treatment from those who notice their behaviour.
Some seem to go around thinking that people are awful, but really people are just awful to _them_. People are generally kind and help each other.
Other life pro tips:
- Stealing is a way to get money
- Murder may be useful
- Fraud is profitable
A bluff, in this context, is claiming or insinuating you're in a better position that you really are.
Eventually, you really do believe the lies you tell and then your ability to see the world for what it is is diminished.
Additionally, when dealing with other people, we tend to fill in the blanks about the things we don’t know about them using information we have picked up from other people as a template. We know ourselves better than any other person, so we often use our own motivations as a stand in for the motivations of others. If we are liars ourselves, we assume other people are liars too, by default (and, due to confirmation bias, once we look for evidence of this, we find it everywhere).
This mindset means we end up trusting people less than is probably optimal, but we are unaware of this fact.
Say you're a physical coward and you're being picked on for a fight by someone. You can puff your chest up, put your shoulders back and put a sneer on your face, possibly scaring the bully off.
Yes you're misrepresenting yourself. You're lying. But how exactly is that corrosive to your character?
The thinking like a poker player is mostly about being upfront about your risk tolerances and then having a culture supporting people who make the best risk adjusted decision, even if it doesn't work out.
Expected Value is complicated because a 50% chance for $100 is the same as a 10% chance at $1000. It's the variance, not the EV that makes a lot of decisions hard.
You (and businesses) need to decide what risk is acceptable, communicate that clearly. Reward people who manage risk in a way that's aligned with the business even when it doesn't work out. Get rid of people who are either take too big risks, and people who don't take risks.
Is it though?
1 in 2 for the first one will pay out. and 1 in 10 for the second one will pay out.
Variance is higher for second than for first.
"50% chance for $100 is the same as a 10% chance at $1000"
should be
"50% chance for $100 is the same as a 10% chance at $500"
But that is when the payoff is 1, of course!
When it's not, we're really looking at payoff² * p * (1-p) and indeed the second situation has much higher variance.
Again I ask, is it really?
Reminds me of: https://en.m.wikipedia.org/wiki/Chinese_word_for_%22crisis%2...
The alternative to getting rid of these people, is offering positions to them that better match their risk tolerances.
EDIT: I see someone else noticed this further down than I looked in comments, sorry for dupe.
They use expected returns in conjunction with size of current pot and size of current holdings.
What you say about variance is right on. It involves the payout and the odds and so on ...
They use something internally like the Kelly Criteria, probably tuned against their own model of what is the probability of each relevant outcome.
The odds of winning a hand with a future card in order to estimate the call's expected value.
Future cards have significant impact on odds and betting.
If you have not yet made your hand (e.g. only 4 spades) then you consider your drawing odds, which is the chance of making your hand.
But say there is a pair on the board. Then you need to consider the probability that your opponent has or will make a full house, which would beat your flush if you were to hit. Likewise the probability that someone will have a higher flush than you.
Analyzing this in aggregate is what gives you the expected value, which is what poker players consider.
If you are simply playing a cash game with unlimited buy-ins. Then assuming your overall bankroll is sufficient, you will always make plays with positive expected value, even with high variance, because it is a continuous game and you expect the total return to be positive over sufficient iterations.
If you are playing a tournament with a set number of places in the money, then you need to take into account the variance as well as the expected return because it is an episodic game and the types of plays you will make will depend on not just expected value of the single play but what place you are in currently and the number of players left.
Implied odds are interesting in that they can make very bad hands more playable because if you do hit, then the likelihood that your opponent believes that they are winning is higher (because the chances of such a bad hand being played rather than folded immediately should be low) and thus the probability that you can elicit more money from them is higher.
Small pairs and low suited connectors are good examples of these kinds of hands. Your pre-flop chance of winning with a pair of 2's on a full table is pretty low. But if the flop is 2-K-A, you will likely get a lot of action.
> ~15k hands last month.
Is that not roughly a full-time jobs' worth of poker? (15k hands > 12k minutes? > 200 hours > 6h/day)
I can play up to 12 at the same time, but my play suffers and I don't feel like I'm able to focus and get better.
1)All other things being equal you should generally prefer a higher expectation decision to a lower-expectation one
2)But you have to avoid decisions which have outcome distribution properties you can't tolerate. Most obviously you should generally ensure that your risk of ruin is zero because even if the probability is very low, that outcome is close to impossible to recover from so must be avoided.
3)For two possible strategies with similar-ish expectations in real-life terms you may well want to choose the one with the lower standard deviation of payoffs. Like say you're choosing between an offer at medical school to train to be a dentist and pursuing your long-shot idea of going to Hollywood and trying to make it as an actor. Imagine that when you look at the outcomes you estimate that in acting you have an insanely low probability of making it but a huge payoff if you do and in dentistry most people do pretty well but no-one's partying on superyachts with Jay-Z and Beyonce. Well even if the expectation of acting works out slightly higher, if they are close enough you should probably pick dentistry because the variance of outcomes is just way lower. If you think about your future as a monte carlo simulation you want to end up "ok" on most of the paths in the simulation even if that means giving up on some "lights out" outcomes.
4)most people agonise the most about decisions where the expectation is really pretty similar so it just doesn't matter that much either way. So don't beat yourself up about whether you made the right decision - just try to learn from the decision and move forward
A huge mistake people make is to evaluate the quality of the decision based on the single possible outcome that crystalised into reality by actually happening rather than using the framework above. So resist that temptation and instead evaluate your decision based on whether you chose to maximise expectation while avoiding risk of ruin and excessive variance.
[1]In the sense of being the weighted average of all possible payoffs where the weights are the probabilities and the payoffs are the utility of each outcome (usually just in cash terms).
Hold 'em, at least, still ain't fully solved.
Well, yes, but the point is that a fully GTO non-exploitable strategy is not solved for Hold'em. At least not outside of limit (and kinda no-limit) heads up play.
https://arxiv.org/pdf/1805.08195.pdf
(There is also an argument that Libratus actually just approaches Nash equilibrium and isn't fully there, since there are 1e+160 decision points in a heads up no-limit game, but compared to human players the difference is probably meaningless. )
The closest we have seen for larger games is Pluribus - https://www.science.org/doi/10.1126/science.aay2400 - but the researchers there aren't even attempting to say they have solved 6max, just that they could be less exploitable than some of the best in the world. It is not particularly useful for online play because it does not try to counter variance or manage bankroll, and in further results it began to lose quite heavily as the pros adapted to it, losing 700BB over the 10k hands.
By no means am I saying GTO-based strategy is ineffective - I'd probably be wasting my DTO subscription if I did think so ;) - but we're only part of the way there and the missing bits mean that you can't play perfectly unexploitable poker.
People try this.
> …by increasing their bet size to a point where I can’t profitably call.
A good poker player will adjust your expected range based on your sizing and frequency of bets/raises, and they will adjust their calling/raising ranges accordingly.
Most poker games, especially big bet games are played against a range/distribution of hands, not a specific hand.
I'll also point out that constructing ranges and thinking in terms of ranges/ hand distributions is a relatively new development in poker. I mean, the very good players were subconsciously doing it all along, but it's only been written down and analyzed in the last 10 years or so.
I agree with you overall points.
For historical reference for folks who are interested in the poker boom of the 00s but were not there, it’s more like 20 years, if not more.
The book Let There Be Range was published in 2008.
High stakes players were discussing the issues covered in that book amongst ourselves (live and online) with quite a bit of sophistication for several years before that.
All that said, and supporting the point made above, advanced computational support for ranges (esp. via “solvers”) didn’t really start becoming a thing until the early 2010s.
Annie Duke has cashed 39 times at the WSOP and has won at least four million dollars at tournaments alone.
Any good poker player would understand that the example was simplified for a non-playing audience. It’s so obvious it didn’t need to be stated.
This statistic is almost certainly not counting tournament buy-ins. It doesn't represent net profit. If you want to evaluate somebody's performance as a poker player, you want to look at net profit (among other things). Literally every poker player will have some wins, so if you only count the wins without counting losses, it will sound impressive but not actually mean anything.
This is most certainly not true, because it doesn't subtract the cost of all the entry fees, to the tournaments she cashed and the dozens or hundreds she didn't.
"alone" implies she has won additional money in non-tournament play, when there is no evidence she is a winning cash player.
Finally, there is a LOT more to the Annie Duke story. Morally, ethically, business-wise. A winner? A little research goes a long way.
It is fairly standard in the poker world to discuss total winnings in tournaments rather than net winnings.
When discussing net winnings, it’s usually discussed using an ROI measure (e.g., 30% roi over their past 100 MTTs between $1000 and $5000) rather than just dollars, since the biggest winners in dollars are almost all winners of super high roller tourneys like the Big One for One Drop.
Furthermore, live MTT results in general are often skewed due to the unmatched (?) softness of the $10k wsop main event for a field that size.
With regards to Annie Duke in particular, I never considered her a “good” pro poker player. That said, she was pretty good at fleecing amateurs, and she was able to play a tight game when she was outclassed.
But this is an article aimed at the general public.
An additional comment that I copied from another reply:
With regards to Annie Duke in particular, I never considered her a “good” pro poker player. That said, she was pretty good at fleecing amateurs, and she was able to play a tight game when she was outclassed.
You'll also become predictable if you only overbet when bluffing. If you try to deal with that by making all your bets larger, then you lose out on value from your strong hands because I can now profitably fold some hands that I would have had to call you down with otherwise.
Put another way: you can't exploit someone who makes decisions based on pot odds just by changing your bet size, because changing your bet size changes the pot odds.
In real life the best marksman can get wiped out on the first day of a war, the best business strategist can get done in by a once in a generation disaster, and a hapless idiot in business can get an early big break. And so on.
I feel like it feels much more natural and intuitive to us to play games that are a mix of skill and random chance.
[1] https://www.youtube.com/watch?v=dSg408i-eKw
In truth, managing randomness is actually one of the core skills of the game. Not just on the decision in front of you, but on future decisions as well. It's a hard skill to master. Moreover, the timelines to evaluate one's skill are long, sometimes hundreds or thousands of games, incorporating tens of thousands of decision points. It's not a wonder it's often misunderstood.
Except that Magic: The Gathering is particularly bad for locking into specific "meta" (metagame) which then persists until new cards come out/old cards rotate out. Hardly random and hardly difficult to master.
The best MTG players are really good at what I call the "game mechanics corner cases". The know all the weird ways to play and resolve the cards--this gives them a quite persistent couple percentage point advantage.
However, it won't give them enough advantage to beat the "meta".
It does not matter what hand an opponent has as long as you can convince him you have a better hand.
Also, it is handy to recognize the situations when an opponent might bluff, so you can call.
Bluffing is about making non-optimal choices for the express purpose of making it more expensive for someone else to make optimal choices. If someone knows you're engaging in bluffing, that's fine, because by bluffing you're still forcing them onto non-optimal lines.
On the other hand, you would prefer people not to know that you're engaging in deception.
So what you need to work on is how you define "success" in life. Money is a high variance objective, whereas self-improvement is a low variance objective. I think finding a good balance of variance and tolerance to risk is a key to happiness.
Rejecting the role of luck in our outcomes is delusional, and people who support it have something to peddle (if only their own personal brand as thought leaders).
In the non-poker real world, most people maybe get to play one hand in life, and that’s it. One at-bat in a baseball game, to use a different analogy. You dump 20 years of savings into buying that laundromat, and that’s your only shot of making it. If we are really lucky we might get one or two more hands to play but that’s it for the vast majority of us.
The rich get as many hands/at-bats in life that they want to play, so they will eventually win one. They can just serially start business after business until one of them lucks out and succeeds, then do talks for the rest of their lives on how to be good at business.
Money buys you chances to make money.
Measuring your net worth against those born rich is a fools errand, but you can definitely make your children have a better shot against their children. True wealth is almost never built up in a single generation, despite the fairy tales about the "American Dream" that we tell our children.
https://en.wikipedia.org/wiki/Expected_value#Expected_values...
Probability distributions are complicated. Even determining what kind of distribution you are look at is difficult in many cases. Such distributions are also skewed in real life by things like insider information (in poker, that would be cheating). Even so the list is rather intimidating, assuming fair play:
Bernoulli, Binomial, Poisson, Geometric, Uniform, Exponential, Normal, Standard Normal, Pareto, Cauchy
If financial institutions (and crypto players) are using these kind of approaches to make their bets, then isn't the individual investor hopelessly outgunned in the vast majority of cases? Plus, not having a big pool of capital to absorb temporary losses makes that situation even worse.
Investment capitalism, in other words, is just a casino for the uber-wealthy. Letting it rule the economy is a serious mistake.
So you're not even dealing with a somewhat countable list of distributions -- everything has its own unique distribution!
Fortunately, you don't need to model things exactly. A very rough approximation often gives me miles better results than plain gut feeling.
>If financial institutions (and crypto players) are using these kind of approaches to make their bets, then isn't the individual investor hopelessly outgunned in the vast majority of cases?
When its your job, you're better at it compared to someone for whom its a hobby. And also, larger firms are able to manipulate the market which tilts the balance in their favor.
In Ancient Greece, you were a hero for what you attempted to do and not for what you actually accomplished. This was because the Greeks considered the outcome to be primarily governed by the whim of the Fates. What you attempted to do was a function of your courage especially given that a big task could lead to a big failure outside of your control.
I think of this quote often anytime the "outcome vs process" discussion comes up.
Business and life decisions aren't as simple as calculating pot odds and outs. Anyone who has estimated a complex and unfamiliar programming task knows that the unknown-unknowns are the biggest part of any equation.
It's a useful framework for thinking in various situations, but it is almost never going to reduce to an equation that can tell you some objectively correct answer or decision.
After all, although we don't know which the unknown unknowns are, the possibility of them is known. And they do, in my experience, tend to increase the required effort by, say, 1--30 × depending on task complexity and familiarity.
So even in the most complex and unfamiliar of tasks, you can adjust the upper end of your estimate by 30× and there you go! Unknown unknowns accounted for in your effort estimation.
(Simpler or more familiar tasks require smaller adjustments to their upper end. Knowing how much adjustment is appropriate is a matter of deliberate practise.)
I noticed because I read what seems to be the exact article there, with the same illustrations and everything on fs a couple weeks ago. Maybe it was a link in the newsletter to this article.
Looking around there is a lot of overlap in content.
Here is what I'm thinking: How about we go up a level, and accept that life is basically random and a lot is decided by luck (your gene is luck, your childhood education is also luck, these two pretty much decide a lot of things), as indicated in the article, but maintain a (meta) mindset that can manoeuvre around or even mitigate negative emotions?
For example:
- Realize that probability is useful in poker but not that useful in life, thus it is impossible to calculate mathematical expectation.
- Learn to harden one against impulsive emotional acts (impulsive purchase, suicidal thoughts, etc.).
- Learn to control one's material requirements and save some $$ for rainy days.
- Keep connections active so when really bad days come maybe someone can get you out.
It also helps to know some probabilistic identities to derive the same probability multiple ways and see if there's consistency in your assessment.
That's very backwards. Sure, "exact" expectation no. But you can add errors bars to all your estimates, that's the basis of statistics. Being able to correctly estimate things and adjust for uncertainty is incredibly useful in life. Unlike poker, you have time to actually punch things into a calculator and consider different scenarios if you want to be super diligent but most things can be intuited if you have a solid grasp of probability/statistics. To be super pragmatic, the whole thing with fake news and misinformation wouldn't exist if people just applied Bayes theorem. For instance, people will say something like P(vaccine is dangerous | bad pharma company) = 1 but then P(bad pharma) = 0 in the sense that Pfizer doesn't want to kill people.
Just this morning I was reading Scott Adams's How to Fail at Almost Everything and Still Win Big, and he was talking about making bets in Life that don't cost any money (they only cost time) and if we do this there's ~100% certainty of winning. Of course he's more eloquent and elaborate. It's also another way to look at "we miss 100% of the shots that we don't take".