Something I've always wondered about: randomness and success

Things that don't belong anywhere else. (Check first).

Moderators: Moderators General, Prelates, Magistrates

Sabredog
Posts: 55
Joined: Sat Jun 01, 2013 6:02 pm UTC

Something I've always wondered about: randomness and success

Postby Sabredog » Fri May 29, 2015 4:36 pm UTC

Let's consider a bunch of things that can gain "success" in the world: an idea, a startup, a fashion trend, a new word in the language, an entrepreneur, a podcast, Google+ versus Facebook, Whatsapp versus Viber.

Of all of these things, there is a tempting psychological tendency to slap a narrative on it to explain its success. (cf: "Survivorship bias").

According to a book called "A Matter of Taste" by Stanley Lieberson, there are some persistent false beliefs about cultural change​, such as how trends happen: that trends have a narrow purpose and meaningful end-goal, in the same way a person who changes has a goal-directed purpose ("I want to lose weight, so I'm going to diet").

Instead, cultural change reflects billions of people in an internally dynamic system making personal decisions in ways contingent upon other people's personal decisions (there are feedback processes) in ways that elude prediction and neat explanation by social scientists.

Now, without doubt, for a product to be successful it must be good: it cannot a piece of shit. A great product will beat a bad product.

But what I'm interested in is (and I'd love to get your thoughts): what if there are two great products, which are pretty much--for the sake of argument--identically good. Let's imagine Google+ is technically just as good as Facebook--equally good (or bad) interface and features, equally good marketing campaigns behind them, etc.
And they both go up against one another, with their success being measured by sheer scale.

What happens? Some would predict they'd both be successful, but I don't think there's a clear link between quality and success. Quality is necessary but not sufficient. You can do everything "right" but still not succeed.

Sometimes, great products just die and nobody can predict why--something zigged instead of zagged, some random or chaotic force thwarts the product getting momentum, something kicks the herd mentality (of the user base) in one direction and not the other. We can slap a story on afterwards (everything, it seems, can be explained afterwards) to try to explain it: we can "explain" why something failed or succeeded, but factoring in random processes makes this messier and less reliable. And we hate messy randomness, we like neat stories, but life just doesn't work that way.

User avatar
ucim
Posts: 6550
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Something I've always wondered about: randomness and suc

Postby ucim » Fri May 29, 2015 4:48 pm UTC

Sabredog wrote:Now, without doubt, for a product to be successful it must be good: it cannot a piece of shit. A great product will beat a bad product.
This is false. There are many examples of bad products that beat good products. Sometimes the reason they beat the good products is that they are bad. They either do nothing new (so are easy to adopt), are hard to use (so once learned, nobody wants to quit) or they break (and need replacement, keeping the company in business). Part of business strategy is just how bad can you make something before people stop buying it.

Sabredog wrote:Let's imagine Google+ is technically just as good as Facebook
The product being "sold" isn't Google or Facebook; it is the network of users, whose value depends on its size. First mover advantage is strong, assuming the first mover doesn't ch*rp it up. The early days are far more susceptible to randomness; one bad review of a good new show could kill it, but in a show that's been going on for months, it takes more than that.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Please help addams if you can. She needs all of us.

User avatar
Deva
Has suggestions for the murderers out there.
Posts: 2018
Joined: Sat Feb 26, 2011 5:18 am UTC

Re: Something I've always wondered about: randomness and suc

Postby Deva » Fri May 29, 2015 5:50 pm UTC

Quality: Still wonders how Goat Simulator performed so well. Credits publicity somewhat. Noticed tons of Let's Plays on it. May have appeared after its success, though.

Argues for aesthetics sometimes. See: Pokemon games. Sold 1.63 million copies of Pokemon X and 1.54 million of Pokemon Y in North America at one point. Caused neither to fail, obviously. Represents one of many factors, however.

Considers company/brand name a factor too. Would you rather own a Chimera or a Doldrum? Do you trust Blizzard over an unknown company?
Changes its form depending on the observer.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Something I've always wondered about: randomness and suc

Postby Izawwlgood » Fri May 29, 2015 6:05 pm UTC

You might be interested in part in this guys series.I linked the one on why Betamax lost to VHS, but he's got a bunch that are really interesting from a design/engineering perspective.


Annnnnnnnd there went 30m of my afternoon while I watched a few.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
SecondTalon
SexyTalon
Posts: 26256
Joined: Sat May 05, 2007 2:10 pm UTC
Location: Louisville, Kentucky, USA, Mars. HA!
Contact:

Re: Something I've always wondered about: randomness and suc

Postby SecondTalon » Fri May 29, 2015 9:03 pm UTC

I never liked the "Betamax was superior!" argument, as it is the same sort of argument some video gamers will put forward as to why Game X is better than Y - it looks and sounds better. Before considering the cost, VHS recorded longer, that always seems to be ignored. The VHS camcorders, while heavier, allowed for in-camera review and the Betamax one didn't.

People will gladly accept a reduction in end-result quality if it means the ability to do it longer. Or, in the case of a process, people will often gladly accept a reduction in end-result quality for a faster result.

Or, you know, what that guy in that video said now that I've watched it.
heuristically_alone wrote:I want to write a DnD campaign and play it by myself and DM it myself.
heuristically_alone wrote:I have been informed that this is called writing a book.

User avatar
K-R
Posts: 1564
Joined: Thu Mar 15, 2012 6:42 pm UTC
Location: Australia

Re: Something I've always wondered about: randomness and suc

Postby K-R » Fri May 29, 2015 9:19 pm UTC

ucim wrote:
Sabredog wrote:Let's imagine Google+ is technically just as good as Facebook
The product being "sold" isn't Google or Facebook; it is the network of users, whose value depends on its size. First mover advantage is strong, assuming the first mover doesn't ch*rp it up. The early days are far more susceptible to randomness; one bad review of a good new show could kill it, but in a show that's been going on for months, it takes more than that.
This is true of basically everything. Thing gets popular, everyone knows about thing, other better thing comes along, nobody cares. People assume first thing is better because it's more popular, and lots of people like it so it must be good. Actually better thing never gets eyeballs, is run by an idiot, gets canceled by two different TV networks in six months, dies.

Seriously, WWE is the fifth-best wrestling sports entertainment show on US television, behind TNA Impact, NXT, ROH and Lucha Underground. But nobody's heard of those four.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Something I've always wondered about: randomness and suc

Postby Izawwlgood » Fri May 29, 2015 10:21 pm UTC

SecondTalon wrote:Or, you know, what that guy in that video said now that I've watched it.
And to think I stretched my indignant typing muscles for nothing.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Tyndmyr » Mon Jun 01, 2015 6:04 pm UTC

Sabredog wrote:But what I'm interested in is (and I'd love to get your thoughts): what if there are two great products, which are pretty much--for the sake of argument--identically good. Let's imagine Google+ is technically just as good as Facebook--equally good (or bad) interface and features, equally good marketing campaigns behind them, etc.
And they both go up against one another, with their success being measured by sheer scale.

What happens? Some would predict they'd both be successful, but I don't think there's a clear link between quality and success. Quality is necessary but not sufficient. You can do everything "right" but still not succeed.

Sometimes, great products just die and nobody can predict why--something zigged instead of zagged, some random or chaotic force thwarts the product getting momentum, something kicks the herd mentality (of the user base) in one direction and not the other. We can slap a story on afterwards (everything, it seems, can be explained afterwards) to try to explain it: we can "explain" why something failed or succeeded, but factoring in random processes makes this messier and less reliable. And we hate messy randomness, we like neat stories, but life just doesn't work that way.


Quality isn't a single metric. All markets are far more complex than that.

And yes, with sufficient data, you absolutely can predict success and failure. "Random" or "Luck" is code for "I don't know the actual reason" when used in this fashion.

For instance, markets are frequently disrupted not by something that is more badass, but by something that is good enough, and a lot cheaper. This is a well known thing, but it isn't strictly a "quality" issue in a direct sense. Walmart didn't win because it was the best store ever, it won because it was good enough for many people, and it was cheap.

And, in my experience, failure usually not only has a cause, it has many causes. When a company starts getting desperate, sometimes they start taking risks. Flailing, really. One change after another trying to desperately make things right, but often not even allowing significant time between changes, or taking the time to consider if the change is really likely to work. So, what starts as a bad situation rapidly gets worse.

Usually, the winner isn't perfect or anything. They're just the party that fucked up least.

Sabredog
Posts: 55
Joined: Sat Jun 01, 2013 6:02 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Sabredog » Mon Jun 22, 2015 6:57 pm UTC

Apologies for delay in responding- great responses, thanks. Food for thought definitely.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Tyndmyr » Tue Jun 23, 2015 3:41 pm UTC

Also of interest. Statistics are what you use to summarize complex deterministic activity. Sometimes even when you don't understand the deterministic activity. However, that doesn't mean the cause is actually non-deterministic. Much of what we call randomness is actually not very random at all.

Yeah, you may have only x% chance to get in a car accident today, but if you are in a car accident, there will be a distinct chain of causality leading up to the collision.

Sabredog
Posts: 55
Joined: Sat Jun 01, 2013 6:02 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Sabredog » Sat Jun 27, 2015 12:20 am UTC

True. Isn't there a quote that roughly goes chaos is defined by an observer's limited ability to understand

User avatar
Djehutynakht
Posts: 1546
Joined: Thu Feb 10, 2011 1:37 am UTC

Re: Something I've always wondered about: randomness and suc

Postby Djehutynakht » Sat Jun 27, 2015 12:25 pm UTC

Machiavelli said something like this in Il Principe.

He said that there are two things that determine outcomes in life: virtue and fortune.

Virtue largely being how well you've prepared, and fortune being, well, luck. Chance.

The less virtue you have.. i.e., the less you've prepared, strengthened yourself, gathered resources, planned, etc. the more the result is left to luck. The more you plan and refine and etc., the less the outcome is left to luck. But in the end, there's always going to be some element of chance which you just can't eliminate. You could be the best military commander in the world, per say, but if lighting hits your army camp and burns it to the ground then you're just outright screwed anyways.

Sabredog
Posts: 55
Joined: Sat Jun 01, 2013 6:02 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Sabredog » Sat Jun 27, 2015 11:59 pm UTC

Djehutynakht wrote:Machiavelli said something like this in Il Principe.

He said that there are two things that determine outcomes in life: virtue and fortune.

Virtue largely being how well you've prepared, and fortune being, well, luck. Chance.

The less virtue you have.. i.e., the less you've prepared, strengthened yourself, gathered resources, planned, etc. the more the result is left to luck. The more you plan and refine and etc., the less the outcome is left to luck. But in the end, there's always going to be some element of chance which you just can't eliminate. You could be the best military commander in the world, per say, but if lighting hits your army camp and burns it to the ground then you're just outright screwed anyways.


Really good point. The other posts didn't really capture this point that I was making about success, and focused more on the quality of a product: interesting, but not quite what I was getting at.

As far as I know, there's a book called The Darwin Economy that argues the point that a great deal of success in any economic endeavour is hugely dependent on good luck (among other things obviously)

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Tyndmyr » Mon Jun 29, 2015 10:13 pm UTC

Sabredog wrote:As far as I know, there's a book called The Darwin Economy that argues the point that a great deal of success in any economic endeavour is hugely dependent on good luck (among other things obviously)


Not generally. Yeah, at a certain level, luck matters. Happening across just the *perfect* thing to make a billion dollars involves a good degree of luck. If you consider "having a million dollars from your parents to start with" to be luck, anyways. That's a huuuge factor. So is education. I mean, yeah, you can label money and knowledge as "luck", but at a certain point, luck is becoming a magic explanation like "it's gods will", which has absolutely crap-all for predictive value, but can be used to explain any outcome.

On the flip side, there are very definitive crap decisions you can do that will reliably tank economic endeavours. Certain reasons show up with extreme consistency as to why businesses fail.

If you want the answer "it's all luck", yes, you can find some way to justify that. But that justification will lead you absolutely no closer to success or understanding. A deterministic approach will.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7494
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Something I've always wondered about: randomness and suc

Postby Zamfir » Tue Jun 30, 2015 10:58 am UTC

Tyndmyr wrote:If you want the answer "it's all luck", yes, you can find some way to justify that. But that justification will lead you absolutely no closer to success or understanding. A deterministic approach will.

I am not 100 percent sure of that. It's true that if you are engaged and comitted in some effort, it's usually wise to act as if the outcome is controllable, to focus on the things you can actually do to make a difference.

But it's different once you're a step removed, not (yet?) deeply engaged in the activity itself. For example, when you're deciding how to commit resources and effort to various options. At that stage, it's important to understand the role of luck in your activity. Should you commit a base level of effort to 5 vaguely promising options, or commit fully to the most promising option? If luck and uncertainty play a large role, you should go more to multiple options. "Fail early, fail often" is a good strategy for luck-dependent endeavours.

Another example: learning from the past. When luck is a large factor, you should be careful before emulating successes from the past. Those are likely biased towards overly risky strategies that only look good due to selection.

In general, I'd say it's very important to consider when you can predict or control an outcome, and when not.

User avatar
ucim
Posts: 6550
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Something I've always wondered about: randomness and suc

Postby ucim » Wed Jul 01, 2015 2:35 pm UTC

Zamfir wrote:In general, I'd say it's very important to consider when you can predict or control an outcome, and when not.
... and that's what "luck" is: a measure of what you can predict and control.

Note also the "and". If you can't control it, it's luck. If you can't predict it, to you it's luck even if to a fully informed outside observer it is plain as day.

Preparation tends to minimize the luck aspect by increasing what you can predict and control, but some things will remain outside that realm, and Zamfir's points are well taken.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Please help addams if you can. She needs all of us.

Sabredog
Posts: 55
Joined: Sat Jun 01, 2013 6:02 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Sabredog » Wed Jul 01, 2015 3:55 pm UTC

Survivorship bias is an interesting cognitive shortcoming.

From http://www.ceobootcamp.com/why-you-cant-learn-to-be-successful-by-studying-successful-people/

"Learning about success comes from understanding the difference between successes and failures. Unfortunately, in business that comparison is rarely done. Largely because it doesn’t make an entertaining speaker series, or an easy to write article. And because business failures tend to disappear and not subject themselves to scrutiny.

The term for the logical failure in this is called “survivor bias” and if you don’t think you’re guilty of it – you better read this article.

You’ll also learn why these things are true:

“A stupid decision that works out well, becomes a brilliant decision in hindsight”
“If you group successes together and look for what makes them similar the only real answer will be luck.”

The quotes are from Daniel Kanneman. I think he overstates it, possibly, though there is definitely truth to it. Hmm.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Something I've always wondered about: randomness and suc

Postby Tyndmyr » Wed Jul 01, 2015 5:15 pm UTC

ucim wrote:
Zamfir wrote:In general, I'd say it's very important to consider when you can predict or control an outcome, and when not.
... and that's what "luck" is: a measure of what you can predict and control.

Note also the "and". If you can't control it, it's luck. If you can't predict it, to you it's luck even if to a fully informed outside observer it is plain as day.

Preparation tends to minimize the luck aspect by increasing what you can predict and control, but some things will remain outside that realm, and Zamfir's points are well taken.

Jose


If it's something you absolutely can't predict or control, well, sure, luck is a reasonable term. But what you can predict and control are largely alterable. So, it's not entirely the same as randomness. Just because a variable is unknown doesn't make it random.

Sabredog wrote:Survivorship bias is an interesting cognitive shortcoming.

From http://www.ceobootcamp.com/why-you-cant-learn-to-be-successful-by-studying-successful-people/

"Learning about success comes from understanding the difference between successes and failures. Unfortunately, in business that comparison is rarely done. Largely because it doesn’t make an entertaining speaker series, or an easy to write article. And because business failures tend to disappear and not subject themselves to scrutiny."


Everyone has biases, of which this is but one. There's truth in here, in that "why I was successful" speeches are often not full explanations. Certainly, motivational speakers, etc shill an awful lot of bullshit(and I note that "ceobootcamp" sounds a little like such a place).

But...studying is essential for comprehension. Studying is NOT just "listen uncritically to what they say", however. So, I'd say their premise is wrong out of the gate, largely because they can't handle basic definitions, or...more likely, opted to go for a more dramatic headline rather than a factual one. Which, of course, feeds back into motivational speaker and crap being mostly bullshit.


Return to “General”

Who is online

Users browsing this forum: Link and 25 guests