How to Price an Election: A Martingale Approach- Discussion with Dhruv Madeka

Discussion with Dhruv of the paper: Election Pricing: a Martingale Approach and other related works.
arxiv.org/pdf/1703.06351.pdf

[@Jebcbeb]

This is a true Nassim video. Starts in the middle of a sentence and filmed in a room containing only hard services using a webcam bought from a farmer in Mongolia in 2005. Makes me pay extra close attention so I can understand.

[@Senecamarcus]

In fact he mentioned once that he does this so people will pay more attention. Nassim can basically prove anything, once he proved why it’s great to be lazy!

[@conornoble6205]

Lol

[@g13flat]

@@Senecamarcus I would have replied sooner but, well, you know, I was practising being lazy. Very important to practice. ;)

[@Ph4nt4sm4ge]

I watched this on the first nokia phone that was able to play video #lindy, while also preparing the crops in my backyard for winter season and wearing a coat I made from the fur of a bear I hunted recently.

[@ashnur]

I think he overshoots the artificially induced dyslexia when I can only guess what the mumble-muble-muble is and when the glare completely clears all information from the blackboard.

[@YamahaC7SRG]

Wow, this is thought-provoking! Let me see if I get this. If we accept: 1) The outcome of the upcoming US Presidential election is a highly volatile event; and 2) The outcome is binary (either Trump or Biden will win), both of which seem almost axiomatic, then the proper pricing (forecasting of the outcome) should be close to 50% over the remaining time to the election. In other words, it's a toss-up. Anyone with skin in the game, would be crazy to price bets on the basis of forecasts such as 'Biden has an 80% chance of winning so I'll pay you $1 for a 25 cent bet for Trump.' If I've got this right, it makes forecasting (other than 50/50) this election almost completely pointless- not because polls are no good - but because of volatility and the binary nature of this election. For example, what if Biden had a stroke today or if Trump's Covid comes back in a bad way? These are real possibilities with potentially huge impact but are largely unpredictable and only 2 of the infinite possibilities between now and election day. Something MUST BE WRONG with a forecast that says, today, Biden has an 80% chance of winning. The model is not capturing the uncertainty adequately, I would argue. I once heard a very smart guy say, "Don't tell me what you think [about the markets], tell me what's in your portfolio!" Thanks so much, yet again, for sharing!!

[@Ardgancrow]

No, it doesn't mean it should be 50/50, just that it's mean reverting and can't be 0 or 100 if there's any uncertainty in the forecast. The larger the uncertainty, the stronger the pull towards 50/50 (in a binary game) and you should be skeptical of extreme values. E.g. you can bet Trump right now at over 2/1 but I wouldn't touch that with a ten foot pole and have in fact a position on Biden at about -110 (1.91). I guess the main takeaway is people with skin in the game tend to get these things pretty quickly or you fall out of the game (sports betting, trading).

[@abbottmd]

to be true statistical probabilities the 80% would need an error estimate. Rather than saying the 80% is wrong, it might be more accurate to say it has a wide error range.

[@rdh_333]

dude, you got this right

[@jeanshawn]

Yes !!!! Thank you!!! I was going to request this!

[@joffy1234]

Great example and explanation to the lay man starts at around 20:40

[@miraclemaxicl]

Nassim, thank you for making your technical book available as a free download.

[@cptnbrrycrnch5194]

This needs to be the top comment, its fantastic to be able to read through and think about it instead of yet another paywall.

[@abbottmd]

definitely is great of him. I hope people purchase his other ones to repay his generosity. Dynamic Hedging from ~1996 is fantastic too

[@davidhcruz25]

Hello, where is this book available?

[@miraclemaxicl]

@@davidhcruz25 fooledbyrandomness.com/

[@hoots187]

Thank you for this Dhruv

[@takashikashiwase3461]

as always thank you

[@pie3566]

What's the name of the simplified version of Nassim's paper?

[@fisherh9111]

Beautiful. Thankyou.

[@falst573]

Awesome!

[@dymka2006]

The audio uncertainty is almost as bad as the election uncertainty.

[@djad3115]

Great, thank u!

[@SimpleTechVidz]

Does the process of determining anti-fragility rely on the set of shock “parameters” to be in mediocristan? Otherwise can’t there be a dimension of shock which was completely unforeseen?

[@williamnichols6253]

What book do I have to read to understand what you guys are talking about?

[@djcuvcuv]

wow this is very very fascinating

[@virtuerse]

Thank you sir

[@substanceandevidence]

Is there any chance we will see an extension with non-Brownian noise? I'm looking at polling for a different election which resembles a Levy flight more closely

[@nntalebproba]

Hi. The martingale process in [0,1] cannot be Levy, since it is bounded.

[@substanceandevidence]

@@nntalebproba thanks, what I mean is, it exhibits long periods of small changes and then sudden jumps which are uncharacteristic for Brownian processes (so I would imagine the hidden unbounded process is Levy)

[@NitinRanjan123]

Fat Tony & Dr. John !! P.S. --a reference to Skin in the Game characters

[@ericherbert8253]

i clicked on this and thought it said how to price an electron, i was seriously curious lol

[@ashnur]

same

[@tomchips8297]

BTC prices electrons. 🤪

[@mak529]

Brilliant

[@briandeleonardis6578]

Does anyone know the spelling of Nassim's friend Bruno's last name? I wanted to look him and see what work he's done, but the spellings Im trying to google aren't finding anything.

[@saurabhbcn]

en.wikipedia.org/wiki/Bruno_Dupire

[@bebopguitar]

Imagen Pythagoras would have presented his famous theorem to the world with that sound quality, and then written the essential formula just inside the glare on the blackboard ...the world would have had to wait a 1000 years longer to do that triangle calculations *lol*

[@Edy5656]

Would this mean Silvers model is still accurate when used right before the election? Because volitility becomes way lower as you approach election day?

[@xXchutterzXx]

I think no? Because the issue is that 538’s model fundamentally doesn’t integrate the information that the can affect the ‘true’ variation in probabilities. But I’m unsure if I’ve understood!!

[@Edy5656]

@@xXchutterzXx right, but most of the true variation comes from uncertainty of the future , much more certainty when you get to election eve/day. But I also didn't understand everything he was saying.

[@paulolivier7346]

Silver’s model is accurate the day after the election. Otherwise...

[@WmTyndale]

A Martingale Approach? That is some kind of a mixed drink right? They are gonna need it!

[@aubreyclayton7158]

LOL, I am not "someone apparently close to Nate Silver." Funny how the journal saw fit to publish my response paper despite it being so obviously faulty. Anyone wanting to actually understand the flaws in Taleb and Madeka's arguments is welcome to read it here: arxiv.org/abs/1907.01576 Less technical write-up here: nautil.us/blog/nassim-talebs-case-against-nate-silver-is-bad-math Tl;dr -- Taleb's intuition about volatility is wrong and he doesn't know the meaning of the word "arbitrage."

[@KartikChugh]

Props to you for showing up here lol

[@HillofNeedles]

The glare on the chalk board! Ha

[@carloscarpizo5845]

👀

[@dexterdexter402]

Nasim my friend, you have a tendency to take over conversation. Track the minutes of who spoke how much during this time. In these formats don't be afraid to take your time and make longer videos rather than cutting people. Interested people will watch it. .... Thanks for the upload and I have read parts of some of your books. Indeed valuable. Thanks.

[@TerryPullen]

The sound was too degraded for me to keep up. Thumbs up but couldn't listen through.

[@fallinginthed33p]

Listen on headphones and EQ the bass down.

[@TerryPullen]

@@fallinginthed33p Thanks. Although I find Taleb interesting I have no real need to hear what he has to say, so I don't have much incentive to try.

[@khaledsrrr]

My father is a politician and his forecast are bad. From time to time he changes his mind but he loses nothing haha 😂. I am an options trader and I lose if I want to change my mind. I love my dad haha 😂

[@johnmurphy5258]

Nassim please buy a better camera !!

[@ashnur]

kzbin.info/www/bejne/qnamaaRvnpeZhdk here is an easy list to follow on how to not suck at youtube

[@ericherbert8253]

i clicked on this and thought it said how to price an electron, i was seriously curious lol

[@DAWMiller]

Super easy, just find its position and velocity and you’re good to go.