spent a bit too long on this one to not post..

if the first two bullets are blanks, you do not want to spin it one click, because there is only one pair of consecutive blanks where the next shot is a bullet.

when spinning it , there are two possible pairs of blanks on the revolver where the shot 2 clicks after it is a bullet.

you are twice as likely to shoot a bullet if you spin it one click, but the exact odds depend on the number of chambers

edit: or does spin mean to spin the chamber randomly? if thats the case, of course you want to spin...

@nedprz 6 chambers in a revolver

2/6 if you spin
1/3 if you don't

so the answer is it's 50/50, either you spin or you don't

http://www.mathgoespop.com/2012/02/i...-roulette.html

damn..

woops i misread the question. thought it was fired twice. 1/4 if you don't spin

the question itt and the one on the link are different. i am still having trouble convincing myself my answer is incorrect but im going to try it again

sorry all for the derail

i didn't read the link but i'm pretty sure my answer is correct, unless i'm still misreading the question

(and yes, spinning it means it randomizes the position)

This sounds like Assent/GPC in Austin. I was there in '08 and it was a crazy time. We all traded the Arca imbalances as a group and it was close to free money for 3 months. I think if you're starting out now at a prop firm, without a proper training program or mentor who is making money, you basically have 0 chance of making it.

Let's say chambers are numbered 1 to 6 and bullets are in chambers 1 and 2.

If you spin, 2/6 because it's now random.

Now if you don't spin, the fact the last shot cannot = 1 or 2 means you are now at 1, 4, 5, or 6. Therefore you're 1/4 to get shot if you pull again without spinning.

This is really a logic test on whether you can identify useful information

90% of the confusion off the bullet question stems from the fact that the question in the thread and the one in the link are different. Stinkypete had the right answer. This part of the thread should be over.

So you're saying there's a lower chance of getting shot by not spinning? You clearly fail at logic. Think about it.

Odds:

P(no shot no spin) = C (3,2) / C (4,2) = 0.5
which is also simply 2/4

P(no shot if spin) = C (5,2) / C (6,2) = 0.67
which is also 4/6

Sent from my SAMSUNG-SM-G925A using Tapatalk

I guess the link was wrong?

I quoted grimreaper's post and I guess it was off.

Oops, for some reason I thought the trigger was pulled twice. Anyway, it doesn't change the fact you want him to always spin for the logical reason that when he shoots and misses, he "exhausts" one empty chamber, making it more likely that he'll shoot a bullet on the next shot then before he shot the first bullet.

P(survive if no spin) = C (4,2) / C (5,2) = 0.6
which is also simply 3/5

P(survive if spin) = C (5,2) / C (6,2) = 0.67
which is also simply 4/6

Article is tldr, I'll get to it later

strike 2 for grimreaper, strikeout coming shortly

Unfortunately we're now seeing the other 10% play out

I know douchebags that ask something similar to the NYC market making question, basically as a challenge to the candidate. It ends up being a personality/gut test unless you say something logically inconsistent.

Another asinine question I've been hearing a lot recently is "do you think r-squared of .46 is good or bad?"

That's not nearly as bad of a question imo. It should make it pretty easy to judge if the candidate has an understanding of statistics beyond mathematical definitions. The market making question sucks because you have to either make bad assumptions or ask a neverending stream of questions about assumptions, and usually the way it's framed the candidate is sort of tricked into making bad assumptions.

The worst I've seen personally was a Credit Suisse director asking me about the myanus of an option. Somehow I kept a straight face when I asked him to explain the myanus, only to realize he was trollololing me and was actually talking about moneyness.

That was my first response, but then they just smile and say 'good point, but lets say you just have to do it anyways', so you still need to do some mathematical acrobatics.

After pushing back on some flawed assumptions, one guy ended up saying that the buyer was forced to trade with me for a business reason, which makes it a whole new type of problem

Yeah he wanted very specific answers. I gave him a market of 8-12 MM then he took the ask and i made a new market at 12-15MM then he took the ask and i made a new market at 15-17MM etc. This went on and on for like 5 rounds and he wanted market range and volume offered at each round.

This is correct. IRT to the bullets question, the assumption is that the trigger was only pulled once when he asked the question, sorry for the confusion.

This is not correct. Odds of survival are better on not spinning the chamber if bullets are loaded consecutively.


A couple other questions i've been asked:

Let's play a game, the goal of the game is to say the number 50. The rules of the game are that you say a number between 1 and 10, and the next player can add a number between 1 and 10. It's a heads up game. Do you want to go first or second, and how do you guarantee you win the game?

A deck of cards is face up, we play this game heads up as well. You can pick any 5 card hand that you want, then the 2nd player can. Then you have the choice of replacing any or all of your cards, 1 time, from the remaining cards in the deck, as does the 2nd player. Do you go first or second, and how do you guarantee you win the game?

Game 1:
You want to race to 39 (only way to force 40+=) which guarantees win.
Which means you want to force opponent to play 29 to 38. That in turn means you are racing to 28. So force opponent to play 18 to 27. Which means race to 17. Which means you want to force opponent to play 7 to 16.

Go first and play 6 to guarantee a win.


Intermediate proofs would be:
If you play 39, opp can't prevent you from 50
If you play 28, opp can't prevent 39
If you play 17, opp can't prevent 28
If you play 6, opp can't prevent 17

Game 2:
Go First and take all the tens, guaranteeing you can form at least ten high straight flush while preventing opp from getting anything better than 9 high straight flush, assuming replaced cards don't go back in deck.

If replaced cards do go back in, then a tie is inevitable as first player will pick up royal flush right away. Player 1 cannot win because if player 1 forms ANY straight flush on first or second move, player 2 must have a way to form a royal on his first or second. To block two from forming a royal, player 1 must have at least one broadway card from each suit after second move, capping his hand value at quad aces, which means player 2 can form a straight flush

You snap answer "it depends"?

That is just about the only intelligent answer, yes.

I'm not sure I can give much information so I was thinking about whether to write at all, but I guess why not. I got into the SIG trading internship program for this summer in Philadelphia. I'm on my penultimate year studying for a BSc in Finance. The interview process consists of two rounds of phone interviews and a final interview round live at SIG. I don't know for a fact, but from what I've read and heard, I think a larger % of the applicants are passed trough to the first phone interview, compared to the larger investment banks for example, but then they drop a ton of more people during the first and second phone interviews.

The interviews consist of both the type quantitative problems/brain teasers similar to what has been mentioned in previous posts in this thread, and just normal job interview questions about you. The reason I decided to apply specifically to SIG was that I heard about their poker interest. They were founded by a poker player 30 years ago, and they include poker in their trader training. I think you can find some descent information about the application process if you google hard enough. I'd consider myself very sharp and fast at these kinds of quantitative problems, but I did work on similar problems for a few full days before the interviews. Conditional probability and especially the Bayes' theorem is important. I did very well in the problems, but I do think my poker background helped a ton and we spent most of the normal part of the interview talking about it. I explained that the biggest achievement was my cash game results which are fairly easy to confirm in highstakesdb, but not sure how well I was able to explain that. I feel the most important things to them were my WSOP bracelet and EPT victory. Got a list of the other interns a few days ago, and it does seem like jobs at SIG are in extremely high demand:

Congrats on the internship!

If their process is still similar to what it used to be, I think once you've gotten through the initial resume filters they care very little about the poker stuff and what really matters is how you perform on their probability based brain teasers.

Interesting to see the list of schools. It has either become way more competitive (likely) or they're targeting top schools more effectively these days. When I was there nine years ago the list looked like this:



Have fun in Philly. Try not to get shot.

Congrats!

Also the first thing I saw when googling them is that they brought in Annie Duke to teach a poker class so maybe you can learn something from her