Neutrino
Neutrino
  • Threads: 84
  • Posts: 515
Joined: Feb 20, 2014
April 13th, 2014 at 7:06:58 PM permalink
I studied linguistics a little bit in college and I learned that sometimes a language is missing a concept. That is called a "lexical gap" and it is very difficult/frustrating to deal with.

So who else is sick of using "long term result" in place of EV to the average person? The concept of EV does not exist in average-person-english and just about everybody uses "long term result" in place of it when addressing a non-mathematically educated population.

But the problem is EV and "long term result" does not necessarily mean the same thing... the part of EV that means "average" has no way of showing up, except tends to deviate to the mean in the long run. But that is missing the point, a very key point. EV does not care about the results. It does not care about either the result of your current hand/game/run nor the result of a "long run", a year, 10 years, whatever. It does not care if you bought lottery tickets for 50 years and never hit a jackpot, it only cares you were supposed to lose $0.6 for every $1 you put in.

So analyzing it a bit linguistically, it seems so miserable to me that "long term result" substitutes EV in the average-person-english not because it means the same, because "long term result" TENDS to EV the bigger your sample size gets. so "long term result" is kinda an effect of EV, not EV itself, of sorts...

This gets into more serious problems as it might make a person think "Oh, i lost a hand, because i'm supposed to gain long term (or not lose so much long term), some winning hands probably are due" No. When you lose a hand that does not change the EV of your next hand, your money is gone for good. Deduct however much you lost from your long term expected return and that is your new long term return (Don't actually do this because you'll likely to experience selection bias and only do it on your losses). This "my win is overdue" could lead into much more serious problems such as utilizing a betting system for the reason of taking advantage of the overdue wins, and getting pwnd on WoO forums for saying you have a winning betting system.

Anyway EV being called "long term result" makes me quite disgusted. Anyone sick of it?
MathExtremist
MathExtremist
  • Threads: 88
  • Posts: 6526
Joined: Aug 31, 2010
April 13th, 2014 at 10:56:26 PM permalink
It's wrong, so I don't actually hear it much. When I'm explaining house edge to my friends and they don't already understand probability distributions, I start with the roll of a single die. The expected number of pips on the roll of a single die is 3.5. You can't actually roll a 3.5 on a die, but that's still the average number of pips per roll. Once you explain that, explaining that the expected loss on a $5 passline bet is $0.07 becomes a lot easier. You can never actually lose 7 cents on a $5 line bet, but that's the average loss for the bet. And that's not a long term result, it's for every roll.

Then, expected losses are additive, so if you play 100 times on a $5 passline bet, your expected loss is $7.

Finally, expected losses don't matter once real results are known. If you were going to play 100 times on a $5 pass bet, and then you actually did play 10 times and won 9 bets (losing 1) you now have $40 in winnings. Your expected results of the 100 plays, starting from the 11th bet, are $40 in actual winnings - $6.30 in future expected loss = $33.70. That last part helps explain why things don't become "due."
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice." -- Girolamo Cardano, 1563
FleaStiff
FleaStiff
  • Threads: 265
  • Posts: 14484
Joined: Oct 19, 2009
April 14th, 2014 at 12:05:49 AM permalink
Its always been a bit of a bother to me to say " a green eye shade type " doesn't even see the 5.00 bet on roulette, he sees only an imaginary shiny quarter and bright shiny penny that are not in fact there. No one can make a five dollar roulette bet and then lose twenty six cents.

This 3.5 pips per edge is somehow reminiscent of a bio statisticians 0.4 persons. Although I admit it was informative to have heard 3.5 pips as an "average roll".

The probability of past events is a troubling concept. If the event has taken place its 1.0. If it hasn't its 0.0.
Yet somehow we feel a roulette wheel that has spun in a forest and one that we just happened to come across are somehow to be described in the same language. Ten reds that we didn't know about or ten reds that we do know about still don't affect that 11th spin.
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
April 14th, 2014 at 4:01:22 AM permalink
I like how Mathextremist worded it. Gave a clear example how the average roll (3.5) can never be made in a single toss.
endermike
endermike
  • Threads: 7
  • Posts: 584
Joined: Dec 10, 2013
April 14th, 2014 at 5:25:23 AM permalink
Long term result has never bothered me. I grew up in a very statistically literate (and pro gambling) family so I quickly internalized the realities of these mathematical terms for myself. In essence, my definitions of words in English often grew to fit my understanding of the math, as opposed to understanding math via the definitions of the words in English. I assume this is quite rare (based on how the average person struggles with math and probability in particular).

If asked to teach these concepts to someone from the ground up (when "long term result" confuses them) I prefer to use "average of many future results." I also focus on the word "tendency." Focusing on that word helps emphasize the probabilistic nature of these concepts.
pacomartin
pacomartin
  • Threads: 649
  • Posts: 7895
Joined: Jan 14, 2010
April 15th, 2014 at 5:29:45 AM permalink
Quote: Neutrino

I studied linguistics a little bit in college and I learned that sometimes a language is missing a concept. That is called a "lexical gap" and it is very difficult/frustrating to deal with.



The "expected value" of the number of times you have to flip a fair coin to get tails is 2. If you flip a coin 2 times, the "expected value" of the number of times you get a tail is 75%.

So you do have to be very careful that you understand the question.
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
April 15th, 2014 at 5:37:40 AM permalink
"Long term result" is not that much different (if at all) from "EV" or "Expected Value". If you asked someone to flip a coin 1000 times and asked what the expected long term result would be...they would say about 500/500 heads/tails. I think most people, who are at least somewhat educated (graduated high school with B's or higher), can tell you that just because you flipped tails (any number of times in a row) that heads becomes due. Then again, I could be biased since I do more statistical-stuff than most people and most of the people I do interact with are more educated than most others (I assume, at least). But then again, I find myself shocked when I hear people say superstitious things or talk about how "this that and the other" factor into something, especially by people that I think should know better. [ie: "Why do I have a better chance at getting a flush in hearts than any other suit?"]

Quote: pacomartin

The "expected value" of the number of times you have to flip a fair coin to get tails is 2. If you flip a coin 2 times, the "expected value" of the number of times you get a tail is 75%.

So you do have to be very careful that you understand the question.



Do you mean "you expect to get AT LEAST 1 tail 75% of the time" ? TT, TH, HT, HH
Ibeatyouraces
Ibeatyouraces
  • Threads: 68
  • Posts: 11933
Joined: Jan 12, 2010
April 15th, 2014 at 5:57:26 AM permalink
deleted
DUHHIIIIIIIII HEARD THAT!
Nareed
Nareed
  • Threads: 373
  • Posts: 11413
Joined: Nov 11, 2009
April 15th, 2014 at 1:09:15 PM permalink
Quote: Neutrino

So who else is sick of using "long term result" in place of EV to the average person? The concept of EV does not exist in average-person-english and just about everybody uses "long term result" in place of it when addressing a non-mathematically educated population.



Is this kind of how people use "theory" when they really mean "hypothesis" or "guess"?
Donald Trump is a fucking criminal
  • Jump to: